ADAPTING TO CLIMATE CHANGE: THE EFFECT OF DESERTIFICATION ON THE PHYSIOLOGY OF FREE-LIVING UNGULATES Robyn Sheila Hetem A thesis submitted to the Faculty of Science, University of the Witwatersrand, in fulfilment of the requirements for the degree of Doctor of Philosophy Johannesburg, South Africa, 2009 i DECLARATION I declare that the work contained in this thesis is my own, unless otherwise specified. It is being submitted for the degree of Doctor of Philosophy in the University of the Witwatersrand, Johannesburg. The work herein has not been submitted before for any degree or examination in any other university. ________________________ Signed on the _______ day of ________________, 2009 ii ABSTRACT For long-lived species, physiological plasticity provides the best option to counter extinction and survive climate change, yet we understand very little about long-term physiological adaptations of arid-zone artiodactyls. Variation within a morphological trait, for example, may provide pre-adaptation to changing climatic conditions. Using intra-abdominal miniature data loggers, I measured core body temperature in female springbok (Antidorcas marsupialis) of three colour morphs (black, normal and white) and showed that the pelt colour does indeed have thermoregulatory significance. The black springbok seem able to reduce energy expenditure in winter, but experience higher solar heat load in hot conditions. Therefore lighter coloured individuals may be selected for in the future as conditions get progressively hotter and drier with climate change. Small individuals also may be selected for in the future. Hypothetically, small artiodactyls would have the advantage of smaller resource requirements and greater access to refuge sites than do larger artiodactyls, but they are likely to be disadvantaged by a high mass-specific metabolic rate, high water turnover and less capacity to store heat. I measured body temperature, activity and microclimate selection in free-ranging Arabian oryx (Oryx leucoryx, ? 70 kg) and the smaller Arabian sand gazelle (Gazella subgutturosa marica, ? 15 kg) inhabiting one of the hottest and driest regions, Arabian Desert environment, at the same time. Despite the oryx having a body mass more than four-fold that of the sand gazelle, both species responded remarkably similarly to changes in environmental conditions. Both species employed heterothermy and cathemerality, and selected the same cool microclimates during times of heat stress. In combination with high ambient temperatures, water stress appeared to be the primary driver towards heterothermy in the Arabian oryx, potentially resulting from dehydration or the combined effects of dehydration hyperthermia and starvation hypothermia. iii To investigate the physiological consequences of habitat transformation, of the kind expected to occur with climate change, I monitored body temperature and activity patterns of Angora goats inhabiting both desertified and intact sites. I was able to demonstrate physiological changes in response to desertification. Following shearing, goats that inhabited the transformed site displayed an increased 24-h amplitude of body temperature rhythm and were generally less active compared to goats that inhabited the intact site, which may reflect a trend towards heterothermy and cathemerality, as was observed in the Arabian oryx and sand gazelle. Finally, the physiology studies required to better understand the mechanisms of phenotypic plasticity underlying responses to climate change cannot be confined to the function of healthy animals as pathogens are predicted to spread with climate change. I therefore investigated the physiological consequences of infection in free- living kudu (Tragelaphus strepsiceros). Not only did I record quantitative evidence for autonomic and behavioural fever, but I also recorded the first evidence of sickness behaviour, in the form of decreased activity, in a free-living artiodactyl. Artiodactyl hosts are likely to have to contend with an increased costs of immunity superimposed on the chronic physiological stress of having to adapt to the climatically unsuitable areas to which they are confined. In conclusion, I have revealed some of the physiological mechanisms that will be brought into action if long-lived artiodactyls are able to adapt phenotypically to climate change in Africa. Activity patterns and microclimate selection are flexible behavioural processes which are likely to represent an animal?s primary defence to changes in climatic conditions. If behavioural processes are insufficient to maintain homeothermy, we may observe changes in an animal?s body temperature, a sensitive indicator of infection, dehydration, nutrition and environmental stress. Such physiological measurements need to be incorporated into long-term physiological monitoring projects, and bioclimatic envelope models, so that we can better predict how species will respond to climate change. iv ACKNOWLEDGEMENTS I am indebted to many people without whom my research would not have been possible. First to my supervisors, Prof. Andrea Fuller, Prof. Graham Kerley and Prof. Duncan Mitchell, for teaching me about research, for promoting independent thinking and for allowing me to pursue various research interests, even when that meant flying to Saudi Arabia. Throughout my PhD I have had the privilege of collaborating with many colleagues for whom I have great respect, and I wish to extend my sincere appreciation to all those who have helped me. I am grateful to my co-authors: Ms Brenda de Witt for her friendship and support and the many hours we spent chasing goats together, Dr. Linda Fick for her support and understanding and whose practical thinking and organisational skills made many of the projects viable, Dr. Shane Maloney, for his continued enthusiasm and patience in teaching me about heat transfer and water turnover, Dr. Leith Meyer for his unsurpassed surgical skills in the field and his professional composure despite many stressful situations, Dr. Mohammed Shobrak for supporting my research in Saudi Arabia and for all his efforts in trying to obtain a visa for a single foreign female, Mr Maartin Strauss, for allowing me to feel safe in a country where women have few rights, and for loving me despite the penguin suit. I would also like to thank the staff and my fellow postgraduate students within the Brain Function Research Group for their advice and support. Finally, I am thankful to my family who supported me financially and emotionally throughout this endeavour and my friends who helped me maintain a sense of humour and who so often provided a shoulder to cry on. Research for this thesis was funded by the National Research Foundation of South Africa, the Medical Faculty Research Endowment Fund of the University of the Witwatersrand and a START/NORAD African PhD fellowship, for which I am extremely grateful. v TABLE OF CONTENTS DECLARATION ................................................................................................................................. i ABSTRACT ....................................................................................................................................... ii ACKNOWLEDGEMENTS.................................................................................................................iv TABLE OF CONTENTS .....................................................................................................................v LIST OF ABBREVIATIONS...............................................................................................................x LIST OF FIGURES.............................................................................................................................xi LIST OF TABLES .............................................................................................................................xv RESEARCH OUTPUTS ...................................................................................................................xvi AUTHOR CONTRIBUTIONS..........................................................................................................xix CHAPTER 1 INTRODUCTION............................................................................................................................. 1 1.1 RANGE SHIFTS ................................................................................................................. 4 1.1.1 Bioclimatic envelope models....................................................................................... 5 1.2 MICROEVOLUTION ........................................................................................................... 9 1.3 PHENOTYPIC PLASTICITY.................................................................................................13 1.3.1 Physiological adaptation ...........................................................................................16 1.3.1.1 Artiodactyls..................................................................................................................20 1.3.1.2 Thermoregulatory trade-offs..........................................................................................28 1.3.2 Future applications ...................................................................................................31 1.4 THESIS AIMS ...................................................................................................................33 CHAPTER 2 MORPHOLOGICAL ADAPTATION TO CLIMATE CHANGE: THE EFFECT OF PELT COLOUR ON BODY TEMPERATURE AND THERMOREGULATORY BEHAVIOUR OF SPRINGBOK (ANTIDORCAS MARSUPIALIS) ......................................................................................................39 2.1 ABSTRACT......................................................................................................................40 2.2 INTRODUCTION ...............................................................................................................40 2.3 MATERIALS AND METHODS .............................................................................................43 2.3.1 Body temperature and thermoregulatory behaviour....................................................43 2.3.1.1 Study area ....................................................................................................................43 2.3.1.2 Animals........................................................................................................................44 2.3.1.3 Surgery ........................................................................................................................45 2.3.1.4 Body temperature measurements...................................................................................46 2.3.1.5 Behavioural observations ..............................................................................................47 vi 2.3.1.6 Climate measurements..................................................................................................48 2.3.1.7 Body temperature analysis ............................................................................................48 2.3.2 Pelt heat transfer characteristics ...............................................................................49 2.3.2.1 Pelt samples .................................................................................................................49 2.3.2.2 Spectral reflectance ......................................................................................................50 2.3.2.3 Thermal conductance....................................................................................................50 2.3.2.4 Radiant heat load..........................................................................................................51 2.3.2.5 Data analysis of pelt characteristics ...............................................................................52 2.4 RESULTS ........................................................................................................................54 2.4.1 Body temperature and thermoregulatory behaviour ...................................................54 2.4.1.1 Climate ........................................................................................................................54 2.4.1.2 Body temperature .........................................................................................................54 2.4.1.3 Environmental effects on the nychthemeral rhythm of body temperature ........................57 2.4.1.4 Behavioural thermoregulation .......................................................................................61 2.4.2 Pelt heat transfer characteristics ...............................................................................64 2.4.2.1 Spectral reflectance ......................................................................................................64 2.4.2.2 Thermal conductance....................................................................................................65 2.4.2.3 Heat load from radiation ...............................................................................................65 2.4.2.4 Potential energetic consequences...................................................................................67 2.5 DISCUSSION ...................................................................................................................68 2.6 ACKNOWLEDGEMENTS ...................................................................................................73 CHAPTER 3 ADAPTING TO ARID ENVIRONMENTS: THERMOREGULATION AND CATHEMERALITY OF FREE-LIVING ARABIAN ORYX (ORYX LEUCORYX) IN A HYPER-ARID DESERT............74 3.1 ABSTRACT .....................................................................................................................75 3.2 INTRODUCTION...............................................................................................................76 3.3 MATERIALS AND METHODS .............................................................................................78 3.3.1 Animals and habitat ..................................................................................................78 3.3.2 Surgery .....................................................................................................................79 3.3.3 Temperature and activity measurements ....................................................................82 3.3.4 Climatic data measurements......................................................................................83 3.3.5 Data analysis ............................................................................................................83 3.4 RESULTS ........................................................................................................................85 3.4.1 Climate .....................................................................................................................85 3.4.2 Body temperature......................................................................................................86 vii 3.4.3 Activity......................................................................................................................92 3.4.4 Microclimate .............................................................................................................94 3.4.5 Selective brain cooling...............................................................................................97 3.5 DISCUSSION..................................................................................................................100 3.5.1 Heterothermy ..........................................................................................................102 3.5.2 Cathemerality..........................................................................................................107 3.5.3 Selective brain cooling.............................................................................................109 3.5.4 Conclusion ..............................................................................................................111 3.6 ACKNOWLEDGEMENTS..................................................................................................112 CHAPTER 4 DOES SIZE MATTER? COMPARISON OF BODY TEMPERATURE AND ACTIVITY OF FREE- LIVING ARABIAN ORYX (ORYX LEUCORYX) AND THE SMALLER ARABIAN SAND GAZELLE (GAZELLA SUBGUTTUROSA MARICA) IN THE SAUDI DESERT ............................113 4.1 ABSTRACT....................................................................................................................114 4.2 INTRODUCTION .............................................................................................................114 4.3 MATERIALS AND METHODS ...........................................................................................117 4.3.1 Animals and habitat.................................................................................................117 4.3.2 Surgery ...................................................................................................................117 4.3.3 Temperature measurements .....................................................................................120 4.3.4 Meteorological data measurements..........................................................................120 4.3.5 Data analysis...........................................................................................................121 4.4 RESULTS ......................................................................................................................122 4.4.1 Climate ...................................................................................................................122 4.4.2 Body temperature ....................................................................................................123 4.4.3 Activity....................................................................................................................127 4.4.4 Microclimate ...........................................................................................................129 4.5 DISCUSSION..................................................................................................................131 4.6 ACKNOWLEDGEMENTS..................................................................................................140 CHAPTER 5 EFFECTS OF DESERTIFICATION ON THE PHYSIOLOGY OF ANGORA GOATS: TESTING GLOBAL CHANGE PREDICTIONS.............................................................................................141 5.1 ABSTRACT....................................................................................................................142 5.2 INTRODUCTION .............................................................................................................142 5.3 MATERIALS AND METHODS ...........................................................................................145 viii 5.3.1 Animals and habitat ................................................................................................145 5.3.2 Surgery ...................................................................................................................145 5.3.3 Temperature measurements.....................................................................................147 5.3.4 Meteorological data measurements..........................................................................148 5.3.5 Isotope analysis.......................................................................................................148 5.3.6 Blood variables .......................................................................................................150 5.3.7 Data analysis ..........................................................................................................150 5.4 RESULTS ......................................................................................................................151 5.4.1 Climate ...................................................................................................................151 5.4.2 Hot and cold days....................................................................................................152 5.4.3 Shearing..................................................................................................................155 5.5 DISCUSSION .................................................................................................................162 5.5.1 Body temperature....................................................................................................163 5.5.2 Autonomic thermoregulation ...................................................................................165 5.5.3 Activity....................................................................................................................166 5.5.4 Water influx ............................................................................................................168 5.5.5 Water turnover rate.................................................................................................169 5.5.6 Parasite load...........................................................................................................170 5.5.7 Conclusion..............................................................................................................171 5.6 ACKNOWLEDGEMENTS: ................................................................................................172 CHAPTER 6 ADAPTING TO THE SPREAD OF PATHOGENS: FEVER AND SICKNESS BEHAVIOUR DURING AN OPPORTUNISTIC INFECTION IN A FREE-LIVING ANTELOPE, THE GREATER KUDU (TRAGELAPHUS STREPSICEROS)...................................................................................173 6.1 ABSTRACT ...................................................................................................................174 6.2 INTRODUCTION.............................................................................................................175 6.3 MATERIALS AND METHODS ...........................................................................................177 6.3.1 Animals and habitat ................................................................................................177 6.3.2 Surgery ...................................................................................................................178 6.3.3 Temperature measurements.....................................................................................181 6.3.4 Meteorological data measurements..........................................................................182 6.3.5 Data analysis ..........................................................................................................182 6.4 RESULTS ......................................................................................................................183 6.4.1 Climate ...................................................................................................................183 6.4.2 Fever ......................................................................................................................184 ix 6.4.3 Sickness behaviour ..................................................................................................188 6.4.4 Selective brain cooling.............................................................................................189 6.5 DISCUSSION..................................................................................................................193 6.6 ACKNOWLEDGEMENTS..................................................................................................199 CHAPTER 7 CONCLUSION..............................................................................................................................200 7.1 MORPHOLOGICAL ADAPTATIONS ...................................................................................201 7.2 AUTONOMIC ADAPTATIONS ...........................................................................................205 7.3 BEHAVIOURAL ADAPTATIONS........................................................................................208 7.4 ADAPTING TO DESERTIFICATION ....................................................................................211 7.5 ADAPTING TO THE SPREAD OF PATHOGENS.....................................................................213 7.6 PERSPECTIVES AND SIGNIFICANCE .................................................................................215 LITERATURE CITED...................................................................................................................217 APPENDIX 1.................................................................................................................................259 Hetem R.S., Maloney S.K., Fuller A., Meyer L.C.R. and Mitchell D. (2007). Validation of a biotelemetric technique, using ambulatory miniature black globe thermometers, to quantify thermoregulatory behaviour in ungulates. Journal of Experimental Zoology A: Comparative and Experimental Biology 307: 342?356. APPENDIX 2.................................................................................................................................274 Hetem R.S., de Witt B.A., Fick L.G, Fuller A., Kerley G.I.H., Maloney S.K., Meyer L.C.R. and Mitchell D. (2009). Summer shearing affects body temperature of Angora goats (Capra aegagrus) more than does winter shearing. Animal 3: 1025-1036. x LIST OF ABBREVIATIONS A/D analog/digital aut autumn C pelt conductance (W.m-2.?C-1) HFTs Heat Flux Transducers i.d. inner diameter I.M. intramuscularly I.V. intravenously o.d. outer diameter PE Port Elizabeth PTFE polytetrafluoroethylene Q or HF heat flow (W.m-2) S.C. subcutaneously spr spring sum summer Ta air temperature Tp plate temperature Ts skin surface temperatures VPDB Vienna Pee Dee Belemnite V-SMOW Vienna-Standard Mean Ocean Water W mass (kg) win winter xi LIST OF FIGURES Figure 1.1. Observed current (small map) and predicted (large map) distribution of the scimitar-horned oryx (Oryx dammah) for 2050 (adapted from Thuiller et al., 2006)...............................................8 Figure 1.2. Mean brain temperature at each 0.1?C category of blood temperature for a single oryx (Oryx gazella) over a time period of 15 days (adapted from Maloney et al., 2002)......................21 Figure 1.3. Body temperature rhythm of a well-hydrated (grey line) and dehydrated (black line) camel over a two day period (adapted from Schmidt-Nieslen, 1957).....................................................22 Figure 1.4. Correlation of the amplitude (the difference between maximum and minimum) of the nychthemeral pattern of body temperature versus the mean or mid-point of the range of ambient temperature over the study period (n = 14).................................................................................28 Figure 2.1. The pelt colour variations of the black, normal and white springbok (adapted from Skinner and Louw, 1996). ......................................................................................................................42 Figure 2.2. Experimental set-up for measuring thermal conductance and radiant heat load in the springbok pelts. .........................................................................................................................52 Figure 2.3. Mean ? SD of the mean, minimum, maximum, amplitude and rate of body temperature rise of the nychthemeral rhythm of body temperatures of black (closed bars), normal (hatched bars) and white (open bars) springbok over four seasonal periods .......................................................56 Figure 2.4. Mean ? SD of body temperature of black (black line), normal (dark grey line) and white (light grey line) springbok, as a function of time of day, over the 10 coldest days in winter and the 10 hottest days in summer of the one-year study period. .......................................................58 Figure 2.5. Correlation of the amplitude (the difference between maximum and minimum) of the nychthemeral pattern of body temperature, averaged for all animals in the group, for black (black circle), normal (dark grey circles) and white (light grey circles) springbok, versus the 24-h range of air temperatures on the same day...........................................................................................60 Figure 2.6. Activity, expressed as the proportion of the total number of observations for which the animals were active, during each hour of the 24-h day (open bars), and the mean body temperature (solid line) of black, normal and white springbok, as a function of time of day ........63 Figure 2.7. The spectral reflectance of a black (solid black line), normal (dashed dark grey line) and white (dotted light grey line) springbok pelt over the spectral range, from ultraviolet to infrared 64 Figure 2.8. Thermal conductance of the pelt (A), and heat load from radiation conveyed through the pelt as a percentage of incident radiant load (B), for a black (black circles), normal (dark grey circles) and white (light grey circles) springbok pelt, as a function of wind speed. ......................66 Figure 2.9. Metabolic savings of black (closed bars), normal (hatched bars) and white (open bars) springbok over four seasonal periods, when animals were orientated parallel (upper panel) and perpendicular (lower panel) to solar radiation. ...........................................................................67 xii Figure 3.1. The top two panels show the original record of 15-min recordings of body temperature from a single free-living male oryx (Oryx 2, left panel) and the captive male oryx (right panel) , which had access to water ad libitum, over the 11-month study period (April 2006 to February 2007). The bottom two panels show the air temperature recorded at nearby weather stations. ..... 87 Figure 3.2. The top two panels show the nychthemeral rhythm of body temperature (mean ? SD), averaged for all five free-living oryx, over the four periods. The bottom two panels show the nychthemeral rhythm of air temperature (mean ? SD) over the four periods. .............................. 88 Figure 3.3. 24-h mean (panel A), minimum (panel B), maximum (panel C), and amplitude of the nychthemeral rhythm (panel D) of body temperature (mean ? SD, n = 5) over the four periods... 90 Figure 3.4. 24-h mean of body temperature plotted against mean 24-h air temperature (panel A), and the maximum daily body temperature plotted against maximum daily air temperature (panel B) for 313 individual days, when the oryx were living free. Minimum daily body temperature was correlated poorly with minimum daily air temperature (panel C) and with maximum daily air temperature (panel D). So too, amplitude of nychthemeral rhythm of body temperature was correlated poorly with the 24-h amplitude of air temperature (panel E) but better with photoperiod (panel F)................................................................................................................ 91 Figure 3.5. Nychthemeral rhythm of body temperature (black line) and activity (grey bars) of a representative oryx (Oryx 3) over the warm wet (panel A), hot dry (panel B) and warm dry (panel C) periods....................................................................................................................... 93 Figure 3.6. Mean 24-h activity, averaged for all five free-living oryx, plotted against maximum daily air temperature (panel A), and photoperiod (panel B), for 224 days, and mean diurnal activity, calculated as activity between 06:00 and 18:00 expressed as a proportion of total 24-h activity, plotted against maximum daily air temperature (panel C), and photoperiod (panel D)................. 95 Figure 3.7. Scatter diagram showing the relationship between collar miniglobe temperatures at the site chosen by a single representative female oryx (Oryx 1) and temperatures recorded by an identical miniglobe at a nearby weather station, which was exposed to the sun. .......................... 96 Figure 3.8. Brain (light line) and carotid blood (dark line) temperatures of a single free-living male oryx (Oryx 2) over four-day periods during the warm wet period (A) and the hot dry period (B).97 Figure 3.9. Hypothalamic temperature as a function of carotid blood temperature (top panels) and the frequency distribution of blood temperature (bottom panels), in a male oryx (Oryx 3, panel A), and a female oryx (Oryx 1, panel B), when both animals were captive and had ad libitium access to food and water, and a free-living male oryx (Oryx 2), in the warm wet period (panel C) and in the hot dry period (panel D). ..................................................................................................... 99 Figure 4.1. Standard black globe (mean ? SD, black line) and air dry-bulb temperature (mean ? SD, grey line) as a function of time of day, for the warm wet (left panel) and hot dry (right panel) periods.................................................................................................................................... 123 xiii Figure 4.2. Nychthemeral rhythm of body temperature for five free-living oryx (mean ? SD, grey line) and four free-living sand gazelle (mean ? SD, black line) over both the warm wet (left panel) and hot dry (right panel) periods ....................................................................................................124 Figure 4.3. The top two panels show mean 24-h amplitude of the nychthemeral rhythm of body temperature, averaged for four sand gazelle (panel A) and for five oryx (panel B) plotted against the concurrent 24-h amplitude of air temperature. The bottom four panels show the mean maximum 24-h body temperature for sand gazelle (panel C) and oryx (panel D), and mean minimum 24-h body temperature for sand gazelle (panel E) and oryx (panel F), plotted against concurrent maximum 24-hour air temperature..........................................................................126 Figure 4.4. The top two panels show minimum 24-h body temperature, averaged for four sand gazelle (panel A) and for five oryx (panel B) plotted against the mean 24-h activity for the preceding day. The bottom two panels show the fraction of total activity which took place during daylight hours, for six sand gazelle (panel C) and five oryx (panel D), plotted against concurrent maximum 24-h air temperature. .......................................................................................................................127 Figure 4.5. Nychthemeral rhythm of activity for six sand gazelle (white bars) and five oryx (black bars) over both the warm wet (left panel) and hot dry (right panel) periods. ......................................128 Figure 4.6. Nychthemeral rhythm of microclimate selection, expressed as the difference between miniglobe temperature at the site chosen by four free-living sand gazelle (mean ? SD, black line) and two oryx (mean ? SD, grey line) and the temperature of an identical miniglobe exposed to the sun at a nearby weather station...........................................................................................129 Figure 4.7. Scatter diagram showing the relationship between miniglobe temperatures at the site chosen by a male sand gazelle (left panel) and a male oryx (right panel) and the miniglobe temperatures recorded at a nearby weather station, during warm wet (upper panel) and hot dry (lower panel)...........................................................................................................................130 Figure 5.1. Nychthemeral rhythm (mean ? SD) of abdominal temperature, subcutaneous temperature and the difference between abdominal and subcutaneous temperature of 12 goats that inhabited the intact site (solid line) and 12 goats that inhabited the transformed site (dotted line), as a function of time of day, over the five coldest days in winter and the five hottest days in summer. The lowest panel represents the nychthemeral activity rhythm (mean ? SD) of a single goat that inhabited the transformed site (white bars) and another goat that inhabited the intact site (black bars) over the same periods .....................................................................................................154 Figure 5.2. Nychthemeral rhythm (mean ? SD) of abdominal temperature, subcutaneous temperature and the difference between abdominal and subcutaneous temperature for 12 goats that inhabited the transformed site (dotted line) and 12 goats that inhabited intact site (solid line) over 10 days after summer shearing. The lowest panel shows the nychthemeral rhythm of activity for a single xiv goat that inhabited the transformed site (white bars) and another goat that inhabited the intact site (black bars) over the same 10 day period. ................................................................................ 156 Figure 5.3. Scatter diagram showing the relationship between miniglobe temperatures recorded on a single Angora goat that inhabited the transformed site (grey dots and regression line) and another goat that inhabited the intact site (black dots and regression line) plotted against miniglobe temperatures recorded at a nearby weather station, during a 10 day period post-shearing in summer................................................................................................................................... 158 Figure 5.4. Water influx (upper panel) and water turnover rates (lower panel) of 12 goats that inhabited the transformed (white bars) and intact (black bars) site, after summer shearing, over a seven-day period during which the goats were denied access to water. ..................................................... 159 Figure 5.5. Angora goat serum carbon isotope ratios (A), total plasma lipid (B), serum albumin to globulin ratio (C) and serum glucose concentration (D) of goats that inhabited the transformed (open circles) and intact (closed circles) site ............................................................................ 161 Figure 6.1. Field site weather station data, showing the standard black globe (black line) and dry-bulb air (grey line) temperatures (A), as well as radiation (black line) and wind speed (grey line) (B), as a function of time of day ..................................................................................................... 184 Figure 6.2. Abdominal temperature for a single free-living female kudu (kudu 6) over 11 days, including 5 days of fever. ........................................................................................................ 185 Figure 6.3. Mean ? SD of the nychthemeral rhythm of abdominal temperature for seven female free- living kudu during febrile (grey line) and afebrile (black line) states. ....................................... 185 Figure 6.4. Scatter diagram showing the relationship between standard black globe temperatures at the site chosen by a single animal (kudu 5), converted from the miniglobe temperatures recorded on the collar, and standard black globe temperatures recorded at a nearby weather station, during febrile (grey) and afebrile (black) states, for a single kudu.. ..................................................... 187 Figure 6.5. Difference between abdominal and subcutaneous temperature over the first day of fever (grey line), and for an average day of non-febrile states (black line), as a function of time of day for seven kudu (mean ? SD).................................................................................................... 188 Figure 6.6. Nychthemeral activity rhythm of a single kudu (kudu 2) during febrile (white bars) and afebrile (black bars) states (mean ? SD)................................................................................... 189 Figure 6.7. The pattern of brain (black line) and carotid blood (grey line) temperatures of a single kudu (kudu 3) over a four-day febrile period, and for the same kudu over a four-day afebrile period. 190 Figure 6.8. The mean, minimum, maximum and standard deviation of hypothalamic temperature for each 0.1?C class of carotid blood temperature during both febrile and afebrile states for four kudu. ...................................................................................................................................... 192 xv LIST OF TABLES Table 1.1. Body temperatures of unrestrained wild artiodactyls ...........................................................23 Table 1.1. Continued ..........................................................................................................................24 Table 1.1. Continued ..........................................................................................................................25 Table 1.1. Continued ..........................................................................................................................26 Table 1.1. Continued ..........................................................................................................................27 Table 2.1. Environmental conditions (mean ? SD) prevailing during the four seasonal periods during which the springbok were free-living in the Steytlerville Karoo region, during the three-month winter study at Port Elizabeth (PE win) and during the 10 hottest days in summer and the 10 coldest days in winter in the Steytlerville Karoo region. .............................................................55 Table 2.2. Association between partitioning of thermoregulatory behaviour and the three springbok colour morphs, over a 24-h day, as well as over nocturnal and diurnal periods, during the short term Port Elizabeth winter study. ...............................................................................................62 Table 3.1. Environmental conditions (mean ? SD) during the four periods in which the oryx were free- living in the Mahazat as-Sayd Protected Area ............................................................................86 Table 4.1. Environmental conditions (mean ? SD) prevalent during the warm wet and hot dry periods.123 Table 4.2. Body temperature and daytime activity (mean ? SD) for five Arabian oryx and four Arabian sand gazelle during the warm wet and hot dry period (see text for statistical analyses). .............124 Table 5.1. Environmental conditions prevailing at the field site for the five hottest days, the five coldest days, and the 10-day period after summer shearing.......................................................152 xvi RESEARCH OUTPUTS Published papers Hetem R.S., Mitchell D., Maloney S.K., Meyer L.C.R., Fick L.G., Kerley G.I.H. and Fuller A. (2008). Fever and sickness behavior during an opportunistic infection in a free-living antelope, the greater kudu (Tragelaphus strepsiceros). American Journal of Physiology - Regulatory, Integrative and Comparative Physiology 294: 246-254 Hetem R.S., de Witt B.A., Fick L.G., Fuller A., Kerley G.I.H., Meyer L.C.R., Mitchell D. and Maloney S.K. (2009). Body temperature, thermoregulatory behaviour and pelt characteristics of three colour morphs of springbok (Antidorcas marsupialis). Comparative Biochemistry and Physiology A: Molecular and Integrative Physiology 152: 379-388 Hetem R.S., de Witt B.A., Fick L.G, Fuller A., Kerley G.I.H., Maloney S.K., Meyer L.C.R. and Mitchell D. Summer shearing affects body temperature of Angora goats (Capra aegagrus) more than does winter shearing. Animal 3: 1025-1036 Submitted papers Hetem R.S., Strauss W.M., Fick L.G., Maloney S.K., Meyer L.C.R., Shobrak M., Fuller A. and Mitchell D. Thermoregulation and cathemerality of free-living Arabian oryx (Oryx leucoryx) in a hyper-arid desert. Physiological and Biochemical Zoology (April 2009) International conference presentations Hetem R.S, Mitchell D., Maloney S.K., Meyer L.C.R., Fick L.G., Kerley G.I.H. and Fuller A. Fever and sickness behaviour in a free-ranging antelope, the greater kudu. Presented at the 2nd international meeting on Physiology and Pharmacology of Temperature Regulation, Phoenix, Arizona 3 - 6 March 2006 (poster presentation) xvii Hetem R.S., de Witt B.A., Fick L.G., Kerley G.I.H., Maloney S.K., Meyer L.C.R., Mitchell D. and Fuller A. Habitat transformation influences the physiology of Angora goats and may have conservation implications for wild ungulates. Presented at the 21st annual meeting of the Society for Conservation Biology, Nelson Mandela Metropolitan University, Port Elizabeth, South Africa. 1 - 5 July 2007. (poster presentation) Hetem R.S., de Witt B.A., Fick L.G., Kerley G.I.H., Maloney S.K., Meyer L.C.R., Mitchell D. and Fuller A. The effect of habitat transformation on the physiology of Angora goats. Presented at the 4th International Geosphere-Biosphere Programme Congress, Cape Town, South Africa. 5 - 9 May 2008. (poster presentation) Hetem R.S., Strauss W.M., Fick L.G., Maloney S.K., Meyer L.C.R., Shobrak M. Fuller A. and Mitchell D. Coping with extremes: thermoregulation and cathemerality of free-living Arabian oryx (Oryx leucoryx). Presented at the 4th international conference of Comparative Physiology and Biochemistry in Africa, Masai Mara, Kenya. 19-25 July 2008. (invited oral presentation) Hetem R.S., Strauss W.M., Fick L.G., Maloney S.K., Meyer L.C.R., Mitchell D., Shobrak M. and Fuller A. Coping with extremes: adaptive heterothermy and cathemerality in Arabian ungulates. Presented at the 4th international conference of Comparative Physiology and Biochemistry in Africa, Masai Mara, Kenya. 19-25 July 2008. (poster presentation) Local conference presentations Hetem R.S., Mitchell D., Maloney S.K., Kerley G.I.H., Meyer L.C.R., Fick L.G., de Witt B.A. and Fuller A. Feasibility of using remote measurements to assess ecological physiology in free-living ungulates. Presented at the 32nd conference of the Zoological Society of Southern Africa, Rhodes University, Grahamstown, South Africa. 13 - 15 July 2005. (oral presentation) xviii Hetem R.S., Strauss W.M., Fick L.G., Maloney S.K., Meyer L.C.R., Mitchell D., Shobrak M. and Fuller A. Coping with extremes: adaptive heterothermy and cathemerality in the Arabian oryx (Oryx leucoryx). Presented at the 33rd bi-annual conference of the Zoological Society of Southern Africa, North-West University, Potchefstroom, South Africa. 8 ? 11 July 2007. (oral presentation) Hetem R.S., de Witt B.A., Fick L.G., Fuller A., Kerley G.I.H., Maloney S.K., Meyer L.C.R. and Mitchell D. Cold stress in Angora goats after shearing. Presented at the 35th meeting of the Physiological Society of Southern Africa, Glenburn Lodge, Cradle of Humankind, South Africa. 9 ? 12 September 2007. (oral presentation) Hetem R.S., Strauss W.M., Fick L.G., Maloney S.K., Meyer L.C.R., Mitchell D., Shobrak M. and Fuller A. Coping with extremes: adaptive heterothermy and cathemerality in Arabian ungulates. Presented at the Southern African Wildlife Management Association Symposium, Didima, South Africa. 18 ? 21 September 2007. (poster presentation) xix AUTHOR CONTRIBUTIONS Where chapters have been modified from submitted or published papers, the contributions of each author are listed below. Chapter two Hetem R.S., de Witt B.A., Fick L.G., Fuller A., Kerley G.I.H., Meyer L.C.R., Mitchell D. and Maloney S.K. (2009) Body temperature, thermoregulatory behaviour and pelt characteristics of three colour morphs of springbok (Antidorcas marsupialis). Comparative Biochemistry and Physiology, A: Molecular and Integrative Physiology 152: 379-388. The original idea for the project was formulated from discussions with Prof. Graham Kerley. The initial three-month winter study, with the behavioural observations, formed part of Ms. Brenda de Witt?s honours (4th year) project, which I helped to supervise. We subsequently expanded the project to collect data for one year, for which I made up the equipment, organised the surgery, analysed the data. I wrote the paper encompassing both projects. Together with Ms. de Witt, I made the measurements and calculations of the heat transfers characteristics of the pelts. Dr. Shane Maloney provided guidance and invaluable expertise with analysis of the pelt heat transfer characteristics and calculations. Dr. Leith Meyer performed the surgical procedures. Dr. Linda Fick assisted with the surgery and project logistics. Prof. Andrea Fuller and Prof. Duncan Mitchell assisted with the planning and execution of the experimental work. All authors edited the manuscript. Chapters three and four Hetem R.S., Strauss W.M., Fick L.G., Maloney S.K., Meyer L.C.R., Shobrak M., Fuller A. and Mitchell D. Thermoregulation and cathemerality of free-living Arabian oryx (Oryx leucoryx) in a hyper-arid desert. Submitted to Physiological and Biochemical Zoology (April 2009) xx Hetem R.S., Strauss W.M., Shobrak M., Fick L.G., Maloney S.K., Meyer L.C.R., Fuller A. and Mitchell D. Does size matter? Comparison of body temperature and activity of free-living Arabian oryx (Oryx leucoryx) and the smaller Arabian sand gazelle (Gazella subgutturosa marica) in the Saudi desert. Intend to submit to Journal of Comparative Physiology B: Biochemical, Systemic, and Environmental Physiology The idea for these projects arose from discussions with my co-authors. I set up the collaboration between the University of the Witwatersrand and the National Wildlife Research Center, Saudi Arabia, organised surgery, prepared the equipment, analysed the data and wrote the manuscripts. Mr. Maartin Strauss made the collaboration viable, assisted with animal capture and checked on the animals when I was unable to do so. Dr. Linda Fick helped to instrument the animals and assisted with surgical logistics. Dr. Shane Maloney assisted with experimental design and helped to instrument the animals. Dr. Leith Meyer performed the surgical procedures. Dr. Mohammed Shobrak took responsibility for project logistics in Saudi Arabia. Prof. Duncan Mitchell assisted with experimental design and surgery. All authors edited the manuscript. Chapter five Hetem R.S., de Witt B.A., Fick L.G., Fuller A., Maloney S.K., Meyer L.C.R., Mitchell D. and Kerley G.I.H. Effects of habitat transformation on the physiology of Angora goats: testing global change predictions. Intend to submit to Journal of Arid Environments The idea for this project arose from discussions with my co-authors. I organised surgery, prepared the equipment, routinely caught and monitored the animals, drew blood, analysed the data and wrote the manuscript. Ms. Brenda de Witt assisted with animal capture and execution of experimental work. Dr. Linda Fick helped to instrument the animals and assisted with surgical logistics. Prof. Andrea Fuller assisted with surgery and planning of experimental work. Dr. Shane Maloney assisted xxi with experimental design and helped with analysis of water turnover. Dr. Leith Meyer performed the surgical procedures. Prof. Duncan Mitchell assisted with experimental design and surgery. Prof. Graham Kerley initiated the collaboration with Blaauwkrantz and provided an ecological perspective. All authors edited the manuscript. Chapter six Hetem R.S., Mitchell D., Maloney S.K., Meyer L.C.R., Fick L.G., Kerley G.I.H. and Fuller A. (2008) Fever and sickness behavior during an opportunistic infection in a free-living antelope, the greater kudu (Tragelaphus strepsiceros). American Journal of Physiology - Regulatory, Integrative and Comparative Physiology 294: R246? R254. The study was designed to investigate the thermal responses of kudu to climatic stress, but after release the kudu acquired a spontaneous infection. I organised surgery, prepared the equipment, monitored the animals, analysed the data and wrote the manuscript. Prof. Duncan Mitchell assisted surgery and data interpretation. Dr. Shane Maloney helped to instrument the animals. Dr. Leith Meyer performed the surgical procedures. Dr. Linda Fick helped to instrument the animals, assisted with surgical logistics and animal monitoring. Prof. Graham Kerley initiated the collaboration with Blaauwkrantz, assisted with surgical procedures and provided logistical support. Prof. Andrea Fuller assisted with surgery. All authors edited the manuscript. 1 ___________________________________________________________ CHAPTER 1 ___________________________________________________________ 1 Introduction Some of the ideas presented in this chapter have appeared in conference proceedings: Mitchell D., Fuller A., Hetem R.S. and Maloney S.K. (2008). Climate change physiology: the challenge of the decades. In: 4th Comparative Physiologists & Biochemists meeting in Africa: Mara2008. "Molecules to Migration: The Pressures of Life". Morris S. and Vosloo A. (Eds.), Medimond Publishing Co, via Maserati 6/2, 40124 Bologna, Italy, pp. 275-281. 2 During his acceptance speech for the 2007 Nobel Peace Prize, Al Gore warned that climate change is "the greatest challenge we've ever faced". Many species are unlikely to survive climate change and a pivotal study by Thomas et al. (2004) predicted that, under mid-range climate change scenarios, a quarter of terrestrial plants and animals may be extinct by 2050. By extrapolating such predictions globally, the authors predict that well over one million species could be threatened with extinction as a result of climate change. However, these bioclimatic models may overestimate future extinction risk because the majority of these models provide no measure of species' adaptability to environmental change. For long-lived species, which are limited in their dispersal capacity as a result of the modern human- dominated landscape, phenotypic plasticity provides the only option to counter extinction and cope with climate change. But whether long-lived species display sufficient phenotypic plasticity to adapt to climate change remains obscure. This introductory chapter contextualises my research within what is currently known and relevant about climate change, identifies key gaps in the research and poses the questions that are addressed in my thesis. The Intergovernmental Panel on Climate Change 2007 synthesis report predicts an increase in global temperatures of between 1.1?C and 6.4?C during the 21st century (IPCC, 2007). Such temperature projections for the end of the century may represent the warmest global climate in over 2 million years. Nevertheless, global warming has been a common occurrence throughout the Earth?s 3.5 billion year history. Modelling of the current episode predicts a temperature rise of the same order of magnitude as that evident at the end of the Permian, when mass volcanism increased global temperatures by 6?C and resulted in the extinction of nearly 95% of species (Benton and Twitchett, 2003). Since the rate of climate change may be more critical than its magnitude and duration (Davis et al., 2005), the unprecedented rate of climate change predicted to occur over the next 50 years is likely to threaten the persistence of numerous species (Davis and Shaw, 2001). Although the current rate of warming has not yet exceeded the rate of past global warming episodes (Barnosky et al., 2003), the 3 so-called ?background rate?, the current episode of warming certainly is distinct from all previous episodes in one respect: because its cause is anthropogenic, it is the first of the global warming events that could be mitigated. If, as now seems inevitable, the episode will not be mitigated adequately, we must expect another major wave of extinctions of species (Myers, 1993; McLaughlin et al., 2002; Thomas et al., 2004; Malcolm et al., 2006). Of the nearly 30 000 documented trends in physical systems and biological characteristics of plants and animals between 1970 and 2004, 90% are in the direction consistent with environmental temperature increases (Rosenzweig et al., 2008). Meta- analyses, such as those by Rosenzweig et al. (2008), have helped to establish a plausible link between recent changes in climate and observed changes in species and communities. Yet, climate change may exceed the natural response capacity of many species and different species are likely to show different climatic tolerance and to be affected differently by the arrival of new species. Thus, climate change is likely to have a significant impact on biodiversity (Malcolm and Markham, 2000; Sala et al., 2000; Lovejoy and Hannah, 2005), and current conservation strategies, which attempt to conserve communities and ecosystems as they currently exist, will not be effective in the face of climate change (Hughes, 2000; Hannah et al., 2002a; Hannah et al., 2002b). We therefore need a better understanding of the potential impacts of climate changes on living organisms so that we may better anticipate the future responses of plant and animal populations (Millien et al., 2006) and better conserve biodiversity. Simplistically, species have only three prospective scenarios when faced with environmental changes. Firstly, species may migrate or shift their current distribution range to more suitable habitats where the climate is within the species? tolerance limits. Secondly, animals may adapt to regional variation in climatic regimes either through a change in the genetic composition of the population or by a change in the expression of that genotype, so called phenotypic plasticity. Either of these adaptation strategies may result in changes in the timing of events (phenology), morphology (e.g. 4 colour patterns, body shape and size) or physiology of a species. Thirdly, if neither migration nor adaptation is attainable, extinction will result. 1.1 Range shifts One of the predicted responses to climate change is that individuals will track suitable climates. In the temperate zone, for example, a 1?C increase in mean annual temperature is predicted to correspond to a shift in isotherms of ~ 160 km in latitude or 160 m in elevation (Parmesan and Yohe, 2003; Thuiller, 2007). Thus, species are expected to follow these shifting climatic zones and move polewards in latitude and upwards in elevation (Hughes, 2000; Walther et al., 2002). These predictions are based largely on palaeoecological evidence as numerous mammalian species tracked consistent climate profiles as climates warmed during the Quaternary Period (Graham and Grimm, 1990; Potts and Deino, 1995; Graham et al., 1996; Lyons, 2003; Martinez-Meyer et al., 2004). Such biotic responses imply that natural systems were resilient to past climate change and re-colonization following glacial retreat often exceeded expected rates of dispersal, as estimated from mean dispersal and time to reproductive maturity (Hannah et al., 2002a). Numerous recent reports have documented shifts in the geographical distribution of extant species (for review see Parmesan and Yohe, 2003; Root et al., 2003) and these observations parallel expectations based on observations of warming events of the past (Barnosky et al., 2003). In the Northern hemisphere, for example, recent northward movement of species? range boundaries has been observed in numerous terrestrial taxonomic groups, including plants (Grabherr et al., 1994; Walther et al., 2005), butterflies (Parmesan, 1996; Hill et al., 1999; Parmesan et al., 1999), birds (Thomas and Lennon, 1999; Brommer, 2004; La Sorte and Thompson, 2007), reptiles and amphibians (Pounds et al., 1999), and mammals (Hersteinsson and MacDonald, 1992; Myers et al., 2009). Global meta-analyses revealed that about 80% of these species range shifts are consistent with climate change predictions (Parmesan and 5 Yohe, 2003; Root et al., 2003). However, of the 105 reports of range shifts in these meta-analyses, only three were included from the Southern hemisphere. The dearth of data from long-term research in the Southern hemisphere (Midgley et al., 2007; Rosenzweig et al., 2008) is of concern because of hemisphere asymmetry in ecosystem processes. Whereas ecosystem processes are largely controlled by energy-related factors in the northern hemisphere, water-related factors appear to control ecosystem processes in the Southern hemisphere (Hawkins et al., 2003; Chown et al., 2004b). Such hemisphere asymmetries are likely to be exacerbated with climate change. While many northern species are expected to expand their ranges northwards in response to an increased primary productivity (Nemani et al., 2003), longer growing season and less severe winters, the increased aridity predicted for many arid and semi-arid regions is likely to further limit the distributions of southern hemisphere species. 1.1.1 Bioclimatic envelope models Although observed range shifts from long-term data sets in the Southern hemisphere are limited (Midgley et al., 2007), various envelope models have attempted to predict species? distributions with climate change (Erasmus et al., 2002; Midgley et al., 2002; Meynecke, 2004; McClean et al., 2005). Niche-based or bioclimatic envelope models are essentially static models that correlate current species distributions with climate variables and project future spatial shifts in species? climatic envelopes (Peterson, 2003; Huntley et al., 2004; Austin, 2007). Although these models may provide a useful first estimate of the potential impact of climate change on the distribution of species, their validity has been questioned recently because of the assumptions on which they are based (Pearson and Dawson, 2003; Thuiller, 2004; Luoto et al., 2005; Heikkinen et al., 2006). The first assumption is that climate is the key limiting factor on species distributions (Ara?jo and Luoto, 2007). For simplicity, most of these models predict species range shifts based on predicted average temperature changes, but species may prove to be more sensitive to changes in precipitation than to 6 changes in average temperature, as has been found for birds, amphibians and reptiles (Pounds et al., 1999). In addition, unlike the average temperature increases which are incorporated into the majority of these model predictions, it is likely that the predicted increase in weather extremes, such as droughts and heat waves, will impose the dominant stress on species in the future (Parmesan et al., 2000; Hallett et al., 2004). Furthermore, these bioclimatic envelope models do not account for the so-called ?non-climatic? influences on species? distributions, such as geology and biotic interactions. Biotic interactions between species are often overlooked in bioclimatic modelling, yet such species interactions are likely to have important consequences for future species distributions (Davis et al., 1998; Mustin et al., 2007). For example, the climate-driven northward range expansion of the red fox (Vulpes vulpes) has been associated with a decrease in the distribution range of the arctic fox (Alopex lagopus) as a result of an increased interspecific competition (Hersteinsson and MacDonald, 1992). Since individual species differ in their response to changing climatic conditions, species may shift their ranges independently of each other which would result in changes in community structure and ultimately lead to ecosystem disruption (Schneider and Root, 1996; McCarty, 2001; Stenseth et al., 2002; Walther et al., 2002). Such secondary effects of climate change may be more severe than the direct consequences of increased temperature (Humphries et al., 2004) and investigation of such biotic interactions has lead to a new scope of ecological research termed ?global change ecology? (Schlesinger, 2006). These responses need to be incorporated into bioclimatic models to better predict future species distributions (Guisan and Thuiller, 2005). An additional non-climatic factor which is likely to impede the movement of many species is the modern human-dominated landscape. Human activity has resulted in habitat loss and fragmentation, and this process is likely to be exacerbated by climate change as previously fertile areas are transformed. The impassable obstacles created 7 by human activity lower the likelihood of species finding compatible distant habitats to colonise, ultimately increasing the risk of extinction. Under mid-range climate change scenarios, climate envelope models predict that 45% of terrestrial species are likely to be committed to extinction by 2050 if their dispersal is limited (Thomas et al., 2004). Similarly, 25-40% of a representative sample of 277 African mammalian species are likely to be critically endangered or extinct by 2080 (Thuiller et al., 2006). South Africa?s mammalian species are predicted to be particularly sensitive to climate change because the majority of national parks are fenced, further limiting the dispersal abilities of large species. The extinction risk of South African mammals is estimated to be as high as 69% by 2050, if dispersal is limited (Thomas et al., 2004). South Africa?s flagship Kruger National Park, for example, may lose two thirds of its current animal species (Erasmus et al., 2002) and long term population monitoring has already observed declines in seven out of 11 ungulate species studied in the Kruger National Park between 1977 and 1996 (Ogutu and Owen-Smith, 2003). Not only do fences limit dispersal abilities of large mammals, but in some cases the future location of a suitable bioclimatic envelope may be impossible to reach. One such example where range shift is unlikely to be a feasible option is the scimitar- horned oryx (Oryx dammah, Fig. 1.1). In order to track its current bioclimatic envelope, this species will have to move thousands of kilometres, from the Sahara desert to Namibia, a distance unattainable without human assistance (Thuiller et al., 2006). In such circumstances where range shifts are unfeasible, either as a result of unattainable travelling distances or loss of habitat connectivity, assisted colonization may provide a conservation option in the future (Hunter, 2007; McLachlan et al., 2007; Hoegh- Guldberg et al., 2008). Yet, moving species to areas where they do not currently occur is not without risk. The introduced organisms can carry disease, displace native species and thereby challenge ecosystem stability or alter the genetic structure of local populations. Therefore, an in-depth knowledge of species? biology and accurate climate change predictions would be required before assisted colonization can become a feasible conservation option (Davidson and Simkanin, 2008). 8 Figure 1.1. Observed current (small map) and predicted (large map) distribution of the scimitar-horned oryx (Oryx dammah) for 2050. Light grey areas indicate current climatically suitable habitats predicted to be unsuitable in the future. Moderate grey areas indicate the current climatically suitable habitats which are predicted to stay suitable. Dark grey areas indicate the current climatically unsuitable habitats which are predicted to be suitable by 2050 (adapted from Thuiller et al., 2006). Another limitation of bioclimatic envelope models is the assumption that species lack sufficient plasticity to adapt to climates beyond those under which a given species occurs today (Levinsky et al., 2007). Yet, physiological plasticity may allow animals to adapt to changing climatic conditions. Some bioclimatic envelope models have attempted to incorporate physiological limitations to determine the climatic requirements of a diverse group of taxa including plants (Prentice et al., 1992; Sykes and Prentice, 1996; Dole et al., 2003), invertebrates (Crozier and Dwyer, 2006; Musolin, 2007), marine (P?rtner et al., 2004; Gilman et al., 2006; Osovitz and Hofmann, 2007) and terrestrial ectotherms (P?rtner, 2001; Kearney and Porter, 2004), and mammals (Taulman and Robbins, 1996; Johnston and Schmitz, 1997). Such models require an understanding of species? physiological responses to climate (Franklin, 1995; Mack, 1996; Guisan and Zimmermann, 2000), yet we lack an understanding of the ecophysiological tolerances of many species. Although these physiological models still have limitations, since they also do not take non-climatic factors into account, they are considered to be superior and more robust than those bioclimatic envelope models based on correlations between observed distributions Loss of suitable habitat Stable suitable habitat Gain of suitable habitat 9 and current climate variables (Prentice et al., 1992; Sykes and Prentice, 1996; Hodkinson, 1999). Despite such limitations, bioclimatic envelope models reveal the potential threat that climate change poses on global biodiversity. These models, however, may overestimate future extinction risk because the majority of these models provide no measure of species' adaptability to environmental change. Natural selection, for example, may allow animals to adapt to changing climatic conditions. However, evolutionary change is often considered a slow process and the current rate of climate change may be too fast to allow evolutionary adaptation, as such, bioclimatic envelope models assume that the tolerance range of a species remains stable as it shifts its geographic range (Bennett, 1990; Jackson and Overpeck, 2000). Despite such predictions, evolutionary change is likely to have accompanied range shifts in the past, since changing climates would change the so-called ?fitness optimum? for different populations throughout the species range (Davis and Shaw, 2001; Davis et al., 2005), making the fundamental niche unstable over time. Indeed, numerous studies have documented morphological changes in response to past climate change events (Martin and Barnosky, 1993). 1.2 Microevolution In combination with shifting species ranges, morphological changes are believed to constitute the first response of mammals to climate change (Barnosky et al., 2003). In general, the extent of the morphological adaptation seen in the past fell within the range of morphological variation found in the species across its current geographical range (Davis and Shaw, 2001; Huntley, 2007). Thus, evolutionary responses require genetically based variation among individuals and range shifts may act to increase gene flow, thereby increasing a species chance of adapting to changing conditions. Northern range shifts, for example, may have the advantage of introducing genotypes that are better adapted to warmer conditions, thus promoting adaptation of existing 10 populations (Garant et al., 2007; Visser, 2008). Conversely, range shifts may also decrease genetic variability as a result of outbreeding of distinct populations. For example, climate change may result in genetic mixing among subspecies of the black bear, which is likely to inhibit or even reverse sub-speciation (Kerr and Packer, 1998). True speciation events may take millennia, leading to the perception that genetic adaptation is a slow process relative to climate change (Bradshaw and McNeilly, 1991; Etterson and Shaw, 2001; Davis et al., 2005). Although the current global warming episode probably does not yet exceed the normal background rate, continued warming over the next few decades will exceed the background rate of change by one or more orders of magnitude (Barnosky et al., 2003). With the unprecedented rate of climate change anticipated to occur in the future it becomes questionable whether evolutionary adaptation will be sufficient to mitigate the effects of climate change (Bradshaw and McNeilly, 1991; Davis and Shaw, 2001; Huntley, 2007). Nevertheless, a theoretical analysis recently concluded that animals must use the option of genetic adaptation to survive climate change (Rice and Emery, 2003; Visser, 2008). Such conclusions are based on evidence that genetic adaptation need not involve the slow process of speciation. Evidence has been advanced that ?microevolution?, that is heritable shifts in allele frequencies in a population (without speciation), already has occurred, in directions predicted by climate change (Bradshaw and Holzapfel, 2006; Bradshaw and Holzapfel, 2008). Evidence for rapid evolutionary change is particularly prevalent for short-lived species with fast generation times (for examples, see Rodr?guez-Trelles and Rodr?guez, 1998; Thompson, 1998; Hendry and Kinnison, 1999; Kinnison and Hendry, 2001; Reznick and Ghalambor, 2001; Yoshida et al., 2003). Such rapid evolutionary change may be an important adaptation to climate change. For example, two species of bush crickets, Conocephalus discolor and Metrioptera roeselii, are becoming more dispersive because of an increase the proportion of longer-winged (more dispersive) individuals, in Britain over the past 20 years 11 (Thomas et al., 2001a). In addition to changes in the size of appendages, ecogeographical rules predict that animals may adapt to local conditions and changing climates by displaying variation in colour (Millien et al., 2006). For example, the Western White Butterfly, Pontia occidentalis, in the United States and Canada, displays seasonal plasticity in wing melanin, which is believed to enhance the butterflies? thermoregulation in seasonally varying thermal environments (Kingsolver and Huey, 1998). Although an animal?s colour, in itself, may have thermoregulatory advantages, often an animal?s colour has fitness advantages because of an association with other morphological features. For example, the coat colour in Soay sheep is related to fitness because of an association with body size (Clutton- Brock et al., 1997), and it is because of this genetic association between coat colour and body size that homozygous dark coloured sheep have decreased in frequency over a 20-year time period (Gratten et al., 2008). However, analyses by Maloney et al. (2009) imply that increases in ambient temperatures may provide a more parsimonious explanation for the recent decline in dark-coloured Soay sheep. These authors suggest that recent warming trends would offset the potential metabolic savings associated with dark colouration, negating any selective advantage for dark- coloured sheep. An additional ecogeographical rule which has been used to predict how animals will respond to climate change is Bergmann?s rule. Bergmann?s rule predicts a positive correlation between body mass of terrestrial endotherms and latitude, and, by inference, an inverse correlation between body mass and environmental temperature. With current models predicting an increase in mean surface air temperature, Bergman?s rule predicts that smaller individuals would be favoured in a warmer world (Millien et al., 2006). Indeed, several species of passerine birds caught in ringing operations in England over 25 recent years (Yom-Tov et al., 2006) and museum specimens from Israel over 50 recent years (Yom-Tov, 2001) have exhibited a linear decrease in body mass. Similarly, the body mass of the bushy-tailed woodrat (Neotoma cinerea) has decreased with recent increases in temperature in New Mexico 12 (Smith et al., 1995; Smith et al., 1998; Smith and Betancourt, 2006). Although these trends have been interpreted as microevolutionary responses to climate change, the majority of these studies provided no evidence that the observed trends had a genetic basis (Teplitsky et al., 2008; Wolf et al., 2009). The general lack of genetic evidence to account for such morphological variations, at least partially, is because molecular genetic techniques remain inadequate to properly reveal how genetic sequences relate to ecological importance (Holt, 1990; Gienapp et al., 2008). Similar to morphological variations, variations in phenology, or the timing of life- history events, have also been attributed to a climate-driven evolutionary response. The changing migration pattern in German blackcaps (Sylvia atricapilla), for example, has been attributed entirely to evolution (Berthold and Pulido, 1994; Bearhop et al., 2005). Yet, to date, there are only two studies which have tested for the genetic basis of breeding time advancements in birds, and neither of these found evidence for a genetically-based response (Sheldon et al., 2003; Gienapp et al., 2006). The most convincing example of microevolutionary response to climate change has been provided by data from 325 marked female North American red squirrels (Tamiasciurus hudsonicus) in Yukon from 1989 to 2001, a period over which mean lifetime parturition date advanced by six days per generation, associated with a mean spring temperature rise of 2?C and a decreased precipitation at the location (R?ale et al., 2003a). R?ale et al. (2003b) used an ?animal-model? to demonstrate that 13% (0.8 days per generation) of the observed phenological changes in parturition date could be attributed to microevolution. However, since R?ale et al. (2003b) did not account for systematic environmental variation across years, they actually may have overestimated the amount of genetic change (Postma, 2006). Thus, potentially more than 60% of the observed changes in parturition date of the squirrel could be attributed to phenotypic plasticity. 13 1.3 Phenotypic plasticity By definition, phenotypic plasticity is the process by which a single genotype gives rise to more than one phenotype when exposed to different conditions (Via et al., 1995; Pigliucci, 2005; Garland and Kelly, 2006). Phenotypic plasticity to a change in environment may invoke acclimatization, acclimation and learning (Garland and Kelly, 2006). Thus, unlike genetic adaptation, phenotypic plasticity allows an organism to respond rapidly to environmental change and is thus likely to be an organism?s primary response to the rapidly changing environmental conditions predicted for the future. Although the red squirrel study provides the first quantification of the role of phenotypic plasticity in climate-induced development of a functional character (R?ale et al., 2003b), it is not the first to document changes in phenology, or the timing of seasonal events, in response to changing climatic conditions (Harrington et al., 1999a; Hughes, 2000; Walther et al., 2002). In a meta- analysis of nearly 700 species, spring phenology shifted between 2.3 days (Parmesan and Yohe, 2003) and 5.1 days (Root et al., 2003) earlier on average per decade over the last 50 years. Such phenological changes are often considered as evidence that species are adapting to the changing environmental conditions. For example, over nearly 50 years, during which springs have been warming, the laying dates of great tits (Parus major), near Oxford in the United Kingdom, have advanced by 14 days, a change closely matched to earlier peaks in food insect biomass, and attributed entirely to phenotypic plasticity (Charmantier et al., 2008). However, unlike the great tits in Britain, those in the Netherlands have not advanced their phenology to match the peak in caterpillar availability (Visser et al., 1998; Visser et al., 2003). A similar mismatch between food abundance and offspring needs within the Dutch population of pied flycatchers (Ficedula hypoleuca) led to a 90% decline in the population over the last two decades (Both et al., 2006). Thus species may show different rates of phenological change than the species on which they depend, potentially resulting in asynchrony or a mistiming of key ecological events (Hughes, 2000; Thomas et al., 2001b; Visser et 14 al., 2004; Visser and Both, 2005). Numerous studies have demonstrated the ecological and metabolic costs of these mistimed ecological events (Thomas et al., 2001b; Stenseth and Mysterud, 2002), which may ultimately lead to loss of biodiversity. One possible cause of such a mismatch is the different environmental cues to which different species respond (Voigt et al., 2003). For example, whereas most plants and insects respond to seasonal changes in temperature, most vertebrate species are more sensitive to changes in photoperiod. Thus many vertebrates electing, or forced, to stay at the historic locations may face a mismatch between photoperiod and food availability, disrupting reproduction and migration, and those dispersing latitudinally will have to adjust to an unfamiliar annual cycle of photoperiod in their new habitat (Bradshaw and Holzapfel, 2008; Visser, 2008). Species which are unable to match the timing of key life-history events to the phenology of the species on which they depend may be forced to show plasticity in other life-history traits if they are to maintain their lifetime reproductive success. For example, flexibility in phenology of the Antarctic fur seal (Arctocephalus gazelle) is likely to be limited because of a long interval between conception and weaning of the pups and a highly variable environment. Instead, female Antarctic fur seals appear to be adapting their life cycles by not breeding in years of low krill supply, thus increasing adult survival and fitness (Forcada et al., 2008). Another species which is changing its life-history strategy in response to stochastic environmental conditions is the pronghorn antelope (Antilocapra americana). Frequent severe weather events result in an increase in male mortality, which favour precocial maturation in male pronghorn and may ultimately lead to a life-history strategy of faster development (Mitchell and Maher, 2006). Not only can severe weather events directly affect the life-history strategy of male antelope, but the climatic conditions to which a female is exposed during her pregnancy can also influence life-history traits of her offspring, the so-called ?maternal effects? (Bernardo, 1996). 15 Body mass, for example, is a major determinant of lifetime reproductive success and numerous species of antelope in the Northern hemisphere display plasticity of offspring birth mass in response to changing climatic conditions (Stenseth et al., 2002). The high costs of thermoregulation during wet and windy winters is proposed to compromise the condition of pregnant ewes, to the extent that individuals born after deep-snow winters are consistently smaller and have a lower lifetime reproductive success than those born after cold dry winters (Post et al., 1997; Post and Stenseth, 1998; Forchhammer et al., 2001). Although these maternal effects may promote survival and enhance reproductive success of the mother, such plasticity in birth mass has long-term consequences for the offspring. Like many morphological traits, body mass at birth is a ?non-labile? trait as it is expressed only once in an individual?s lifetime (Nussey et al., 2007). The majority of ?non-labile? traits are traits which show plasticity only during development. House sparrows, for example, show morphological plasticity during development and fledglings can alter the lipid composition of the outer layer of their skin, the stratum corneum, depending on the water vapour pressure of their environment (Mu?oz-Garcia and Williams, 2008). Since the lipid composition of the stratum corneum affects the cutaneous water loss (Mu?oz-Garcia and Williams, 2005), these fledglings are able to reduce their cutaneous water loss, and thus conserve body water, when exposed to dry desert environments. However, such ?non-labile? plasticity can only be adaptive if the trends for changes in climatic conditions at the time of development remain similar throughout the individuals? lifetime. Although understanding the physiological mechanisms underlying changes in life- history traits is essential for predicting responses to climate change (Visser and Both, 2005; Visser, 2008), the vast majority (> 80%) of studies of phenotypic selection have focused on morphological plasticity without incorporating physiological and behavioural responses to climate change (Kingsolver et al., 2001). Such a preponderance may not reflect a true biological trend but rather represent the ease of recording of morphological features, like body mass. Gathering physiological and 16 behavioural data, on the other hand, is labour-intensive and requires long periods of observation and monitoring. Nevertheless, given that natural selection primarily works at the level of physiology and behaviour (Berteaux et al., 2004) it is concerning that we understand so little about the direct links between a species? physiology and its vulnerability to climate change. We need to improve our understanding of the physiological mechanisms which determine an animal?s thermal tolerance and its acclimatory capacity in order to better predict the impact of climate change on a particular species (McCarty, 2001; Stillman, 2003; Helmuth et al., 2005; Calosi et al., 2008). 1.3.1 Physiological adaptation That physiological mechanisms are responsible for the capacity of animals to adjust to new environments sounds self-evident, but according to Carey (2005) the concept is just over 50 years old; she attributes it to Adolf (1956). In order to accurately predict the direct physiological effects of climate change on a species, two elements are required. Firstly, we need an understanding of the thermal physiological sensitivity of the species and how close the species is to its thermal limit. Secondly, we need an understanding of the relationship between climate and body temperature experienced by the species and the degree to which the species can adjust, or acclimatize, its thermal sensitivity (Stillman, 2003; Gilman et al., 2006). Because of the clearly defined thermal niches which occur in the marine environment, the majority of studies which have investigated the physiological principals underlying thermal limits and thermal sensitivity have focused on marine ectotherms. Marine ectotherms have contributed substantially to our understanding of the key mechanisms of thermal adaptations and limitations, including the finding that thermal tolerance windows are limited by the capacity of circulatory and ventilatory tissues (P?rtner, 2004). Because of similarities in the molecular and cellular mechanisms of oxygen supply, numerous animal phyla are likely to show similar limitations in their capacity for adaptation (P?rtner, 2001; P?rtner, 2002). Even endotherms remain 17 fundamentally, and potentially detrimentally, affected by temperature (Post and Stenseth, 1999; Porter et al., 2000; Humphries et al., 2004). With the temperatures predicted to occur in the future, endotherms may well reach their thermal limits. In fact, endotherms may be more sensitive than ectotherms to rising ambient temperatures, since endothermy evolved during cold climatic conditions (P?rtner, 2004) and enhanced organismic complexity is often accompanied by increased thermal sensitivity (P?rtner, 2002). Theoretically, ectotherm physiology is superior to that of endotherms in warm climates as they are able to stay active for longer in the heat (Martin and Nagy, 2002), therefore ectotherms will be less compromised by global warming than global cooling. For example, a warmer climate is proposed to favour the Australian nocturnal lizard (Heteronotia binoei) and by 2100 its range is predicted to expand by up to 170 km as a result of the increased number of degree- days for egg development and the increased potential time available for activity (Kearney and Porter, 2004). In addition, these lizards may well be able to buffer any adverse effects associated with the increased metabolic costs of high body temperatures through appropriate thermoregulatory behaviour (Kearney et al., 2009), as nocturnal ectotherms are known to select thermally suitable sites within the soil profile (Kearney and Predavec, 2000). Nevertheless, whether endothermic or ectothermic, thermal tolerance windows differ between species and are likely to determine the range of environmental temperatures under which species can survive (P?rtner, 2002). Generalist species, with large geographic ranges, wide thermal tolerance windows and greater physiological plasticity, are less likely to be affected by climate change than are species which are physiologically specialised with respect to temperature (Potts, 1996; Bale et al., 2002; Visser and Both, 2005; Codron et al., 2008). Indeed, terrestrial ectotherms in the tropics appear to be operating near the upper limit of their thermal range and are believed to be particularly sensitive to climate change because they are physiologically specialised with respect to temperature and have a limited acclimation capacity (Addo-Bediako et al., 2000; Deutsch et al., 2008; Tewksbury et 18 al., 2008). Similarly, of the 71 European bird species monitored during the 2003 French heat wave, those species with small thermal ranges showed the sharpest decrease in population growth (Jiguet et al., 2006). Mammalian species with narrow thermal tolerances, such as tropical arboreal marsupials (Isaac et al., 2009), also are likely to be adversely affected by climate change. For example, the white lemuroid possum (Hemibelideus lemuroids), a species endemic to the mountain forests of northern Queensland, cannot maintain its body temperature when exposed to heat and risks extinction as a result of the recent 0.8?C increase in ambient temperature. Conversely, mammalian species with wide thermal tolerances, such as the hibernating mammalian species in the Canadian Arctic region (Humphries et al., 2002; Humphries et al., 2004), are predicted to show an increase in abundance and distribution in response to climate change. In summary, climate change is therefore predicted to favour species with wide thermal tolerances, short generation times and genotypic variability (P?rtner and Farrell, 2008). Although studies on short-lived species have contributed substantially to our understanding of the importance of physiological plasticity, long-lived species are likely to respond differently to changes in environmental conditions. Long-term population growth models predict that, relative to long-lived species (such as perennial plants, birds, ungulates), short-lived species (such as insects and annual plants and algae) may be particularly vulnerable to the predicted increase in climatic variability because enough members of the species must successfully survive and reproduce each year (Morris et al., 2008). However, short-lived species have the advantage of fast generation times, which may improve their chance of survival as they can undergo microevolution (Hendry and Kinnison, 1999; Hendry and Kinnison, 2001; Kinnison and Hendry, 2001; Berteaux et al., 2004; Visser, 2008). Large-bodied species with long generation times are predicted to have less ability to respond genetically to new selective pressures (Rosenheim and Tabashnik, 1991), possibly making them more susceptible to extinction (Gomulkiewicz and Holt, 1995), than species with short generation times. Indeed, numerous studies have demonstrated that 19 large mammalian species are usually more vulnerable to extinction than are smaller species because of slower reproductive rates and greater range requirements (Cardillo et al., 2005). There are many species of mammals with longevity sufficient such that individuals alive now ought still to be alive in 2030, and a few species for which individuals alive now could be alive in 2100. Clearly, the survival of those individuals, and probably those species, cannot depend on genetic adaptation. Instead these species are likely to be entirely dependent on physiological plasticity to survive climate change, because, unlike developmental plasticity, physiological plasticity is ?labile? and has the advantage of changing numerous times within an individuals? lifespan. One example of a long-lived mammalian specialist which is likely to be severely affected by climate change is the polar bear (Ursus maritimus). Without a radical change in their behaviour, the future survival of polar bears is considered bleak (Berteaux et al., 2004). They are heavily dependent on Arctic spring ice, because that is where they discover the seals (on ice to give birth) that are their primary food source at this time (Derocher et al., 2004; Stirling and Parkinson, 2006). The Arctic ice is disappearing under the impact of global warming, and, if they continue with their current lifestyle, the world population of polar bears is likely to drop by two- thirds by 2050 (Amstrup et al., 2007). Polar bears may well survive if they have the capacity to make a major change in lifestyle (which the fossil record shows they have done previously), namely to abandon the ice, and their current food source, and to become land-based (Derocher et al., 2004). An Arctic species which is already showing such plasticity in behaviour is the humpback whale. Some humpback whales (Megaptera novaeangliae) are abandoning their migration habits and remaining in southeast Alaska throughout winter, seemingly in response to the availability of herring (Moore and Huntington, 2008). Although there is likely to be an energetic cost of thermoregulation in the cold waters, such costs could be offset by metabolic savings of not having to undertake one of the longest documented mammalian migrations (Rasmussen et al., 2007). 20 1.3.1.1 Artiodactyls An order of long-lived mammals which may be pre-adapted to the future hot and dry conditions predicted for large regions of Africa is the Artiodactyla. These even-toed ungulates have been evolutionarily successful and hugely speciose, with 90 extant genera. The artiodactyls evolved during the Eocene and speciated during a warming period in the Mid-Miocene Climatic Optimum (Barnosky et al., 2003). Since these taxa evolved under hot and dry conditions of the past, and the first appearance of an artiodactyl correlates with the onset of such a warming trend, most artiodactyls are likely to be adapted to warmer and drier environments than they currently experience (Barnosky et al., 2003; Mitchell and Lust, 2008). In fact, when Burns et al. (2003) modelled 213 mammalian species losses for national parks in the United States in response to climate change, the Artiodactyla were the only order in which no species losses were predicted. Although such a prediction may be because artiodactyls represented a small percentage (6.7%) of the total current species found within the parks, it may be that these species are better adapted to changing climatic conditions. Traditionally, the success of artiodactyls has been attributed to improved locomotion and the evolution of a ruminant digestive tract, which enabled artiodactyls to use widely dispersed forage with a high fibre content (Janis, 1989; Codron et al., 2008). More recently, however, it has been suggested that the current prevalence of artiodactyls could be attributed to the evolution of the carotid rete (Mitchell and Lust, 2008). The carotid rete is an anatomical structure consisting of an intertwining network of arteries which lies within a venous sinus at the base of the brain. Cool venous blood from the nasal mucosa drains into the sinus and cools the arterial blood destined for the brain, resulting in selective brain cooling (Baker, 1982; Mitchell et al., 1987). Selective brain cooling is defined as the reduction of brain temperature below arterial blood temperature (Fig. 1.2, IUPS Thermal Commission, 2003). Although selective brain cooling originally was believed to protect the brain from a rise in body temperature (Mitchell et al., 1987), recent literature proposes that 21 artiodactyls may employ selective brain cooling to attenuate the drive for evaporative cooling. Selective brain cooling results in a reduction in hypothalamic temperature, which reduces evaporative heat dissipation and ultimately conserves water (Jessen et al., 1994; Mitchell et al., 1997; Fuller et al., 1999; Maloney et al., 2002; Mitchell et al., 2002). Such a strategy would have obvious advantages in an environment where water resources are both limited and variable (Mitchell et al., 2002). 37 38 39 40 selective brain cooling no selective brain cooling Blood temperature (?C) B ra in te m pe ra tu re ( ?? ??C ) 40 39 38 37 Figure 1.2. Mean brain temperature at each 0.1?C category of arterial blood temperature for a single oryx (Oryx gazella) over a time period of 15 days. Each point shows mean ? SD. The lines directly above and below the data points represent the maximum and minimum, respectively, recorded brain temperature for each category of blood temperature. The dashed diagonal line marks the line of identity between blood and brain temperatures; all points below the line represent periods of selective brain cooling (adapted from Maloney et al., 2002). If selective brain cooling does indeed increase body temperature by attenuating thermal drive for evaporative heat loss, it may contribute to another physiological mechanism, heterothermy. Adaptive heterothermy, like selective brain cooling, is proposed to reduce evaporative cooling by desert ungulates. It is defined as the storage of body heat during the day, with a consequent rise in body temperature, to reduce both heat gain and evaporative heat loss. This stored heat then can be dissipated non-evaporatively during the colder night, when heat conservation and 22 generation is also suppressed, allowing body temperature to fall so that the scope for heat storage on the day is increased (Fig. 1.3). By actively widening the thermoregulatory dead band under extreme conditions, animals may be able to reduce the energetic costs of thermoregulation (Dawson, 1992) and such a strategy is likely to become increasingly important as conditions get hotter and drier with climate change. 0:00 12:00 0:00 12:00 35 37 39 41 heterothermy homeothermy Time of day B o dy te m pe ra tu re (?C ) Figure 1.3. Body temperature rhythm of a well-hydrated (grey line) and dehydrated (black line) camel over a two day period. The well-hydrated camel displayed greater homeothermy than did the dehydrated camel, which displayed an increased maximum and decreased minimum body temperature with a resultant increase in daily body temperature fluctuation. Such large fluctuations of body temperature of about 6?C, without a major change in mean 24h temperature, have been defined as adaptive heterothermy and heat stored as a result of the large increase in body temperature corresponds to a saving of nearly 5 L of water (adapted from Schmidt-Nielsen, 1957). Yet, the presence of adaptive heterothermy rarely is evident in large free-ranging mammals exposed to hot arid environments, leading many authors to argue that it may be an artefact associated with laboratory animal studies (Finch, 1972; Hofmeyr and Louw, 1987; Mitchell et al., 1997; Fuller et al., 2000; Maloney et al., 2002; Mitchell et al., 2002). Table 1.1 summarises the findings of numerous studies of body temperatures in unrestrained artiodactyls over the last 50 years. 23 Table 1.1. Body temperatures of unrestrained wild artiodactyls Body temperature (?C) Species Mass (kg) Ambient temperature (?C) mean minimum maximum amplitude Body site Location Reference 29?44 38.4?1.3 36.5?1.2 40.5?0.7 4.1?1.7 abdominal Saudi Arabia (Ostrowski et al., 2003) Arabian oryx (Oryx leucoryx) 88-100 13?27 38.4?1.3 37.5?0.5 39.2?0.3 1.5?0.6 abdominal Saudi Arabia (Ostrowski et al., 2003) 25-41 39.5?0.3 38.2?0.5 40.8?0.4 2.6?0.8 abdominal Saudi Arabia (Ostrowski and Williams, 2006) Arabian sand gazelle (Gazella subgutturosa marica) 12-20 11-26 39.4?0.2 38.6?0.3 40.3?0.3 1.7?0.3 abdominal Saudi Arabia (Ostrowski and Williams, 2006) Black wildebeest (Connochaetes gnou) 130 16-31 38.6?0.0 37.0 39.7 0.8 carotid artery South Africa (Jessen et al., 1994) Blue wildebeest (Connochaetes taurinus) 120- 250 - 39.3?0.2 38.4 39.8 0.9?0.1 abdominal South Africa (Fick et al., 2006) 24 Table 1.1. Continued Body temperature (?C) Species Mass (kg) Ambient temperature (?C) mean minimum maximum amplitude Body site Location Reference Boran cattle (Bos indicus) - - - 36.1 39.8 - deep muscle Kenya (McGinnis et al., 1970) Buffalo (Syncerus caffer) 350 10-25 38.7 36.9 40.1 - lower neck tissue Kenya (Bligh and Harthoorn, 1965) Camel (Camelus dromedarius) 450- 500 21-35.5 37.1?0.4 35.1 39.1 - hump tissue Kenya (Bligh and Harthoorn, 1965) 27-44 - 37.5 40.9 - vaginal AZ (USA) (Zervanos and Hadley, 1973) Collared Peccary (Tayassu Tajacu) 12-30 4-18 - 37.5 40.3 - vaginal AZ (USA) (Zervanos and Hadley, 1973) 25 Table 1.1. Continued Body temperature (?C) Species Mass (kg) Ambient temperature (?C) mean minimum maximum amplitude Body site Location Reference 200 14-27 39.1 38.3 39.9 - neck and back tissue Uganda (Bligh and Harthoorn, 1965) 330- 380 6-27 38.2?0.1 37.1 39.4 2.3?0.4 arterial blood Namibia (Fuller et al., 1999) Eland (Tragelaphus oryx) - - - 37.0 40.8 - deep muscle Kenya (McGinnis et al., 1970) 120- 180 15-30 38.8?0.1 37.0 40.5 1.8?0.3 carotid artery South Africa (Maloney et al., 2002) Gemsbok (Oryx gazella) 150 12-26 38.5 36.6 40.0 - lower neck tissue Kenya (Bligh and Harthoorn, 1965) Giraffe (Giraffa camelopardalis) 400 10-22 38.5 37.8 39.1 - lower neck tissue Kenya (Bligh and Harthoorn, 1965) Kongoni (Alcelaphus buselaphus cokei) - - - 37.3 39.7 - deep muscle Kenya (McGinnis et al., 1970) 26 Table 1.1. Continued Body temperature (?C) Species Mass (kg) Ambient temperature (?C) mean minimum maximum amplitude Body site Location Reference Mule deer (Odocoileus hemionus) - 11-31 37.6 36.2 42.1 - deep muscle Wyoming (USA) (Thorne, 1975) 40-50 2-38 38.6?0.3 36.4?0.8 39.5?0.7 3.1?0.4 carotid artery Wyoming (USA) (Lust et al., 2007) - -7.3-4.5 38.5?0.4 37.8?0.2 39.7?0.7 1.8?0.7 carotid artery Wyoming (USA) (Hebert et al., 2008) - 11.7-30.5 37.6 36.2 42.2 - deep muscle Wyoming (USA) (Thorne, 1975) - 5-22 38.5 - - 0.3 flank muscles Wyoming (USA) (Lonsdale et al., 1971) Pronghorn antelope (Antilocapra americana) - -33-46 - 37.5?0.1 39.3?0.3 - abdominal WA (USA) (Sargeant et al., 1994) 27 Table 1.1. Continued Body temperature (?C) Species Mass (kg) Ambient temperature (?C) mean minimum maximum amplitude Body site Location Reference 25-30 6-31 39.2?0.4 37.6 40.2 1.6 carotid artery South Africa (Mitchell et al., 1997) 20-35 10.0?2.1 39.2?0.1 38.6?0.1 39.9?0.2 1.2?0.2 abdominal South Africa (Fuller et al., 2005) 20-35 15.8?3.7 39.4?0.2 38.9?0.2 40.0?0.2 1.2?0.3 abdominal South Africa (Fuller et al., 2005) 20-35 17.6?2.8 39.4?0.1 38.9?0.1 40.0?0.2 1.1?0.2 abdominal South Africa (Fuller et al., 2005) Springbok (Antidorcas marsupialis) 20-35 20.2?2.2 39.5?0.1 39.0?0.1 40.1?0.2 1.1?0.1 abdominal South Africa (Fuller et al., 2005) Body site = site at which body temperature was measured Location = locality where the study was conducted 28 In general, those species that were exposed to high ambient temperatures displayed a larger amplitude of nychthemeral rhythm of body temperature. Using the data represented in Table 1.1, I found a weak positive correlation between body temperature amplitude and mean ambient temperature (r2 = 0.35, P = 0.02, Fig. 1.4). Yet, despite such a relationship and a large variation in ambient conditions between the studies, only the Arabian oryx (Oryx leucoryx) displayed an increase in the amplitude of the nychthemeral rhythm of body temperature consistent with the definition of adaptive heterothermy (Ostrowski et al., 2003). Heterothermy is proposed to be an effective water saving mechanism only for large artiodactyls (Taylor, 1970a; Mitchell et al., 2002), but I found no correlation between body temperature amplitude and body mass (r2 = 0.04, P = 0.50). -10 0 10 20 30 40 0 1 2 3 4 Ambient temperature (?C) B o dy te m pe ra tu re am pl itu de ( ?? ??C ) Figure 1.4. Correlation of the amplitude (the difference between maximum and minimum) of the nychthemeral pattern of body temperature versus the mean or mid-point of the range of ambient temperature over the study period (n = 14). Since the correlation was significant (r2 = 0.35, P = 0.02), I fitted a linear regression (y = 0.06x + 0.6), where y = amplitude of body temperature rhythm and x = ambient temperature. 1.3.1.2 Thermoregulatory trade-offs Since body temperature can be measured more easily than many other variables under field conditions, it may provide a convenient measure of how changes in environmental factors impact an animal?s circadian rhythms (Cable et al., 2007). The temperature-regulating system of artiodactyls, like that of other homeotherms, functions to buffer internal body temperature against fluctuating environmental 29 temperatures by controlling heat production and heat loss. When an organism is in thermal equilibrium with its surroundings, metabolic heat together with heat absorbed from the environment must equal heat loss by radiation, convection, conduction and evaporation (Finch, 1972). Thermoregulation thus involves modifying rates of heat transfer between the core and the environment through a combination of autonomic nervous system, somatic nervous system, endocrine system or behavioural responses. Autonomic responses include polypnea (increased breathing rate, through panting), sweating (increases cutaneous evaporative water loss), vasomotor responses (alter local cutaneous blood or heat flow), ptilo- or pilomotor responses (alter the relative heat flow by controlling body hair erection, (Hammel, 1968; Bakken et al., 1985). Autonomic thermoregulation also involves modifying the level of heat generation either through somatic neural responses (shivering) or endocrine (hormonal) responses. All of the mechanisms involved in the maintenance of homeothermy may disrupt the balance of other homeostatic systems. For example shivering is energetically expensive, whereas panting and sweating threaten to disrupt the osmotic and acid-base homeostasis. For species which are dependent on evaporative cooling to dissipate environmental heat, water limitations may severely compromise such a thermoregulatory strategy and animals may be forced to trade off thermoregulation, osmoregulation and acquisition of energy. Since warming trends are predicted to be accompanied by reduced precipitation and increased frequency of droughts, both in the tropics and in the Southern Hemisphere in general, including most of Africa and Australia, species inhabiting these regions are likely to have to make such trade-offs. The high cost of autonomic thermoregulation, particularly in an environment with inadequate water supplies, may result in an animal preferentially using behavioural thermoregulation (Bustamante et al., 2002). Free-living animals rely less on autonomic thermoregulation than do animals confined to a laboratory, as they have greater access to thermoregulatory behaviour. Behavioural thermoregulation, and not autonomic adaptation, is likely 30 to act as an animal?s primary defence to adverse changes in the environmental heat load. Behavioural responses involve the co-ordinated activity of the whole animal in selecting or creating a microenvironment in which the optimal internal temperature can be achieved without, or with reduced, assistance from autonomic responses (Hammel, 1968). Terrestrial animals, because of their mobility and capacity for complex behaviours, can behaviourally select and use an array of environmental conditions, thus creating their own microenvironment (Bartholomew, 1964; Bartholomew, 1987). Desert artiodactyls, for example, are proposed to be able to maintain temperature and water balance through appropriate microclimate selection, body orientation and by restricting daily activities to time periods that reduce daily heat loads and water loss (Cain et al., 2008). Since it is this microclimate selected by the individual, not the macroclimate, that has a direct influence on an animal?s thermal status (Hodkinson, 1999; Porter et al., 2000; Porter et al., 2002), these microclimates need to be quantified and incorporated into climate change models to better predict the future survival and distribution of species. Behaviour is thus a rapid, extremely flexible and precise mechanism, which may enhance an animal's performance, and presumably its fitness, by incorporating both morphological and physiological characteristics in a variety of ways to optimise body temperature homeostasis (Bartholomew, 1964; Bartholomew, 1987; Huey et al., 2003). Since behavioural changes are generally less costly than autonomic responses, behavioural adjustments are likely to be an animal?s most suitable defence against environment change (Huey and Tewksbury, 2009). However, to date, only two models have evaluated the role of behavioural thermoregulation in buffering the impact of climate change (Mitchell et al., 2008; Kearney et al., 2009), but whether such behavioural adjustments are actually occurring in free-living species remains to be investigated. Nevertheless, implementing behavioural thermoregulation may involve its own costs. For example, a habitat selected purely based on its thermal properties may have an increased risk of predation, parasites, competition, as well as a decreased 31 resource availability, including energy, mates, food and water (Bartholomew, 1964; Huey, 1991; Bakken, 1992). Thus, in an extreme thermal environment, such as that predicted by climate change, heat loss mechanisms may be severely compromised, and obligatory thermoregulatory behaviour may compromise these inferential functions. The moose (Alces alces) provides an example of the potential negative costs associated with behavioural thermoregulation. In the last 20 years, the moose population in Minnesota has halved and the population in the Isle Royale National Park has declined by 75%. Moose are particularly sensitive to heat and seek shelter when ambient temperatures exceed 14?C (Dussault et al., 2004). Over the last 40 years average summer temperatures have increased by 2?C and moose have had to forfeit valuable foraging time as they are forced to become lethargic and immerse themselves in water in an attempt to stay cool. Such behaviour leads to malnutrition and decreases their fat reserves, which are essential for winter survival. This malnutrition is likely to increase their risk of succumbing to parasites, disease and predation by wolves, all factors which are believed to have contributed to the recent decline in the moose population (Thompson et al., 1998; Murray et al., 2006). With further increases in summer temperatures predicted for the future, it seems likely that the moose will be extirpated from its historic southern range within the next 50 years (Murray et al., 2006). 1.3.2 Future applications Though we know so little about it, it will be on their physiological phenotypic plasticity that the future of long-lived mammals, threatened by global warming, will depend. Many authors explicitly extend that position to behavioural plasticity, distinct from physiological plasticity, but in this context I consider behaviour to be one element, and a crucial one, of the repertoire of physiological responses available to animals to adapt to climate change (Bartholomew, 1964; Meyers and Bull, 2002; Huey et al., 2003; Gilmour et al., 2005). Physiological mechanisms are responsible for the capacity of animals to adjust to new environments (Feder and Block, 1991; Carey, 2005) and are fundamental to predicting the consequences of climate change (Helmuth et al., 2005). Physiologists thus have 32 an important role to play in uncovering the mechanisms employed by animals to counter extinction and cope with climate change. Future climate change research will require measurement of physiological characteristics of many identified individual animals for long periods, probably decades (Berteaux et al., 2004; Visser, 2008). Since the responses to climate change are likely to be multifaceted responses to complex interrelated stresses, the approach will have to be that of field physiology (Costa and Sinervo, 2004), namely investigation of the mechanisms that an animal uses to maintain homeostasis in its everyday life. The more-traditional technique of characterising habitats in the field, but studying physiological responses in the laboratory (Huey, 1991), identifies what an animal can achieve physiologically, but not what it actually will do, free-living in its natural habitat, as has been demonstrated, for example, by field studies of how arid-zone antelope use selective brain cooling (Mitchell et al., 2002) or how elephant shrews use torpor (Mzilikazi and Lovegrove, 2004). The studies required fall within the sub-disciplines of conservation physiology (Carey, 2005; Wikelski and Cooke, 2006) and evolutionary physiology (Feder et al., 2000). The growth of these sub-disciplines has resulted not just from the clear need for such an approach, but from the growing availability of suitable technology, such as the use of stable isotopes for field measurement of metabolic rate and water turnover, and osmotic minipumps to deliver substances to free- living animals (Goldstein and Pinshow, 2006). The primary new technology, however, has been biotelemetry or biologging (Andrews, 1998; Cooke et al., 2004; Block, 2005; Wikelski and Cooke, 2006). Physiological variables such as dive behaviour and prey ingestion of marine mammals, as well as body temperature, activity and energetic expenditure of terrestrial mammals, now can be measured routinely in free-living animals. The studies will require long-term monitoring of physiological variables of animals going about their daily business in their natural habitat. Moreover, because the physiological mechanisms that animals employ in the field are confounded by the presence of human observers (Mitchell et al., 33 2002) it is not just autonomic responses that will have to be measured by biotelemetry, but behavioural responses too. It is now possible, for example, to monitor microclimate selection by biotelemetry (Appendix 1, Hetem et al., 2007). Such long-term monitoring studies also would benefit from availability of suitable equipment for sampling blood in the absence of human observers, as described 20 years ago (Hattingh et al., 1988), but still seldom implemented. 1.4 Thesis aims The general aim of the research described in my thesis was to investigate the phenotypic plasticity which may enable long-lived species to cope with climate change. I focused on the physiological plasticity of artiodactyls, which, because of their evolutionary past and the development of the carotid rete, may be particularly good at adapting to hot and arid conditions. It will become increasingly important to understand the physiological mechanisms employed by artiodactyls to adapt to arid regions so that we may better monitor and predict species responses to the increased aridity anticipated for large regions of Africa. Secondly, my thesis aims to investigate whether remote sensing technology is sensitive enough to detect physiological plasticity, such as changes in body temperature and activity. If such technology proves to be successful in detecting physiological changes in the short term it will provide confidence and motivation to extend such studies from their current durations of about a year (for example, Mzilikazi and Lovegrove, 2004; Fuller et al., 2005; Gremillet et al., 2005) to that of decades. Instead of the typical modelling approach, I used an empirical or experimental approach to investigate the physiological mechanisms employed by artiodactyls under circumstances that exist today but at the extremes, and in the direction in which climate change is moving. I chose to investigate whether morphology variations, such as difference in pelt colouration and body size, have physiological consequences. Morphological changes are believed to have constituted the first response of mammals to climate change events of the past (Barnosky et al., 2003) 34 and, because of the ease of recording of morphological features, the vast majority (>80%) of studies of phenotypic selection have focused on morphological plasticity (Kingsolver et al., 2001). Yet, such morphological plasticity can only be adaptive if it results in a change in function, a condition seldom investigated. In addition, I chose to address issues such as how artiodactyls at the current upper end of hot environments cope and the physiological consequences of desertification. Habitat transformation, in the form of desertification, will be a consequence of changing climatic conditions and there is debate about the relative detrimental effects of climate itself and its consequences for habitat. Currently, 46% of the sub-humid and semi-arid regions of southern Africa are vulnerable to desertification and, because of an increased water stress predicted for the future, the proportion of arid and semi-arid regions in Africa is predicted to increase by 5-8% by 2080 (Boko et al., 2007). I therefore investigated whether similar physiological responses were exhibited by artiodactyls inhabiting one of the most stressful of current hot and hyper-arid environments, the deserts of Saudi Arabia, and those artiodactyls, which are increasingly exposed to desertified landscapes in South Africa. Finally, the physiology studies required to better understand the mechanisms of phenotypic plasticity underlying responses to climate change cannot be confined to the function of healthy animals. Infectious pathogens have plagued animals and plants throughout their evolutionary history and have periodically caused population declines and extinctions (Wilson, 1994). According to Lochmiller and Deerenberg (2000), ?immunological competence could very well be the most important determinant of life-time reproductive success and fitness for many species?. Climate change will not only bring emergent pathogens with the relocation of arthropod vectors (Rogers and Randolph, 2000; Olwoch et al., 2003; Cumming and van Vuuren, 2006), but may also disrupt historically stable relationships between co-existing pathogens and their existing hosts, thus increasing mortality risks (Munson et al., 2008). So conservation physiology, in the context of climate change, will need to include the study of innate immunity 35 and sickness behaviour, the physiological processes which animal hosts use to contend with novel pathogens (Wingfield, 2003; Carey, 2005). Yet, very little is known about the immune responses and sickness behaviours of free-living artiodactyls. Chapter 2. Previous climate change events have resulted in changes in morphology and many studies have documented morphological changes with recent climate change. Yet, there is uncertainty about whether changes in morphology are phenotypic or genotypic. If morphological changes are genotypic, the unprecedented rate of climate change predicted to occur over the next 50 years may negate any mitigation effects of morphological adaptation, particularly for long-lived species. On the other hand, phenotypic plasticity in a morphological trait would hypothetically provide potential for long-lived species to adapt to changing climatic conditions. However, whether phenotypic or genotypic, changes in morphology are moot unless they result in a change in function. Few studies to date have investigated the physiological consequences of morphological variation. Natural variation in hair colour of springbok (Antidorcas marsupialis) provides an ideal opportunity to investigate the thermoregulatory advantage of a morphological trait. Anecdotal reports, mainly from game ranchers in the Eastern Cape, suggest that the black springbok are better able to survive cold winter spells than the other springbok colour morphs. If these anecdotal accounts are true, pelt colour may prove to be a functional variable capable of compensating for potential detrimental effects of climatic change. I therefore set out to investigate the effect of pelt colouration on the thermoregulation of springbok free-living in the arid Karoo. If colour variation does indeed have thermoregulatory consequences, certain colour morphs may be selected for in the future, and colour variation within a population may provide an important pre-adaptation to climate change. Chapter 3. The deserts of Saudi Arabia represent one of the most stressful of the current hot and hyper-arid environments and, as such, provide an ideal environment in which to study the physiological mechanisms of species that have 36 adapted to that extreme, an extreme likely to become more prevalent under future climate change. The Arabian oryx inhabits this hot desert environment and is reputedly the most arid adapted of all antelope. It survives most of the year, including the hottest summer months, without access to drinking water. As such, the Arabian oryx provides an ideal species in which to investigate water conservation strategies. By studying the physiological mechanisms by which Arabian oryx compensate for limited water and extreme heat, I aim to gain a better understanding of the kind of physiological mechanism that other artiodactyls may employ to adapt to changing climate conditions. Since many arid regions of Africa are predicted to get both hotter and drier with climate change, understanding the physiological mechanism exhibited by this desert antelope would assist future studies in monitoring physiological adaptations to desertification. I therefore measured body temperature, activity and microclimate selection of free-living Arabian oryx occupying a hyper-arid desert environment, for a one-year period, to ascertain the importance of, and the potential interactions between, reputed water-saving mechanisms. Chapter 4. One of the mechanisms employed by the Arabian oryx to cope with their hyper-arid environment is heterothermy. Heterothermy is proposed to be a more effective water saving mechanism for large mammals than for small mammals, since larger mammals have greater thermal inertia and also because, within a given time, large mammals would gain relatively less heat from the hot environment (McNab, 1983; Phillips and Heath, 1995) and generate less metabolic heat per gram of tissue than would small mammals (Taylor, 1970a; Mitchell et al., 2002). Climate change is predicted to favour smaller individuals and numerous species of endotherms have exhibited decreased body mass in response to recent increases in ambient temperature. Yet, it is unknown whether different sized artiodactyls employ different thermoregulatory mechanisms to adapt to hot and dry environments. I therefore set out to compare thermoregulatory mechanisms of a large and small desert-adapted antelope, namely the small Arabian sand gazelle (? 15 kg) and the larger Arabian oryx (? 70 kg), inhabiting the same hot hyper-arid environment. Although the smaller 37 gazelles may have the advantage of smaller resource requirements and greater access to refuge sites than do the larger oryx (Ostrowski and Williams, 2006), they are likely to be disadvantaged by a high mass-specific metabolic rate, high water turnover and less capacity to store heat (Bartholomew, 1964; Taylor, 1970a). I therefore compared the body temperature and refuge selection of the sand gazelle and oryx living free in the same desert environment, at the same time. Chapter 5. One of the consequences of climate change will be habitat transformation and there is debate about the relative detrimental effects of the climate itself and its consequences for habitat. I therefore set out to investigate the effects of habitat transformation for an arid-adapted artiodactyl. I chose Angora goats as my study species, because, unlike antelope, they are docile and amenable to experimental manipulation. Goats have a wide geographical distribution, surviving in a range of diverse climates and topographies. They have been able to colonise and thrive in some of the most extreme environments in the world (Shkolnik and Choshniak, 1987). Angora goats are particularly well suited to low rainfall areas (Norton and Deery, 1985) and are extensively farmed in the semi- arid regions of the Eastern Cape, helping to put South Africa on the map as being one of the biggest mohair producers worldwide. Since the mid 1800s Angora goat farming has put pressure on the natural resources, resulting in severe habitat transformation in some areas of the Eastern Cape. Such transformation of the natural flora has been classified as desertification, or dryland degradation, and is the result of overgrazing due to human mismanagement of agricultural and pastoral resources (Dean et al., 1995; Kerley et al., 1995). This dryland degradation as a result of overgrazing is likely to test Angora goats? ability to withstand both heat and water stress. Since the desertification within the semi-arid regions of the Eastern Cape succulent thicket habitat mimics at least some of the future consequences of climate change described earlier, it provides an appropriate location at which to analyse potential physiological consequences of habitat transformation. I therefore investigated the effect of habitat transformation, in the form of desertification, on the body temperature of Angora goats inhabiting transformed and intact sites in the Eastern Cape. The transformed and intact sites I 38 selected were positioned adjacent to each other, thus reducing the potential confounding variables associated with geographic location and climate. Chapter 6. Since climate change is predicted to bring emergent pathogens, particularly with the relocation of arthropod vectors (Rogers and Randolph, 2000; Olwoch et al., 2003; Cumming and van Vuuren, 2006), studies of phenotypic plasticity cannot be confined to the function of healthy animals. We need to understand the physiological mechanisms employed by free-ranging antelope to contend with novel pathogens. A herd of free-living antelope (kudu, Tragelaphus strepsiceros), which I had instrumented with temperature and activity data loggers for other purposes, acquired a spontaneous infection. I was able to extract quantitative data reflecting the kudu?s autonomic and behavioural responses to the infection and, for the first time, show how a sick free-living artiodactyl responds to its thermal environment. 39 ___________________________________________________________ CHAPTER 2 ___________________________________________________________ 2 Morphological adaptation to climate change: the effect of pelt colour on body temperature and thermoregulatory behaviour of springbok (Antidorcas marsupialis) Data and ideas presented in this chapter have been published in the paper: Hetem R.S., de Witt B.A., Fick L.G., Fuller A., Kerley G.I.H., Meyer L.C.R., Mitchell D. and Maloney S.K. (2009) Body temperature, thermoregulatory behaviour and pelt characteristics of three colour morphs of springbok (Antidorcas marsupialis). Comparative Biochemistry and Physiology, A: Molecular and Integrative Physiology 152: 379-388. 40 2.1 Abstract Using intra-abdominal miniature data loggers, with the help on my colleagues, I measured core body temperature in female springbok (Antidorcas marsupialis) of three colour morphs (black, normal and white), free-living in the Karoo, South Africa, for one year. During winter, white springbok displayed lower daily minimum body temperatures (37.4 ? 0.5?C), than did both black (38.1 ? 0.3?C) and normal (38.0 ? 0.6?C) springbok. During spring, black springbok displayed higher daily maximum body temperatures (40.7 ? 0.1?C) than both white (40.2 ? 0.2?C) and normal (40.2 ? 0.2?C) springbok. These high maximum body temperatures were associated with larger daily amplitudes of nychthemeral rhythm of body temperature (2.0 ? 0.2?C), than that of white (1.6 ? 0.1?C) and normal (1.7 ? 0.2?C) springbok. Biophysical properties of sample springbok pelts were consistent with these patterns, as the black springbok pelt showed lower reflectance in the visible spectral range, and higher heat load from simulated solar radiation, than did the pelts of the other two springbok. Black springbok had lower diurnal activity in winter, consistent with them having to forage less because their metabolic cost of homeothermy was lower, but were disadvantaged in hot periods. White springbok, by contrast, were more protected from solar heat load, but potentially less able to meet the energy cost of homeothermy in winter. Thus energy considerations may underlie the rarity of the black and white springbok colour morphs. 2.2 Introduction Both lay opinion and early physiological analysis associated the colour and structure of an animal?s pelt with adaptation to the thermal environment (Finch, 1972). Thus, animals may adapt to local conditions and changing climates by displaying variation in colour (Millien et al., 2006). Yet, the thermoregulatory consequences and adaptive significance of colour remains tentative since the thermal function for pelt colour can be confounded by other pressures, such as the requirement for concealment or communication (Cowles, 1967; Cloudsley- 41 Thompson, 1979; Stoner et al., 2003; Caro, 2005; Caro, 2009). The natural variation in hair colour of springbok (Antidorcas marsupialis) provides an attractive opportunity to investigate the thermoregulatory consequences of pelt colour. There are three springbok colour morphs, commonly referred to as normal, white, and black springbok (Fig. 2.1), which show no taxonomic differences. The length, texture and thickness of hairs making up the pelt are the same (Kruger et al., 1979) and so it might be expected that the morphs are similarly insulated, while varying in colour. The familiar normal springbok of southern Africa, with a pale ventral surface, white face and rump, and a buff back, is an arid-adapted species (Estes, 1991; Skinner and Louw, 1996, see Fig. 2.1). Its colouration is proposed to have thermoregulatory importance, dependent on the assumption that the white colouration of the face and rump reflects solar radiation, thus reducing solar heat load when an animal orientates parallel to incident solar radiation, and that the darker sides absorb solar radiation, so that an animal can enhance solar heat gain by orientating perpendicular to the sun (Hofmeyr and Louw, 1987). The far less common black springbok and white springbok have the same body conformation as the normal springbok, and apparently differ only in hair colour. Black springbok are predominantly a dark dull chocolate brown, whilst white springbok are predominantly a dusty white colour (Kruger, 1976; Kruger et al., 1979). Genetic studies on black and white springbok have failed to reveal the pattern of inheritance of the unusual colour morphs, though some believe that the normal colouration results from a double recessive combination (Kruger, 1976). White springbok are known to have occurred naturally, albeit at very low frequencies (Roche, 2005; Skead, 2007). The black colour morph was not recorded historically, and appears to have arisen through inbreeding of an enclosed population on a farm in the Murraysburg district of South Africa in the 1950s (Skead, 2007). Game ranchers have increased the relative abundance of black and white springbok through selective breeding (Skinner and Louw, 1996). 42 Figure 2.1. The pelt colour variations of the black, normal and white springbok (adapted from Skinner and Louw, 1996). The adaptive significance, if any, of the different colour morphs is unknown, as the black and white animals have been bred for aesthetic purposes. However, anecdotal reports, mainly from game ranchers in the Eastern Cape, South Africa, suggest that the black springbok are better able to survive cold winter spells than are the other springbok colour morphs. If these anecdotal accounts are true, it may be that dark-coloured animals conserve metabolic energy better in cold climates by absorbing more solar radiation than their lighter-coloured counterparts, and so Normal springbok Black springbok White springbok 43 offset the energetic costs of homeothermy. Although dark colouration may be an advantage during cold winters, a high absorption of radiant heat may be a disadvantage during the typically hot and cloudless Karoo summers. The purpose of this study was to investigate whether the colour variations among the three springbok colour morphs have thermoregulatory consequences. I, together with the help of my colleagues, recorded simultaneously, for the first time, body temperature and behaviour of all three springbok colour morphs, in a small paddock near Port Elizabeth, in the Eastern Cape, for a three-month winter period. With the help of my colleagues, I subsequently expanded the study to record body temperature from animals of the three springbok colour morphs, for a year, with the animals living free in their natural environment in the Succulent Karoo, South Africa. Additionally, to determine if the pelts of the three colour morphs have different heat transfer characteristics, I measured conductance, reflectance, and radiant heat load of sample pelts of the three springbok colour morphs. As far as I am aware, this study is the first to compare body temperature, behaviour and pelt characteristics of animals with different colour morphs, but which apparently are otherwise identical. 2.3 Materials and methods 2.3.1 Body temperature and thermoregulatory behaviour 2.3.1.1 Study area The short-term winter study, during which my colleagues and I recorded behaviour and body temperature, took place from June to August 2004 (southern hemisphere winter). Representatives of the three springbok colour morphs were housed together in a 3 ha game camp (34?00? S, 25?39? E, at an altitude of 60 m above sea level) at the Nelson Mandela Metropolitan University, Port Elizabeth, South Africa. The camp?s vegetation is classified as Algoa Dune Strandveld (Mucina and Rutherford, 2006). Animals had access to water ad libitum and 44 grazed on natural vegetation in the camp, which was supplemented with lucerne daily, to which all animals had access. In the longer-term study, my colleagues and I recorded body temperatures of representatives of the three springbok colour morphs every 15 min over a one- year period (October 2004 to August 2005), at Hopedale farm, a few kilometres west of Steytlerville (33?20? S, 24?20? E, at an altitude of 480 m above sea level). The camp covered an area of 200 ha and the vegetation is classified as Steytlerville Karoo, part of the Succulent Karoo biome (Mucina and Rutherford, 2006). All animals had access to water ad libitum and grazed on natural vegetation. 2.3.1.2 Animals Seventeen springbok (Antidorcas marsupialis), 15 of which were captured at the beginning of June 2004 at Cradock, South Africa, and supplemented by two normal springbok from the Nelson Mandela Metropolitan University?s nature reserve, were housed in pens at the Nelson Mandela Metropolitan University (~ 200 km south of Cradock) before surgery and release onto the study areas. The study groups for the short-term study consisted of four female white springbok (body mass 23 ? 4 kg), five female black springbok (25 ? 3 kg) and five female normal springbok (21 ? 4 kg). For the one-year study, the study groups consisted of four female white springbok (body mass 22 ? 2 kg), four female black springbok (27 ? 3 kg) and seven female normal springbok (23 ? 3 kg). A single male white springbok (26 kg) was housed with the females during both of the studies, but no data were collected from him. All experimental procedures were approved by the Animal Ethics Screening Committee of the University of the Witwatersrand (clearance number 2004/56/04). 45 2.3.1.3 Surgery The springbok were anaesthetised in the holding pens by intramuscular (I.M.) injection of 3.5 mg.kg-1 ketamine hydrochloride (Anaket-V, Bayer, Isando, South Africa) and 0.05 mg.kg-1 medetomidine hydrochloride (Domitor, Novartis, Kempton Park, South Africa), and once recumbent were carried to a temporary operating theatre set up in an adjacent laboratory. Anaesthesia was maintained with 1-4% halothane (Fluothane, Astra Zeneca, Johannesburg, South Africa) administered in 100% oxygen via a facemask. After ~ 10 min of halothane administration, the action of medetomidine was reversed with 0.25 mg.kg-1 (I.M.) atipamezole hydrochloride (Antisedan, Novartis, Kempton Park, South Africa). Respiratory rate, oxygen saturation, heart rate (pulse oximeters, Nonin 9847V, Nonin Medical, North Plymouth, USA) and rectal temperature (thermocouple thermometer, BAT-12, Physitemp, Clifton, USA) were monitored throughout the surgical procedure, which lasted ~ 15 min. A 200 x 200 mm midline area on the ventral abdominal surface of each springbok was shaved and sterilized with chlorhexidine gluconate (Hibitane, Astra Zeneca, Johannesburg, South Africa) in alcohol. Within this area, a 70 mm cranial-caudal incision was made through the skin and linea alba, and a miniature temperature- sensitive data logger (see description below) was placed into the abdominal cavity, where it floated freely. The skin and muscle layers were sutured closed. The wound was sprayed with a topical antiseptic spray (Necrospray, Centaur Labs, Johannesburg, South Africa) and coated with a topical tick repellent (tick grease, cypermethrin 0.025% m/m, Bayer Animal Health Pty, Isando, South Africa). The springbok received a long-acting antibiotic (900 mg I.M. penicillin, Peni LA Phenix, Virbac Animal Health, Centurion, South Africa), analgesic and anti-inflammatory medication (560 mg I.M., ramiphenazone and 2 mg dexamethazone, Dexa-Tomanol, Centaur Labs, Johannesburg, South Africa), and a long-acting parasiticide (5 mg subcutaneously, doramectin, Dectomax, Pfizer Laboratories, Sandton, South Africa). Animals were marked with different coloured plastic ear tags and a small ring of coloured reflective tape (conspicuity sheeting, 3M Scotchlite Diamond Grade, Rivonia, South Africa) around their 46 horns. These markings enabled us to identify individual animals, even at night. After recovery from surgery, the springbok were released, as a single herd, into the game camp on the Nelson Mandela Metropolitan University campus, where they roamed freely throughout the three-month winter study period, here forth termed ?Port Elizabeth winter?. After the initial three-month study period, the springbok were recaptured, tranquillised with haloperidol (10-15 mg I.M., Kyron Laboratories, South Africa) and perphenazine (50-100 mg I.M., Kyron Laboratories, South Africa), and held in the pens at the Nelson Mandela Metropolitan University. After two weeks, animals again were anaesthetised, the data loggers removed and replaced with new data loggers under the same procedures used for the original implantation. After a two-week recovery period in the pens, the animals were transported ~ 180 km to Hopedale farm and released together. They remained as a single free-living herd. After the one-year study period, the springbok were recaptured, tranquillised and transported back to the original holding pens at the Nelson Mandela Metropolitan University. Animals once again were anaesthetised and the data loggers were removed, under the same surgical procedures. In both studies the data loggers were all physically intact, the animals? wounds had healed, and there were no signs of infection from the implant surgery. Most of the loggers were found intra- abdominally in the pelvic canal and were not encapsulated in adherent tissue. After surgery, the springbok were returned to the field to rejoin conspecifics. 2.3.1.4 Body temperature measurements My colleagues and I implanted miniature temperature-sensitive data loggers (StowAway XTI, Onset Computer Corporation, Pocasset, MA, USA) with dimensions of ~ 50 x 45 x 20 mm and a mass of ~ 40 g when coated in inert wax (Sasol wax EXP986, Sasol, Johannesburg, South Africa). The data loggers had a storage capacity of 32 kb and measured temperatures within the range of + 34 to 47 + 46?C, at a resolution of 0.04?C. All the data loggers were calibrated individually, in an insulated water bath, against a high-accuracy thermometer (Quat 100, Heraeus, Hanau, Germany). Calibrated accuracy was better than 0.1?C. The data loggers were set to record temperature at 5-min intervals over the short-term Port Elizabeth winter study and at 15-min intervals for the one-year study. The 15-min interval should have provided about 338 days of readings, but five of the data loggers stopped recording after about 270 days, because of premature battery failure. Such premature battery failure reduced my sample size to ten individuals during the Karoo ?winter? period, comprising of two female black springbok, four female white springbok and four female normal springbok. 2.3.1.5 Behavioural observations The springbok were observed for one week of each month of the three-month Port Elizabeth winter study period. Brenda de Witt, as part of a student project, made the actual field observations, using binoculars, from a hide on the perimeter of the game camp. Scan samples (Altmann, 1974) of behaviour were made at 5-min intervals, throughout ~ 3?6 h observation periods, which, over the week period, covered all 24 h. A one-zero sampling method (Altmann, 1974) was used to record the behaviour of each individual of the black, white and normal springbok at each scan. Ms de Witt recorded whether animals sought shelter from wind (downwind or in close proximity to vegetation) or stood out in the open, their orientation to solar radiation (long axis of body parallel, perpendicular or oblique to the sun), the proximity of the nearest conspecific (close (< 1 m) or far (> 1 m)), whether animals were in sunshine or shade, standing or lying, foraging, ruminating or neither, and the degree of their locomotor activity (inactive or active). Behaviour was not recorded during the one-year study, because it was not possible to observe the animals properly from the perimeter of the 200 ha camp, and I did not want to enter the camp as I would have disturbed the normal behavioural patterns of the free-living springbok. 48 2.3.1.6 Climate measurements Black globe temperature and air temperature were measured, 1 m above ground, in an open area that was not shaded or protected by vegetation, at each of the study sites. The air temperature probe was housed in a ventilated vane which was painted white to reflect solar radiation. The measurements were made at 15-min intervals by a dual channel temperature data logger (Hobo, Onset Computer Corporation, Pocasset, MA, USA). Due to technical malfunctions, I did not obtain a complete data set, and so I supplemented my weather data with data obtained from a nearby meteorological station (South African Weather Services, Willowmore). 2.3.1.7 Body temperature analysis For analysis of seasonal patterns of body temperature, data from the one-year study were averaged over four periods, based on prevailing climatic conditions, namely ?spring? (October to November), ?summer? (December to February), ?autumn? (March to May) and ?winter? (June to August). I also selected the 10 hottest days in summer, determined by maximum air temperatures, which averaged 37.8 ? 1.3?C, and the 10 coldest days in winter, determined by the minimum air temperature, which averaged -1.9 ? 1.0?C (Table 2.1). The mean ? SD of the mean, minimum, maximum and amplitude of the nychthemeral rhythm of body temperature of the groups of black, white and normal colour morphs were calculated for each ?season?, the 10 hottest and 10 coldest days, as well as for the short-term Port Elizabeth winter study. I calculated the rate of rise in body temperature over consecutive 4-h time intervals and analysed the maximum rate of body temperature rise. I also calculated the mean time of day at which both the minimum and maximum body temperature occurred. I performed a repeated measures two-way analysis of covariance (ANCOVA), with body mass as a covariant, to test for interaction between the body temperatures of the springbok colour morphs and ?season?, as well as between the body temperatures of the springbok colour morphs and hot and cold days. Newman-Keuls multiple comparisons tests were used to identify sources of significant differences in 49 ANCOVAs. I used a one-way ANCOVA to test for differences in the body temperature profile of the three colour morphs during the short-term Port Elizabeth winter study. I used Statistica (kernel release 5.5 for Windows, StatSoft, Inc. (1999), Tulsa Oklahoma, USA) for statistical analyses. To further assess the influence of the ambient temperature on body temperature patterns, I performed a Pearson?s linear correlation between the daily nychthemeral amplitude of body temperature, averaged for all animals in each colour morph group, and the air temperature range on the same day, over each ?season?. I also correlated mean daily maximum body temperature with daily maximum air temperature, and mean daily minimum body temperature with daily minimum air temperature, for each colour morph group over each ?season?. I fitted linear regressions, where appropriate, and compared the slopes of the regression lines among the three colour morphs and across ?seasons? using a repeated measures two-way ANCOVA. A chi-squared test for independence was used to test for differences in the patterns of behaviours exhibited by the colour morphs during the short-term Port Elizabeth study. Data are expressed as mean ? SD, and ? < 0.05 was considered to be statistically significant. 2.3.2 Pelt heat transfer characteristics With the help of my colleagues, I measured pelt heat transfer characteristics using the technique previously described for emu pelts by Maloney and Dawson (1995), as summarized below. 2.3.2.1 Pelt samples I purchased winter pelts of one normal, one white and one black springbok from a taxidermist in the Eastern Cape. The pelts were tanned to standards which allowed transport across international borders. A section, excised from the mid-dorsal region of each pelt and representative of each colour morphs predominant pelt colour, was used to determine spectral reflectance, thermal conductance and heat load from radiation. 50 2.3.2.2 Spectral reflectance Spectral reflectance of the pelt samples was measured with an ultra-violet/visible light spectrophotometer (Varian DMS 80 UV, Belmont, Western Australia, Australia). Reflectance was measured at three sites, randomly chosen on the dorsal surface of each pelt sample, with three replicates at each site. Measurements were made between 350 nm and 950 nm, at 1 nm intervals, with reference to a barium sulphate disc. This range of wavelengths replicates the range of greatest intensity of solar energy (Gates, 1966) and the range of output of the radiation source used in the wind tunnel experiments (described below, Maloney and Dawson, 1995). 2.3.2.3 Thermal conductance To measure thermal conductance, I mounted the pelt sample horizontally, hair upwards, on the upper surface of a metal plate (250 mm diameter), maintained at ~ 38?C via a temperature-controlled circulating water bath (Fig. 2.2). A rectangular glass wind tunnel positioned over the pelt allowed laminar-flow wind to be directed parallel to the pelt surface, at different speeds. Three heat flux transducers (20 mm x 30 mm, model HA13-18-10P, Thermonetics Corporation, U.S.A.) were embedded in the upper surface of the plate. The voltage outputs of the transducers were logged via an analog/digital (A/D) converter (Datataker, Data Electronics, Australia P/L, model 100F). The heat flux transducers were calibrated using two thicknesses of polystyrene (Boral Industries, Sydney, Australia) of known thermal conductance. Copper/constantan (Type-T) thermocouples, 0.7 mm in diameter, were used to measure plate temperature (Tp), air temperature (Ta), and skin surface temperatures (Ts). Skin surface temperatures were measured by feeding three thermocouples to the hair/skin interface, through oblique holes from beneath the skin. Air temperature was measured near the inlet of the wind tunnel so that it was not affected by the boundary layer. Thermocouple outputs were referenced against an isothermal block (Datataker, Data Electronics, Australia P/L) and the 51 temperatures logged via the A/D converter, as above. Thermocouples were calibrated, against a mercury thermometer that was certified by the Australian National Association of Testing Authorities in an insulated water bath, to 0.1?C accuracy. Thermal conductance was measured in a temperature-controlled room (air temperature 20 ? 2?C). Once temperatures had stabilized, heat flow, skin and air temperatures were measured for at least half an hour. Mean values then were used to calculate pelt conductance (C, W.m-2.?C-1) as ( )as TTQC ?=  , where Q is the heat flow (W.m-2) through the pelt as measured by the transducers, Ts is skin surface temperature and Ta is air temperature. Conductance of each sample was determined at four wind speeds (~ 0, 1.5, 3 and 4.5 m.s-1), as measured 20 mm above the pelt surface with a thermo-anemometer (Schiltknecht 39400, Technical & Scientific Equipment Co., Melbourne, Australia). Air flowed in a cranial to caudal direction across the pelt sample. 2.3.2.4 Radiant heat load To determine the heat load on the animal resulting from radiant heat absorption, I repeated measurements of heat flow and temperature profiles across the pelt with the pelt surface subjected to a simulated solar radiant heat load (Fig. 2.2). Short- wave radiation was provided by a spotlight (ARRI daylight 575 W), containing a 575 W metal halide lamp (ILC Technology, DM1575), which radiated 590 W.m-2 of ultra-violet/visible (300 - 1100 nm) radiation. The radiant flux was verified at the pelt surface with a radiometer (Model 8-48, Eppley Laboratory, Rhode Island, USA). The relative spectral distribution of radiation supplied by the lamp had been shown previously to be similar to that of the solar spectrum (Walsberg, 1982; Maloney and Dawson, 1995; Dawson and Maloney, 2004). Radiation was delivered through a 100 mm diameter hole in aluminium foil covering a glass pane, placed between the lamp and the wind tunnel (Fig. 2.2), so that the spotlight did not heat the wind tunnel. 52 The proportional heat load from radiation was calculated as [ ] radiationincidentHFHF radiationwithradiationwithout ? , where heat flow (HF) through the sample was measured by the transducers, with adjustment for the difference in air temperature and plate temperature depending on whether radiation was present or absent (Walsberg, 1982; Maloney and Dawson, 1995; Dawson and Maloney, 2004). Figure 2.2. Experimental set-up for measuring thermal conductance and radiant heat load in the springbok pelts. HFT?s = Heat Flux Transducers. 2.3.2.5 Data analysis of pelt characteristics I calculated mean weighted solar reflectance by averaging measured reflectance at 50 nm intervals multiplied by the relative spectral power of sunlight at each wavelength. I could not measure reflectance above 950 nm; I assumed it to remain constant at the 950 nm value. I correlated thermal conductance with wind speed; I then fitted a linear regression and tested for differences in the slopes and intercepts using ANCOVA. I compared heat load from radiation as a function of wind speed, among the pelts, by fitting and comparing a one-phase exponential decay curve. Tukey-Kramer multiple comparisons tests were used to identify sources of significant differences in ANCOVAs. Data are expressed as mean ? SD and ? < 0.05 was considered to be statistically significant. I used GraphPad Prism (version 4.00 for Windows, GraphPad Software, San Diego California USA) for statistical analyses. spotlight Fan Glass covered with aluminum foil Anemometer Connection to circulating water bath Temperature-controlled plate with embedded HFT?s Insulating foam Pelt window in aluminium foil 53 I calculated potential energetic consequences of pelt colouration using an approach similar to that described by (Maloney et al., 2005a). I used solar radiation intensity incident on a horizontal surface as measured at a nearby weather station (Blaauwkrantz Farm, 33?32?S 25?23?E, Table 2.1). I multiplied hourly solar radiation over each season by the fraction of the heat load which reaches the springbok (Fig. 2.8), at the average wind speed prevailing at the field site during that season, for each colour morph. The fraction of the total body surface area presented to incident solar radiation varies with body orientation and solar elevation. I calculated solar elevation for the latitude and longitude of my field site, using software available at http://www.susdesign.com/sunangle, for each hour of a representative day within each season. The surface area presented at different solar elevations was estimated based on data (Clapperton et al., 1965) obtained for a 35 kg shorn sheep (total surface area 1 m2) orientated either parallel or perpendicular to the solar beam. Estimating surface area (m2) of the skin from 3 2 09.0 WA ?= , where W is mass in kg, I calculated a proportional surface area presented at different solar elevations for a 25 kg springbok model. The best fit equation was an exponential equation for parallel orientation (ln y = -2.02 ? 0.008x, where y is the proportional surface area of the animal exposed to solar radiation and x is the solar elevation) and a power equation for perpendicular orientation (y = 0.27 ? 1.97x10-7x). For each hour of sunlight during each season, the surface area presented to incident radiation was calculated from these equations for the appropriate solar angle. The magnitude of the total absorbed solar radiation (MJ per day) was calculated by multiplying surface area presented to incident radiation by the solar radiation incident on each of the colour morphs, for each hour of sunlight within each season. Metabolic savings were calculated by expressing total absorbed solar radiation as a proportion of field metabolic rates of springbok (Nagy and Knight, 1994), within each season, for each colour morph, assuming that field metabolic rate was the same for all colour morphs. 54 2.4 Results 2.4.1 Body temperature and thermoregulatory behaviour 2.4.1.1 Climate Table 2.1 shows the mean ? SD of environmental variables. Air temperature reached a maximum of 40?C in February and dropped to as low as -4?C in June, in the Karoo, giving a range of 44?C. The minimum air temperatures usually occurred before dawn (03:00 - 05:00) and the maximum air temperatures usually occurred after noon (12:00 - 14:00). Notably, the Karoo ?winter? was colder and drier than the Port Elizabeth winter. 2.4.1.2 Body temperature Figure 2.3 shows mean, minimum and maximum body temperatures, and the amplitude of the nychthemeral body temperature rhythm (maximum - minimum), for the different springbok colour morphs, over the four time periods while they were living free in the Karoo. A repeated measures two-way ANCOVA of data from the one-year study revealed that mean (F3,21 = 23.8, P < 0.0001), daily minimum (F3,21 = 14.2, P < 0.0001) and daily maximum (F3,21 = 42.6, P < 0.0001) body temperatures decreased as ambient temperature became progressively colder from ?summer? to ?winter?. 55 Table 2.1. Environmental conditions (mean ? SD) prevailing during the four seasonal periods during which the springbok were free- living in the Steytlerville Karoo region, during the three-month winter study at Port Elizabeth (PE win) and during the 10 hottest days in summer and the 10 coldest days in winter in the Steytlerville Karoo region. Oct-Nov ?spring? Dec-Feb ?summer? Mar-May ?autumn? Jun-Aug ?winter? Jun-Aug ?PE win? Hot days Cold days Mean globe temperature (?C) no data 28.2 ? 2.8 21.2 ? 4.9 14.9 ? 2.0 17.8 ? 6.1 no data 12.5 ? 2.0 Air temperature (?C) daily mean 20.7 ? 6.6 23.0 ? 5.3 19.3 ? 4.1 13.6 ? 2.1 16.3 ? 2.1 28.6 ? 1.6 8.0 ? 2.4 daily minimum 12.3 ? 4.0 15.8 ? 2.2 13.6 ? 3.9 6.3 ? 6.9 9.5 ? 3.1 17.9 ? 2.5 -1.9 ? 1.0 daily maximum 27.7 ? 5.9 30.0 ? 4.4 26.5 ? 5.6 22.3 ? 3.6 22.5 ? 3.2 37.8 ? 1.2 17.1 ? 3.8 Mean monthly rainfall (mm) 15 ? 14 34 ? 12 27 ? 19 4 ? 3 35 ? 20 n/a n/a Mean daily wind speed (m.s-1)* 2.7 ? 2.5 2.9 ? 2.2 1.6 ? 2.1 2.2 ? 2.7 3.6 ? 1.7 2.6 ? 1.4 1.2 ? 1.1 Mean daily radiation (W.m-2)? 231 ? 75 244 ? 85 156 ? 62 137 ? 33 - - - * Wind speed was recorded at a nearby meteorological station (South African Weather Services, Willowmore) ? Radiation was recorded at a nearby weather station located on Blaauwkrantz farm, within 100 km of the field site 56 38 39 40 mean 37 38 39 minimum * 39 40 41 maximum ** Bo dy te m pe ra tu re (? C) * 1 2 3 amplitude * sprsumaut win PE 0.2 0.3 0.4 0.5 max rate * Ra te of bo dy te m pe ra tu re ris e ( ?C . h- 1 ) hotcold Figure 2.3. Mean ? SD of the mean, minimum, maximum, amplitude and rate of body temperature rise of the nychthemeral rhythm of body temperatures of black (closed bars), normal (hatched bars) and white (open bars) springbok over four seasonal periods (where spr = ?spring?, sum = ?summer?, aut = ?autumn?, win = ?winter?) while the springbok were free- living in the Steytlerville Karoo region, during the three-month Port Elizabeth winter study (PE) while the springbok were free-living in the Algoa Dune Strandveld and during the 10 hottest days in summer (hot) and the 10 coldest days in winter (cold) in the Karoo. Repeated measures two-way ANCOVA with body mass as the covariate, * P < 0.05, ** P < 0.001. 57 Post-hoc analysis of the interaction between body temperature indices and season revealed that daily minimum body temperatures of white springbok were significantly lower than those of both black (P = 0.04) and normal (P = 0.04) springbok in ?winter?, and that black springbok had a daily maximum body temperature significantly higher than that of both normal (P = 0.0002) and white (P = 0.0002) springbok during ?spring?. This higher maximum body temperature of the black springbok during ?spring? resulted in an increase in the amplitude of body temperature rhythm, in black springbok compared to both white (P = 0.005) and normal (P = 0.03) springbok. I assessed whether the differences in body temperature amplitudes among the springbok colour morphs were the result of differences in the maximum diurnal rate of body temperature rise (Fig. 2.3, lowest panel). As expected, black springbok showed a faster rate of body temperature rise, compared to both white (P = 0.01) and normal (P = 0.009) springbok, during ?spring?. This difference in the maximum rate of body temperature rise was not the result of differences in the time at which minimum or maximum body temperatures occurred, as there were no significant differences between the time at which maximum (F2,6 = 0.3, P = 0.75), nor the time at which minimum (F2,6 = 1.8, P = 0.25), body temperature occurred among any of the springbok colour morphs, over the one-year study period. There were no significant differences in any of the body temperature indices among the colour morphs during the short-term Port Elizabeth winter study (Fig. 2.3). The lower minimum body temperature of the white springbok compared to the other two springbok colour morphs, during the Karoo ?winter? period was not evident over the short-term Port Elizabeth winter study in which the animals inhabited the Algoa Dune Strandveld, where ambient temperature was not as low as it was in the Karoo (Table 2.1), and the diet of the springbok was supplemented. 2.4.1.3 Environmental effects on the nychthemeral rhythm of body temperature The core body temperature of the animals showed a nychthemeral rhythm with a temperature nadir shortly after sunrise and peak near sunset (Fig. 2.4). On the 58 coldest 10 days in winter in the Karoo, mean (F1,7 = 48.4, P = 0.0002), daily minimum (F1,7 = 20.9, P = 0.003) and daily maximum (F1,7 = 194.1, P < 0.0001) body temperatures were lower than on the hottest days. The amplitude (F1,7 = 10.1, P = 0.02) of the body temperature rhythm also was dampened on the cold days compared to the hot days (Fig. 2.4). However, the only body temperature variable which differed significantly among the colour morphs was the daily maximum body temperature (F2,6 = 6.9, P = 0.03). Black springbok had a significantly higher daily maximum body temperature than did both white (P = 0.03) and normal (P = 0.01) springbok, and these differences were particularly apparent on the hottest days (Fig. 2.3). 38 39 40 41 cold days 0:00 6:00 12:00 18:00 38 39 40 41 hot days Time of day Bo dy te m pe ra tu re (o C) Figure 2.4. Mean ? SD of body temperature of black (black line), normal (dark grey line) and white (light grey line) springbok, as a function of time of day, over the 10 coldest days in winter (mean air temperature of 8.0 ? 2.4?C) and the 10 hottest days in summer (mean air temperature of 28.6 ? 1.6?C) of the one-year study period. 59 To further assess the influence of the ambient temperature on the nychthemeral body temperature rhythm of the different springbok colour morphs, I correlated the daily nychthemeral amplitude of body temperature, averaged for all animals in each colour morph group, with the daily air temperature range on the same day. The body temperature amplitude for each colour morph was correlated significantly with the daily range of air temperature (Fig. 2.5) during ?spring? (P < 0.001, n = 49) and ?summer? (P < 0.0001, n = 90), but not during ?autumn? (P > 0.15, n = 63) or ?winter? (P > 0.05, n = 74) in the Karoo, nor during ?winter? in Port Elizabeth (P > 0.05, n = 42). For each of the colour morphs the slope of the regression equation did not differ between ?spring? and ?summer? (F1,12 = 0.02, P = 0.90), but these slopes were significantly different among the colour morphs (F2,11 = 7.9, P = 0.007). Black springbok showed the steepest slope of 0.06 ? 0.01?C change in body temperature amplitude, per 1?C change in the range of air temperature, versus 0.04 ? 0.01?C for white and 0.03 ? 0.01?C for normal springbok, in both ?spring? and ?summer?. Similarly, for each of the colour morphs the dependence of amplitude of body temperature on range of air temperature did not differ between ?spring? and ?summer? (F1,12 = 0.00002, P = 0.99), but the dependence was significantly different among the colour morphs (F2,11 = 7.1, P = 0.01). Black springbok had the highest dependence of amplitude of body temperature on range of air temperature (black r2 = 0.34 ? 0.09, white r2 = 0.14 ? 0.1 and normal r2 = 0.13 ? 0.09). Daily maximum body temperature, averaged for all animals in each group of colour morphs, also was significantly correlated to daily maximum air temperature during ?spring? (P < 0.0003) and ?summer? (P < 0.0001). For each of the colour morphs the slopes of the regression equations were significantly steeper in ?summer? than in ?spring? (F1,12 = 16.6, P = 0.002), and the slopes of the regression equations were significantly different among the colour morphs (F2,11 = 4.2, P = 0.04), with black springbok showing the steepest slope. Similarly, mean daily minimum body temperature, for each group of colour morphs, also was correlated with daily minimum air temperature, but only weakly, and only during ?autumn? (P < 0.004; black r2 = 0.12, white r2 = 0.15 and normal r2 = 0.20). 60 0 1 2 3 spring 0 1 2 3 summer 0 1 2 3 autumn Am pl itu de of bo dy te m pe ra tu re ( ? C) 0 10 20 30 1 2 3 winter Range of air temperature ( ?C) Figure 2.5. Correlation of the amplitude (the difference between maximum and minimum) of the nychthemeral pattern of body temperature, averaged for all animals in the group, for black (black circle), normal (dark grey circles) and white (light grey circles) springbok, versus the 24-h range of air temperatures on the same day, for each day in my designated ?season?. Regression lines were fitted for the black (solid black line), normal (dashed dark grey line) and white (dotted light grey line) springbok when the correlations were significant in ?spring? (black y = 0.06x + 1.1; white y = 0.04x + 1.0; normal y = 0.03x + 1.2) and ?summer? (black y = 0.06x + 1.0; white y = 0.04x + 1.1; normal y = 0.03x + 1.2), where y = amplitude of body temperature rhythm and x = 24-h range of air temperature. 61 2.4.1.4 Behavioural thermoregulation During daylight hours in the short-term Port Elizabeth winter study, the white and normal springbok intermingled routinely, and appeared to behave similarly, whilst the black springbok typically stayed in a separate group. However, animals from all three colour morphs herded together at night. The white springbok seemed to initiate behavioural events, such as feeding, ruminating or lying down, and their behaviour was imitated by the normal springbok, and then by the black springbok. The three colour morphs displayed similar partitioning of behaviour, over diurnal hours, nocturnal hours and the full 24-h day (Table 2.2). The only behavioural category in which there was a significant association between springbok colour morph and behaviour was the degree of activity (Table 2.2), with the black springbok being less active during daylight hours, and consequently also over the full 24-h day, than were the other two colour morphs. Figure 2.6 illustrates the average 24-h body temperature and activity pattern for animals of each colour morph at 1-h intervals during the short-term Port Elizabeth winter study. The three colour morphs showed similar nocturnal activity profiles with distinct periods of inactivity soon after dusk and just before dawn, but the black springbok were noticeably less active between 06:00 and 09:00 (Fig. 2.6). It is clear from Figure 2.6 that the diurnal rise in body temperature was not a product of activity because the steepest rise in body temperature occurred when the animals were inactive, after the morning activity period peak, and body temperature decreased during the post-dusk activity peak. 62 Table 2.2. Association between partitioning of thermoregulatory behaviour and the three springbok colour morphs, over a 24-h day, as well as over nocturnal and diurnal periods, during the short term Port Elizabeth winter study. Behaviour category Description P value 24-h Diurnal Nocturnal Wind Tolerance or avoidance 0.94 0.92 0.98 Sun orientation Parallel, perpendicular or oblique 0.70 0.70 - Location Sun or shade 0.17 0.17 - Position Standing up or lying down 0.91 0.99 0.87 Proximity >1 or <1m from each other 0.17 0.49 0.21 Feeding Foraging, ruminating or neither 0.70 0.64 0.95 Activity Stationary or mobile 0.04 0.008 0.80 Values are P values obtained from Chi-squared tests, testing association between behaviour and colour morph. Bold values represent significance. 63 00:00 06:00 12:00 18:00 24:00 39 40 Time of day (h) black 0.0 0.5 1.0 normal 0.0 0.5 1.0 Bo dy te m pe ra tu re (?C ) Ac tiv ity (pr op or to n of ob se rv at io ns ) 00:00 06:00 12:00 18:00 24:00 39 40 Time of day (h) 0:00 06:00 12:00 18:00 39 40 Time of day white 0.0 0.5 1.0 Figure 2.6. Activity, expressed as the proportion of the total number of observations for which the animals were active, during each hour of the 24-h day (open bars), and the mean body temperature (solid line) of black, normal and white springbok, as a function of time of day during the short-term Port Elizabeth winter study. 64 2.4.2 Pelt heat transfer characteristics 2.4.2.1 Spectral reflectance Figure 2.7 depicts the spectral reflectance of three springbok pelts, representing each of the three colour morphs, as a function of wavelength, over a waveband simulating the solar spectrum, except for infrared radiation. Spectral reflectance increased with an increase in wavelength for all pelt samples. Mean weighted solar reflectance was different for the three pelt samples, with the black springbok pelt displaying the lowest solar reflectance (0.19 ? 0.11), the white pelt the highest reflectance (0.76 ? 0.01) and the normal springbok pelt intermediate reflectance (0.49 ? 0.06). 400 600 800 0.0 0.2 0.4 0.6 0.8 1.0 Wavelength (nm) Re fle ct an ce Figure 2.7. The spectral reflectance (where a reflectance of one represents total reflection and zero represents total absorption of light relative to barium sulphate) of a black (solid black line), normal (dashed dark grey line) and white (dotted light grey line) springbok pelt over the spectral range, from ultraviolet (350 nm) to infrared (950 nm). 65 2.4.2.2 Thermal conductance The conductance of each pelt increased linearly as wind speed increased (Fig. 2.8A; black r2 = 0.96, P = 0.02; normal r2 = 0.97, P = 0.01; white r2 = 0.93, P = 0.04). There was no significant difference between the slopes of the conductance, as a function of wind speed, among the springbok pelts (ANCOVA, F2,6 = 2.6, P = 0.15), but there was a significant difference in the intercepts (ANCOVA, F2,8 = 23.0, P = 0.0005), with the black springbok pelt having a conductance significantly higher than that of the pelts from the normal and white springbok. 2.4.2.3 Heat load from radiation In all pelts, the fraction of the incident radiant heat load that was conveyed through the pelt to the skin decreased with increasing wind speed (Fig. 2.8B). Fitting one-phase exponential decay curves revealed significant differences among the pelts of each colour morph (ANCOVA, F6,3 = 14.16, P = 0.03), with the black springbok pelt having the highest, and white springbok the lowest, proportion of incident radiant heat load conveyed through the pelt. 66 2 4 6 8 A Co n du ct an ce (W . m - 2 . ? C- 1 ) 0 2 4 0.0 0.2 0.4 B Wind speed (m.s-1) H ea t l o ad fro m ra di at io n (% in cid en t) Figure 2.8. Thermal conductance of the pelt (A), and heat load from radiation conveyed through the pelt as a percentage of incident radiant load (B), for a black (black circles), normal (dark grey circles) and white (light grey circles) springbok pelt, as a function of wind speed. Linear regressions were fitted to the conductance values for black (solid black line, y = 0.61x + 4.5, r2 = 0.96), normal (dashed dark grey line, y = 0.40x + 3.5, r2 = 0.97) and white (dotted light grey line, y = 0.44x + 3.8, r2 = 0.93) springbok pelts, where y = conductance (W.m-2.?C-1) and x = wind speed (m.s-1). One-phase exponential decay curves were fitted to the proportional heat loads for black (solid black line, y = 16.9e-x + 15.0, r2 = 0.96), normal (dashed dark grey line, y = 13.5e-0.5x + 11.1, r2 = 0.99) and white (dotted light grey line, y = 7.6e-0.7x + 7.5, r2 = 0.99) springbok pelts, where y = radiant heat load (% incident) and x = wind speed (m.s-1). 67 2.4.2.4 Potential energetic consequences If animals are employing metabolic heat to maintain homeothermy, then, theoretically, they can spare metabolic heat production by absorbing solar radiation. Based on my model, the potential energetic savings associated with pelt colour were nearly two-fold higher for the black and normal springbok compared to the white springbok (Fig. 2.9), across all seasons. The smallest potential metabolic savings were those in winter, and it was during the cool winter and autumn months when orientation to the sun had the largest effect on metabolic savings. The proportional contribution of total absorbed solar radiation to metabolic rate was low in summer, which was likely the result of the high metabolic rate for springbok during hot dry months (Nagy and Knight, 1994). 0 5 10 parallel spr sum aut win 0 5 10 perpendicular M et ab ol ic sa vin g (% ) Figure 2.9. Metabolic savings of black (closed bars), normal (hatched bars) and white (open bars) springbok over four seasonal periods (where spr = ?spring?, sum = ?summer?, aut = ?autumn?, win = ?winter?), when animals were orientated parallel (upper panel) and perpendicular (lower panel) to solar radiation. Metabolic savings were calculated by expressing the total absorbed solar radiation of each colour morph per day as a proportion of the daily seasonal field metabolic rates of springbok measured by Nagy and Knight (1994). 68 2.5 Discussion My study provides the first comparison of body temperature and behavioural thermoregulation of three colour morphs of a single mammalian species, while the animals were living free under the same environmental conditions. I supplemented the field studies by measuring the thermal characteristics of sample pelts from the colour morphs. The data allowed me to examine the influence of environmental thermal loads on the body temperature of representatives of the three springbok colour morphs, as well as the relationship between behaviour and thermoregulation, under natural conditions. I found that pelts from the colour morphs differed in more than just colour, and that the pelt of the black springbok conveyed more incident radiant heat to the skin than the pelts of the other springbok colour morphs. Possibly related to that finding, the black springbok displayed a higher peak body temperature, a larger 24-h amplitude of the nychthemeral body temperature rhythm, and a faster rate of body temperature rise, than did both white and normal springbok, in hot conditions. In addition, the body temperature of my black springbok was potentially influenced more by air temperature, as illustrated by a steeper slope and stronger correlation between the amplitude of the nychthemeral body temperature rhythm and the range of air temperatures, than was the body temperature of the other colour morphs. The springbok displayed a nychthemeral rhythm of body temperature, with a minimum body temperature in the morning and a maximum in the afternoon (Fig. 2.4). These body temperature patterns were similar to those reported previously for normal springbok (Hofmeyr and Louw, 1987; Mitchell et al., 1997; Fuller et al., 2005). There were no differences apparent in body temperature patterns during winter between the black and normal springbok. During the Port Elizabeth winter study, my springbok, like those observed by others (Bigalke, 1972; Davies and Skinner, 1986; Skinner and Louw, 1996), foraged throughout most of the day. The black springbok displayed the lowest diurnal activity (Fig. 2.6). Since foraging was a major contribution to activity, this finding is consistent with the black springbok spending less time feeding in the winter than did the other colour morphs. Similarly, dark steers spent less time grazing than did light-coated steers 69 in summer (Finch et al., 1984). The reduction in feeding time might reflect lower metabolic costs of homeothermy in the black colour morphs, resulting from better access to solar energy, offsetting the metabolic cost of homeothermy in cold conditions. Such a finding would be consistent with local anecdotal reports that the black springbok survive cold winters better than the other springbok colour morphs. But the higher solar heat load experienced by the black springbok, compared to the other springbok colour morphs, while of potential benefit in the cold, was associated with hyperthermia greater than that of the conspecific colour morphs in hot conditions. In contrast to the situation with the black springbok, there was an apparent disadvantage to the white colour morph during winter, with the white springbok displaying lower minimum body temperatures than did the other springbok in the Karoo winter. This difference was unlikely to have resulted directly from the lower solar radiant input to the white morph, because the minimum body temperature occurred shortly after sunrise, when radiant heat was not a major source of warmth for any of the animals. Instead, I hypothesise that the lower morning body temperature resulted from a reduced ability of the white springbok, compared to the other springbok colour morphs, to maintain a positive energy balance in the cold Karoo. If the lower energy absorption from solar radiation during the day increased the metabolic cost of homeothermy in the white springbok, while they fed for the same time and had similar energy intake as did the normal springbok, then they were likely in deficit, relative to the other springbok colour morphs, which manifested as an exaggerated nocturnal nadir of body temperature. Evidence supporting that hypothesis is that energy restriction, or starvation, leads to a lower nadir in the nychthemeral rhythm of body temperature in many species, including sheep (Piccione et al., 2002), desert goats (Choshniak et al., 1995; Ahmed and El Kheir, 2004), horses (Kronfeld, 1993), many bird species (Hohtola et al., 1991), rodents (Sakurada et al., 2000; Bae et al., 2003), and primates (G?nin and Perret, 2003). My conclusions about the reduced ability of the white springbok to maintain a positive energy balance in the Karoo winter must remain tentative, however, because some of the loggers 70 stopped recording prematurely which resulted in a low sample size during this winter period. The minimum body temperatures of my white springbok were no different to those of the other colour morphs during the short-term Port Elizabeth winter study, where conditions were milder and animals? diets were supplemented. To ascertain whether the differences I observed in the body temperature profile of the springbok colour morphs indeed were the result of differences in their pelt properties, I investigated pelt heat flow characteristics. The white springbok pelt had the highest reflectance in the visible spectrum, and the black springbok pelt had the highest absorbance (Fig. 2.7). These differences in absorbance contributed to the black springbok having the highest, and white springbok the lowest, heat load from radiation (Fig. 2.8B). The biophysical part of the study was limited, though, because I was able to examine only a single pelt sample of each springbok colour morph, and I do not know if differences in pelt conductance and solar heat gain can be compensated, in living animals, by autonomic responses, such as pilo- erection or vasodilation. Based on the pelt heat flow characteristics of the excised pelts, black springbok should be more prone to losing heat when ambient temperature is cooler than body temperature, because the thermal conductance of the black springbok pelt was higher than that of the other springbok pelts (Fig. 2.8A). In the presence of solar radiation, though, the characteristics of the pelt of the black springbok resulted in the transmission of more solar heat to the skin, and there was no apparent disadvantage to homeothermy of the black springbok, compared to the normal springbok, in winter. Several earlier studies have investigated the impact of pelt colour on heat load from radiation. My data fit the traditional view that animals with darker colouration acquire a larger heat load from solar radiation than do animals with lighter colouration, because they absorb more short-wave radiation (Hamilton and Heppner, 1967; Heppner, 1970; Cloudsley-Thompson, 1979; Walsberg, 1982). Although reflectance of short-wave radiation is colour dependent, the increase in reflectance from the blue end of the spectrum to the near infrared spectrum is 71 independent of pelt colour, and almost all long-wave radiation is absorbed, whatever the visible colour (Dawson and Brown, 1970; Cena and Monteith, 1975; Finch et al., 1984). I analysed pelt reflectance only within the visible spectrum. The reflectance of my black and white springbok pelts were similar to those reported for black goats (Finch et al., 1980), white cattle (Hutchinson and Brown, 1969) and black and white finches (Heppner, 1970). The reflectance of my normal springbok pelt was similar to that reported previously for normal springbok (Hofmeyr and Louw, 1987). However, because of the way they weighted their reflectance calculation to take into account differences in reflectance between the buff and pale areas of the normal springbok pelt, Hofmeyr and Louw (1987) arrived at a slightly higher mean weighted absorbency (0.55) than I did for my normal springbok (0.51 ? 0.06). This weighted absorbency of the normal springbok pelt is lower than that reported for other African ungulates (Finch, 1972), a property proposed to offer adaptive advantages for desert life (Skinner and Louw, 1996). The thermal consequences of pelt structure are dependent not simply on reflectance, but also on other properties that affect the depth to which radiation penetrates the fur or feathers (Kovarik, 1964; Walsberg et al., 1978; Walsberg, 1983). If penetrance is low, as in dense or matted pelts, radiant heat is absorbed at the fur or feather surface and returned to the environment, reducing the potential contribution to the heat load on the animal. Such considerations become especially relevant when wind speed increases and the insulation to heat flow from the point of absorbance to the environment is reduced as the air boundary layer is reduced (see below). In such circumstances, pelt colour may have little impact on the thermal load from solar radiation that actually reaches the animal (Walsberg, 1982; Walsberg, 1990; Maloney and Dawson, 1995; Walsberg and Wolf, 1995; Dawson and Maloney, 2004). However, if all other morphological characteristics of pelts are equal, as I have been led to believe for my springbok colour morphs (Kruger et al., 1979), then penetrance should be inversely related to absorbance (Hutchinson and Brown, 1969; Cena and Monteith, 1975). I discovered, though, that the pelts differed in more than just colour; the pelt of the 72 black colour morph also had a significantly higher conductance. Springbok pelts are thinner and have a higher conductance than predicted for an antelope of its body size (Hofmeyr, 1981). The thin pelage may well be the reason that pelt colour influenced the radiant heat load so strongly in this species. The outcome of the combination of pelt thermal properties was that the white springbok pelt prevented heat load from radiation reaching the animals more than did the normal and black pelts, an outcome that concurs with results obtained for goats (Finch et al., 1980), cattle (Hutchinson and Brown, 1969; Finch et al., 1984; Hansen, 1990) and pigeons (Walsberg et al., 1978). The fact that the lines relating heat load from radiation to wind speed (Fig. 2.8B) converged at higher wind speeds supports the view that the effects of radiation absorbance predominated at low wind speeds, whilst small differences in penetrance exerted an effect as wind speed increased (Hutchinson and Brown, 1969; Walsberg et al., 1978). Although the radiant heat load through the colour morph pelts converged as wind speed increased, the heat load through the white pelt remained lower than that through the black pelt, even at high wind speeds. The greater access to radiant heat predicted, from the biophysical properties of the pelt, for the black animals should have thermoregulatory advantages in the cold, saving metabolic energy. My model of energetic savings (Fig. 2.9) indicates that the black springbok potentially were able to obtain almost twice the energy flux from solar radiation, in the field, as could their white counterparts. The reduced feeding activity Ms de Witt observed in the black springbok, compared to the other springbok, is consistent with such a saving. A reduced energetic cost of homeothermy in dark- coloured animals, via their better access to solar heat, has been reported often (Hamilton and Heppner, 1967; Lustick, 1969; Heppner, 1970; Dmi'el et al., 1980; Finch et al., 1980; Burtt, 1981; Finch et al., 1984). However, as conditions get hotter with climate change dark-coloured individuals may be disadvantaged. White-coated Santa In?s ewes showed better tolerance to heat stress (McManus et al., 2009) and are therefore likely to be better adapted to hot conditions in the future than their darker counterparts. 73 Thus I conclude that there is adaptive significance of colour morphs, and there probably are selective pressures acting against both the white and the black springbok at different times of the year. The black springbok seem able to reduce energy expenditure in winter, as predicted and in accord with anecdotal reports, but experience higher solar heat load in hot conditions. On the other hand, the white springbok apparently is living closer to the energetic edge in winter, possibly because its lower heat load from solar radiation requires a higher metabolic cost of homeothermy. From a physiological perspective, the normal springbok appears to occupy a compromise position, with better energy balance than the white springbok in winter and less overheating than the black springbok in summer, a compromise which may explain why the black and white springbok rarely occur naturally, despite the genes for those colour morphs being present. The crypsis function of the normal springbok?s colouration may further explain the preponderance of its prevalence, as colour is not only important for thermoregulation but also serves an aposematic function (Caro, 2009). Nevertheless, maintaining such genetic diversity within a population, particularly diversity for a trait that adapts individual animals to different thermal environments, may provide an important pre-adaptation to climate change. 2.6 Acknowledgements I thank Mr Ronald Kirkman for his hospitality and the use of his property, Hopedale. I thank the staff and students of the Zoology Department of Nelson Mandela Metropolitan University for their assistance with animal capture and management. I thank Andr? Matthee, the Head of the Lichtenberg Game Breeding Centre of the National Zoological Gardens, whose advice, support and game capture and management skills made this project viable. I thank Georg von Bormann for his help and support on site, and Phil Withers and Chris Cooper from the University of Western Australia for their assistance with the biophysical measurements made on the pelts. 74 ___________________________________________________________ CHAPTER 3 ___________________________________________________________ 3 Adapting to arid environments: thermoregulation and cathemerality of free-living Arabian oryx (Oryx leucoryx) in a hyper-arid desert Data and ideas presented in this chapter have been written up as a scientific paper and submitted to Physiological and Biochemical Zoology. Hetem R.S., Strauss W.M., Fick L.G., Maloney S.K., Meyer L.C.R., Shobrak M., Fuller A. and Mitchell D. Thermoregulation and cathemerality of free-living Arabian oryx (Oryx leucoryx) in a hyper-arid desert. 75 3.1 Abstract In a desert environment, endotherms have to trade off thermoregulation, osmoregulation and energy acquisition. One physiological mechanism proposed to conserve body water at the expense of thermoregulation is heterothermy, a variability in body temperature beyond the limits of homeothermy. Alternatively, endotherms may reduce diurnal activity and seek shade to maintain homeothermy at the expense of energy acquisition. With the help of my colleagues, I assessed the thermoregulatory strategies of the arid-adapted Arabian oryx (Oryx leucoryx) by implanting data loggers to measure body temperature and activity patterns, for a one-year period, in five free-living oryx inhabiting a desert in Saudi Arabia. As predicted for adaptive heterothermy, during hot months compared to cooler months, maximum daily body temperatures were not only higher (41.1 ? 0.3?C vs. 39.7 ? 0.1?C, P = 0.0002) but minimum body temperatures also were lower (36.1 ? 0.3?C vs. 36.8 ? 0.2?C, P = 0.04), resulting in a larger 24-h amplitude of body temperature rhythm (5.0 ? 0.5?C vs. 2.9 ? 0.2?C, P = 0.0007). Body temperature variability was associated not only with ambient temperature, but also with water availability, with oryx displaying a greater 24-h amplitudes of nychthemeral rhythm of body temperature during warm dry months compared to warm wet months (3.6 ? 0.6?C vs. 2.3 ? 0.3?C, P = 0.005). Oryx displayed flexibility in their activity patterns and shifted from a continuous rhythm with crepuscular peaks during warm months to a nocturnal rhythm during hot months. The attenuation in activity over daylight hours during the hot months compared to warm months (25 ? 8% vs. 48 ? 5%, P = 0.0004) was accompanied by the selection of cooler microclimates, up to 12?C cooler than that experienced in the sunshine, during the heat of the day. An additional thermoregulatory strategy proposed to conserve water is selective brain cooling. Although the magnitude of selective brain cooling was small (< 1?C), my oryx frequently employed selective brain cooling. Arabian oryx therefore employ heterothermy, cathemerality and, I believe, selective brain cooling to survive the extremely hot and arid conditions of the Arabian Desert. 76 3.2 Introduction Since many arid regions of Africa are predicted to get both hotter and drier with climate change (Boko et al., 2007), understanding the physiological mechanism exhibited by desert antelope would assist future studies in predicting physiological adaptations to desertification. Amongst large desert-dwelling mammals, the species which arguably faces the greatest challenge to homeostasis is the Arabian oryx (Oryx leucoryx). With a body mass of 80-100 kg, this artiodactyl inhabits one of the hottest deserts in the world, in Saudi Arabia, and survives most of the year, including the hottest summer months, without access to drinking water. Because of the subsequent limitations in the use of evaporative cooling to dissipate metabolic heat and heat gained from the environment, the animals have to trade off thermoregulation, osmoregulation and acquisition of energy. Desert artiodactyls display a summer nadir in metabolic rate (Nagy and Knight, 1994; Williams et al., 2001; Ostrowski et al., 2006a; Ostrowski et al., 2006b). However, these low metabolic rates in summer are associated with low quality and quantity of food (Choshniak et al., 1995; Piccione et al., 2002; Ahmed and El Kheir, 2004), and it has not been determined whether the reduced metabolism occurs in response to reduced energy intake or as an adaptation to reduce the requirements for heat dissipation. Another physiological mechanism exhibited by desert artiodactyls, with the capacity to reduce evaporative cooling, is adaptive heterothermy. It is defined as the storage of body heat during the day, with a consequent rise in body temperature, reducing both heat gain and evaporative heat loss (Schmidt-Nielsen et al., 1957). This stored heat then can be dissipated non-evaporatively during the colder night, allowing body temperature to fall. Animals using heterothermy therefore would display an increase in the amplitude of nychthemeral rhythm of core body temperature under high environmental heat loads. Such heterothermy might result simply from failure of thermoregulation in the heat, but also could be an active process, termed ?adaptive heterothermy?, in which the 24-h nadir of body temperature is depressed, even in well-fed animals, allowing more heat to be stored before body temperature reaches lethal levels. 77 Early studies reporting the existence of adaptive heterothermy were performed on captive animals, often deprived of both drinking water and behavioural thermoregulatory opportunities (Schmidt-Nielsen et al., 1957; Taylor, 1969a), however, adaptive heterothermy appears to be absent in most free-living African artiodactyl species studied (Mitchell et al., 2002), in environments similar to those in which captive artiodactyls do display heterothermy. Only two studies to date have reported the existence of adaptive heterothermy in free-living artiodactyls, namely the Arabian oryx (Ostrowski et al., 2003) and the smaller, sympatric, Arabian sand gazelle (Gazella subgutturosa marica, Ostrowski and Williams, 2006). The Arabian oryx displayed an amplitude of nychthemeral rhythm of body temperature which increased from 1.5 ? 0.6?C in winter to 4.1 ? 1.7?C in summer (Ostrowski et al., 2003), but the recordings of body temperature, especially at night, were sporadic, and night recordings and daytime recordings were not necessarily contiguous. Although recordings for the sand gazelle were continuous, these antelope displayed an amplitude of nychthemeral rhythm of body temperature of 1.7 ? 0.3?C in winter, which increased to only 2.6 ? 0.8?C in summer (Ostrowski and Williams, 2006). So the nychthemeral amplitudes for both of these species, in spite of the thermal stress to which they were exposed with mean daily maximum air temperature exceeding 40?C, were substantially lower than the 6?C variation originally reported for captive camel (Camelus dromedaries, Schmidt-Nielsen et al., 1957), eland (Tragelaphus oryx) and oryx (Oryx gazella beisa, Taylor, 1969a). Nevertheless, if adaptive heterothermy does indeed exist, it is likely to occur in a species able to survive independent of water in an arid environment, such as the Arabian oryx (Tear et al., 1997; Williams et al., 2001; Ostrowski et al., 2002; van Heezik et al., 2003). Another conspicuous thermoregulatory adaptation of artiodactyls, including the genus Oryx (Maloney et al., 2002), is selective brain cooling. Selective brain cooling, the reduction of brain temperature below arterial blood temperature (IUPS Thermal Commission, 2003), originally was hypothesized to protect the 78 brain from a rise in body temperature, such as that associated with heterothermy (Schmidt-Nielsen et al., 1957). However, recent literature proposes that arid-zone mammals possessing a carotid rete may employ selective brain cooling to attenuate thermal drive, because selective brain cooling reduces hypothalamic temperature, which reduces evaporative heat loss and ultimately conserves water (Mitchell et al., 2002). Selective brain cooling therefore may be a mechanism which gives rise to heterothermy, rather than protection from it. Selective brain cooling has not been investigated in Arabian oryx. A third thermoregulatory mechanism that might reduce the need for evaporative cooling is shade-seeking behaviour and a reduction of diurnal activity, both of which have been observed in Arabian oryx in summer (Stanley Price, 1989; Seddon and Ismael, 2002; Ostrowski et al., 2003). Such behaviour, though, must reduce diurnal foraging, and, unless there is a compensatory increase in nocturnal foraging, must compromise energy acquisition, especially because resources are limited in the summer. Whether Arabian oryx increase nocturnal foraging as a result of the high heat loads during summer months is not known. With the help of my colleagues, I therefore measured abdominal, brain and arterial blood temperature, activity and microclimate selection, of free-living Arabian oryx occupying a hyper-arid desert in Saudi Arabia to ascertain the importance of, and the potential interactions between, the reputed water-saving mechanisms. I employed data loggers to obtain regular measurements of body temperatures, activity and microclimate over one year. Such measurements have not been achieved previously for Arabian oryx, or indeed any free-living animal species. 3.3 Materials and methods 3.3.1 Animals and habitat The study took place between March 2006 and April 2007 within the 2200 km2 Mahazat as-Sayd Protected Area (28?15? N, 41?40? E) in the open steppe desert in 79 west-central Saudi Arabia that is both the historical and current habitat for Arabian oryx (Oryx leucoryx, Ostrowski et al., 1998). Three male and two female, adult, wild-born oryx were captured, under veterinary supervision, in the protected area, in mid-March 2006. The oryx were habituated in outdoor pens for two weeks to reduce potential peri-operative stress. An additional male oryx was obtained from the breeding herd at the National Wildlife Research Center in Taif (21?15' N, 40?42' E). This male oryx remained in a partially open pen throughout the 11-month study period with lucerne and water available ad libitum. All experimental procedures were approved by the Animal Ethics Screening Committee of the University of the Witwatersrand (protocol no. 2005/87/5). 3.3.2 Surgery The oryx, at both locations, were darted and anaesthetised in the holding pens with etorphine hydrochloride (2.5 mg intramuscularly (I.M.), M99, C-Vet, Leyland, UK) and, once recumbent, were transported to a temporary operating theatre within 200 m of the pens. At the surgery, the animals were placed in sternal recumbency, with their heads elevated. Animals were intubated and anaesthesia was maintained with 2-6% isoflurane (Aerrane, Astra Zeneca, Johannesburg, South Africa), administered in 100% oxygen. Respiratory rate, heart rate, arterial oxygen saturation (pulse oximeters, Nonin 9847V, Nonin Medical, North Plymouth, USA) and rectal temperature (thermocouple thermometer, BAT-12, Physitemp, Clifton, USA) were monitored throughout the surgery, which lasted approximately two hours. Under sterile surgical conditions, several miniature data loggers were implanted. All loggers were covered in an inert wax (Sasol, Johannesburg, South Africa), and dry-sterilized in formaldehyde vapour before implantation. After administration of a local anaesthetic (0.04 g lignocaine hydrochloride, subcutaneously (S.C.), Bayer Animal Health (Pty) Ltd, Isando, South Africa), incision sites were shaved and sterilized with povidone iodine antiseptic (Vetedine, Vetoquinol, Lure, France). Each oryx was fitted with data loggers connected to thermistor sensors for temperature measurement in the carotid artery, brain, and abdominal cavity. For 80 measurement of carotid temperature, a thermistor inserted in a blind-ended and thin-walled polytetrafluoroethylene (PTFE) catheter (o.d. 1.35 mm, i.d. 0.97 mm; Straight Aortic Flush 4F Catheter, Cordis, The Netherlands) was advanced 60 mm into the left common carotid artery towards the heart, at a position midway along the length of the neck, and secured with a purse-string suture in the artery wall. Outside the artery, the PTFE tube was sealed on a PTFE-coated co-axial cable (150 mm long, o.d. 3 mm, Belden, Richmond, USA) connecting the thermistor to the temperature-sensitive data logger (see temperature measurements). The data logger was positioned subcutaneously, dorsal to the artery. For measurement of brain temperature, a second data logger, connected to the thermistor positioned in the brain, was positioned subcutaneously, caudal to the base of the left ear. Its PTFE-coated cable was advanced subcutaneously over the skull, where it was connected to a head plate and guide tube. The guide tube, constructed from cellulose acetate butyrate tubing (40 mm long, o.d. 3.2 mm, i.d. 1.6 mm; World Precision Instruments, Sarasota, Florida, USA) sealed at the tip by a steel cap, was inserted through a 2 mm diameter burr hole, which was drilled through the cranium, at appropriate co-ordinates pre-determined from head sections of dead oryx of similar size, so that the probe tip would be positioned near the hypothalamus. The brain guide tube was connected to a polyvinyl chloride head plate (20 ? 10 ? 5 mm), which was secured to the skull by two bone screws and covered by skin so that it lay secured in a subcutaneous position. For measurement of abdominal temperature, a third data logger, with an internal temperature sensor, was inserted, via an incision in the paralumber fossa, into the abdominal cavity. The muscle layer was sutured closed and an activity logger (Actical, Mini-Mitter Corporation, Bend, Oregon, USA), for measurement of activity, was tethered to the outer layer of the abdominal muscle before the skin was sutured closed over the logger. Wounds were treated with a topical antiseptic spray (Necrospray, Centaur Labs, Johannesburg, South Africa). Each animal received a long-acting antibiotic (450 mg I.M., penicillin, Norocillin La, Norbrook Laboratories Ltd., Newry, Northern 81 Ireland), a non-steroidal anti-inflammatory analgesic (100 mg I.M., phenylbutazone, Dexaphenylarthrite injectable solution, Vetoquinol Veterinary Pharmaceuticals, Cedex, France), a long-acting parasiticide (2 ml S.C., Ivermectin, Noromectin, Norbrook Laboratories Ltd., Newry, Northern Ireland) and a multivitamin (9 ml I.M., Multivit injectable solution, Univet Ltd., Ireland). Before anaesthesia was terminated, a neck collar (MOD-500 Telonics, Inc. Mesa, AZ, USA) was fitted to each oryx. In addition to a tracking radio transmitter, each collar supported a miniature black globe thermometer (?miniglobe?), to allow for the dynamic measurement of the microclimate that the oryx chose to occupy. This technique has been validated previously on other ungulate species (Appendix 1, Hetem et al., 2007). Miniglobe temperature was measured by a small temperature-sensitive data logger (see temperature measurements) inserted into the centre of a matt-black hollow copper sphere (30 mm diameter, Press Spinning & Stamping Co., Cape Town, South Africa). The globe was attached to a 10-mm diameter, polyvinyl chloride rod, which in turn was attached to the outer surface of the collar. A weight on the ventral side of the collar ensured that the miniglobe remained over the dorsum of the neck and could not be shaded by the animal?s body. Following surgery, the oryx were transported back to their pens, where they became ambulatory within ~ 10 min after the effect of etorphine was reversed with diprenorphine hydrochloride (7.5 mg intravenously (I.V.), M5050, C-Vet, Leyland, UK). After a two-week recovery period, and following veterinary inspection, the five oryx in the Mahazat as-Sayd Protected Area were released into a 2 km2 fenced enclosure with natural forage and water available ad libitum. Ten days later the five oryx were allowed to enter and range freely within the Mahazat as-Sayd Protected Area. The five oryx separated from each other shortly after their release and were left undisturbed, apart from the occasional serendipitous visual contact with rangers, for one year. 82 In April 2007, the oryx were tracked, captured and transported to the holding pens. The animals were anaesthetized once again and the data loggers were removed under a surgical procedure similar to that used for the original implantation. The animals? wounds had healed and there were no signs of infection from the initial surgery. Most of the abdominal loggers were found intra- abdominally in the pelvic canal and were not encapsulated in adhesive tissue. After surgery and a two-week recovery period in the pens, the oryx were re- released into the Mahazat as-Sayd Protected Area. A similar procedure was used to retrieve the loggers from the oryx housed at the National Wildlife Research Center in Taif. 3.3.3 Temperature and activity measurements The miniature thermometric data loggers (StowAway XTI, Onset Computer, Pocasset, Massachusetts, USA) used to measure brain, carotid artery and abdominal temperature had outside dimensions of ~ 50 ? 45 ? 20 mm and a mass of ~ 40 g when covered in wax. These loggers had a resolution of 0.04?C and measurement range from + 34?C to + 46?C. Temperature sensors used to measure brain and carotid blood temperatures were constructed from ruggedized glass- coated bead thermistors with insulated extension leads (bead diameter 0.3 mm; AB0E3-BR11KA103N, Thermometrics, Edison, New Jersey, USA). The scan interval of the brain and carotid blood loggers was set at 5 min and that of the abdominal logger was set at 15 min. Miniglobe temperatures were recorded every hour using a smaller thermometric data logger (iButton DS1922T, Maxim, Dallas Semiconductor, Texas, USA), which weighed ~ 10 g. These loggers had a resolution of 0.5?C and a measurement range from 0?C to 125?C. All temperature sensors and loggers were calibrated against a high-accuracy thermometer (Quat 100, Heraeus, Hanau, Germany) in an insulated water bath. After calibration, the loggers and their sensors measured blood, brain and abdominal temperature to an accuracy of better than 0.05?C and miniglobe temperature to better than 0.5?C. 83 The activity logger recorded movement via a multidirectional, piezoelectric accelerometer which was sensitive to 0.05g. Movement was recorded over 5-min intervals at a sampling rate of 32 Hz. The logger had dimensions of 40 x 40 x 15 mm and weighed ~ 40 g when covered in wax. 3.3.4 Climatic data measurements I collected climatic data from a portable weather station erected near the Mahazat as-Sayd Protected Area, at the Saja/Umm ar-Rimth Protected Area (23?22? N, 42?45? E), and also at the National Wildlife Research Center in Taif. I recorded wind speed (m.s-1), solar radiation (W.m-2), dry-bulb temperature (?C) and relative humidity (%). I also recorded miniature (30 mm diameter) black globe temperature (?C) and rainfall on site at the Mahazat as-Sayd Protected Area for the duration of the study period. Photoperiod was calculated as the daily difference between sunrise and sunset times obtained from the U.S. Naval Observatory website (http://aa.usno.navy.mil/data/). 3.3.5 Data analysis For analysis of seasonal patterns, data from the 11-month study were averaged over four seasonal periods, based on prevailing climatic conditions (Table 3.1), namely warm wet (April to May), hot dry (June to August), warm dry (September to November) and cool dry (December to February). March was not included in my analysis to allow for post-surgical recovery. Although I called the April to May period ?warm wet?, only 24 mm of rain fell during this period (see Table 3.1). For successive 24-h periods, I calculated mean, minimum, maximum and amplitude of nychthemeral rhythm of body temperature as measured in the abdominal cavity. I averaged the body temperature parameters for each individual, for each period, and performed one-way repeated-measures analysis of variance (ANOVA) to test for differences in the body temperature profile across the four periods, with Newman-Keuls multiple comparison tests to identify sources of differences. 84 To further assess the influence of the environment on body temperature patterns, I correlated the various parameters of the body temperature profile and the corresponding prevailing environmental conditions. Since oryx seek shade during the heat of the day (Stanley Price, 1989), dry-bulb air temperature probably provides a better approximate of operative temperature than does the globe temperature in the sun. I therefore correlated the mean 24-h body temperature, averaged for all five oryx, with mean 24-h air temperature, and the amplitude of nychthemeral rhythm of body temperature against the 24-h amplitude of air temperature. I also correlated mean minimum body temperature, within each 24-h period, against the concurrent 24-h minimum and maximum air temperature, and 24-h maximum body temperature against the concurrent 24-h maximum air temperature. I fitted simple non-linear regressions where appropriate. To test the effect of season on the magnitude of the amplitude body temperature rhythm, I correlated the 24-h amplitude of nychthemeral rhythm of body temperature, averaged for all five oryx, against photoperiod. One of my oryx (Oryx 5) died, of unknown causes in February, that is 10 months after surgery, and the data logger which had measured his carotid temperature was not recovered. The data loggers which measured carotid temperature in three oryx (Oryx 1, 3 and 4) failed within 20 days after surgery. The data logger which measured brain temperature in the captive male oryx failed within a month after surgery. I obtained a complete set of brain temperature and carotid blood temperature data for only a single free-living male oryx (Oryx 2). I analyzed the relationship between brain temperature and carotid blood temperature by sorting all 5-min measurements of carotid blood temperature into 0.1?C classes, and determining the mean, standard deviation, maximum, and minimum brain temperature at each class of carotid blood temperature. The battery life of the activity loggers restricted measurements to the warm wet, hot dry and warm dry periods. To account for differences in the sensitivity of individual activity loggers I calculated activity as a percentage of the maximum reading each logger recorded, over the entire period. Since activity levels in light 85 and dark phases may be linked to the duration of the phases (Hill, 2006), I calculated ?diurnal? activity between 06:00 and 18:00 and ?nocturnal? activity between 18:00 and 06:00. I compared mean 24-h activity, actual diurnal activity and diurnal activity expressed as a proportion of 24-h activity, across periods, using a one-way repeated-measures ANOVA with Newman-Keuls post-hoc analysis. The miniglobe thermometers broke off the collars of all three male oryx, so I retrieved data on the microclimate selected only for the two female oryx (Oryx 1 and 4). I correlated collar miniglobe temperatures against weather station miniglobe temperatures using linear Pearson procedures. To assess whether the animals were conforming to ambient conditions or were selecting microclimates, I tested whether the slope of the regression equation was significantly different from one (the slope of the line of identity). In addition, I tested whether the slope and elevation of the regression equations were significantly different across seasonal periods using an analysis of co-variance (ANCOVA). I used Statistica (kernel release 5.5 for Windows, StatSoft, Inc. (1999), Tulsa Oklahoma, USA) and GraphPad Prism (version 4.00 for Windows, GraphPad Software, San Diego California USA) for statistical analyses. Values are expressed as mean ? SD and P < 0.05 was considered significant. 3.4 Results 3.4.1 Climate Environmental conditions over the four periods are shown in Table 3.1. Ambient temperatures were similar during the warm wet and warm dry periods. On average, the hot dry period was 15?C warmer than the cool dry period. Air and black globe temperature varied as a function of time of day, peaking just after solar noon (12:00) and reaching a minimum just before sunrise. Solar radiation showed the expected bell-shaped distribution and wind speed increased in the late afternoon. Rainfall totalled 31 mm over the study period, substantially lower than the ten- 86 year average of 100 ? 60 mm. April was the wettest month (17 mm) but some rain fell during March (7 mm) and May (7 mm). March was not included in my analysis to allow for post-surgical recovery. Table 3.1. Environmental conditions (mean ? SD) during the four periods in which the oryx were free-living in the Mahazat as-Sayd Protected Area Warm wet (Apr-May) Hot dry (Jun-Aug) Warm dry (Sep-Nov) Cool dry (Dec-Feb) Globe temperature (C) 24-h mean 24-h minimum 24-h maximum 33.8 ? 3.4 18.5 ? 3.3 53.3 ? 5.2 37.9 ? 1.7 22.9 ? 2.5 57.1 ? 3.9 32.1 ? 3.8 16.4 ? 3.8 52.3 ? 5.4 23.3 ? 3.6 10.3 ? 3.4 42.6 ? 6.1 Air temperature (C) 24-h mean 24-h minimum 24-h maximum 27.7 ? 4.0 19.9 ? 4.2 34.1 ? 4.0 34.0 ? 1.5 25.7 ? 2.3 40.4 ? 1.3 26.1 ? 4.6 18.2 ? 3.8 33.2 ? 4.8 15.4 ? 3.7 8.9 ? 3.6 22.0 ? 4.3 Mean 24-h wind speed (m.s-1) 4.4 ? 1.4 3.9 ? 1.0 3.9 ? 1.1 4.5 ? 1.4 Mean daytime radiation (W.m-2) 561 ? 93 580 ? 50 456 ? 65 394 ? 79 Total rainfall (mm) 24 0 0 0 Mean time of sunrise 05:36 ? 0:15 05:29 ? 0:12 06:16 ? 0:16 06:58 ? 0:07 Mean time of sunset 18:48 ? 0:10 19:04 ? 0:11 17:48 ? 0:25 17:40 ? 0:18 3.4.2 Body temperature The typical body temperature patterns for a single free-living male oryx and the captive male over the 11-month study period are shown in Figure 3.1. Also shown are the air temperatures prevailing near to where the animals were located. There was a clear increase in the 24-h amplitude of body temperature rhythm during the hot dry period (June to August) in the free-living male oryx, which was not evident in the captive male oryx. The nychthemeral rhythm of body temperature averaged for all five free-living oryx is shown in Figure 3.2. Although ambient temperature had an apparent effect on the nychthemeral rhythm of body temperature, as the oryx showed an increased amplitude of body temperature 87 rhythm during the hot dry compared to cool dry period (Fig. 3.2A), ambient temperature was not the only factor to influence body temperature rhythm. During the warm wet and warm dry periods environmental conditions were similar (Fig. 3.2D, Table 3.1), but there was a difference in the amplitude of body temperature rhythm. In particular, oryx displayed a lower minimum body temperature during the warm dry than during the warm wet period (Fig. 3.2C). 35 37 39 41 43 A B C D B o dy te m pe ra tu re (?C ) A B C D A M J J A S O N D J F 0 10 20 30 40 50 A B C D Month Ai r te m pe ra tu re (?C ) A M J J A S O N D J F A B C D Month Figure 3.1. The top two panels show the original record of 15-min recordings of body temperature from a single free-living male oryx (Oryx 2, left panel) and the captive male oryx (right panel) , which had access to water ad libitum, over the 11-month study period (April 2006 to February 2007). The bottom two panels show the air temperature recorded at nearby weather stations, namely the Saja/Umm ar-Rimth Protected Area (left panel) and National Wildlife Research Center in Taif (right panel), over the same period. The dotted lines separate the data into the four periods analyzed, namely warm wet (panel A), hot dry (panel B), warm dry (panel C) and cool dry (panel D). 88 36 38 40 42 hot dry cool dry A Bo dy te m pe ra tu re (?C ) warm dry warm wet C 0:00 6:00 12:00 18:00 0 10 20 30 40 50 B hot dry cool dry Time of day Ai r te m pe ra tu re (?C ) 0:00 6:00 12:00 18:00 warm dry warm wet D Time of day Figure 3.2. The top two panels show the nychthemeral rhythm of body temperature (mean ? SD), averaged for all five free-living oryx, over the four periods. Panel A shows data from the cool dry (light line) and hot dry (dark line) periods, and panel C from the warm wet (light line) and warm dry (dark line) periods. The bottom two panels show the nychthemeral rhythm of air temperature (mean ? SD) over the four periods. Panel B shows data from the cool dry (light line) and hot dry (dark line) periods, and panel D from the warm wet (light line) and warm dry (dark line) periods. A one-way repeated-measures ANOVA revealed that mean (Fig. 3.3A, F3,9 = 11.2, P = 0.002), minimum (Fig. 3.3B, F3,9 = 15.4, P = 0.0007), maximum (Fig. 3.3C, F3,9 = 41.4, P < 0.0001) and amplitude (Fig. 3.3D, F3,9 = 23.0, P = 0.0001) of nychthemeral body temperature rhythm differed significantly between periods (Fig. 3.3). Maximum body temperatures (Fig. 3.3C) and amplitude of body temperature rhythm (Fig. 3.3D) appeared to be influenced by changes in air temperature, as these variables were higher during the hot dry months than during the other dry periods. However, minimum body temperatures appeared to be influenced by changes in water availability, as minimum body temperature was higher during the warm wet months than during the dry periods (Fig. 3.3B). Notably, minimum body temperature was lower during the warm dry period than during the warm wet period, despite ambient temperatures being similar. This lower minimum body temperature during the warm dry period resulted in a lower 89 mean body temperature (Fig. 3.3A) and a greater body temperature amplitude during the warm dry, than during the warm wet, period. To further assess the influence of ambient temperature on body temperature, I plotted body temperature variables against air temperature (Fig. 3.4). Mean 24-h air temperature accounted for 35% of the variability in mean 24-h body temperature (P < 0.0001, Fig. 3.4A). There was a disjunction in the relationship, however, above a mean air temperature of about 25?C, probably because mean 24- h body temperature was lower during warm dry than during warm wet periods (Fig. 3.3A), even though air temperatures were similar. Maximum 24-h air temperature accounted for 85% of the variability in maximum 24-h body temperature, across all 313 days of the study, when a best-fit power curve was fitted to the data (Fig. 3.4B). Above a maximum 24-h air temperature 36?C, maximum 24-h body temperature increased by 0.25?C per 1?C increase in maximum 24-h air temperature. Conversely, over the full range of air temperatures, minimum 24-h body temperature was independent of both minimum 24-h air temperature (Fig. 3.4C), and maximum 24-h air temperature (Fig. 3.4D). There was a trend for minimum 24-h body temperatures to increase with increasing minimum 24-h air temperatures up to 15?C (r2 = 0.57, P < 0.0001), and increasing maximum 24-h air temperatures up to 30?C (r2 = 0.47, P < 0.0001). Again there was a disjunction in the relationship when minimum 24-h air temperature was between 15?C and 25?C and when maximum 24-h air temperature was between 30?C and 38?C, corresponding to the low minimum body temperatures displayed by the oryx during the warm dry, and not during the warm wet, period. 90 38 39 * A * * * M ea n 24 - ho u r bo dy te m pe ra tu re (?C ) 35 37 39 B * ** * M in im u m bo dy te m pe ra tu re (?C ) 39 40 41 42 ** C ** * * ** M ax im u m bo dy te m pe ra tu re (?C ) warm wet hot dry warm dry cool dry 1 3 5 7 9 D * ** ** * * Seasonal period Am pl itu de o f b o dy te m pe ra tu re (?C ) Figure 3.3. 24-h mean (panel A), minimum (panel B), maximum (panel C), and amplitude of the nychthemeral rhythm (panel D) of body temperature (mean ? SD, n = 5) over the four periods, namely warm wet (black bar), hot dry (hatched bar), warm dry (white bar) and cool dry (hatched bars). * P < 0.05, ** P < 0.001, Newman-Keuls multiple comparison tests. 91 0 10 20 30 40 37 38 39 40 A Mean air temperature ( ?C) M ea n 24 - ho u r bo dy te m pe ra tu re (?C ) 5 10 15 20 25 0 2 4 6 E Amplitude of air temperature ( ?C) A m pl itu de o f b o dy te m pe ra tu re (?C ) 10 20 30 40 50 35 36 37 38 39 D Maximum air temperature ( ?C) M in im u m bo dy te m pe ra tu re (?C ) 0 10 20 30 35 36 37 38 39 C Minimum air temperature ( ?C) M in im u m bo dy te m pe ra tu re (?C ) 10 20 30 40 50 38 39 40 41 42 B Maximum air temperature ( ?C) M ax im u m bo dy te m pe ra tu re (?C ) 10 11 12 13 14 0 2 4 6 F Photoperiod (h) A m pl itu de o f b o dy te m pe ra tu re (?C ) Figure 3.4. 24-h mean of body temperature (mean, n = 5) plotted against mean 24-h air temperature (panel A, r2 = 0.35, y = 0.02x + 38.0), and the maximum daily body temperature plotted against maximum daily air temperature (panel B, r2 = 0.85, y = 39.7 + 9.2x10-14x8.2) for 313 individual days, when the oryx were living free. Minimum daily body temperature (mean, n = 5) was correlated poorly with minimum daily air temperature (panel C, r2 = 0.13, y2 = 1368 ? 0.003x3) and with maximum daily air temperature (panel D, r2 = 0.15, y2 = 1379 ? 0.0008x3). So too, amplitude of nychthemeral rhythm of body temperature (mean, n = 5) was correlated poorly with the 24-h amplitude of air temperature (panel E, r2 = 0.07) but better with photoperiod (panel F, r2 = 0.31, y = 0.50x - 2.5), although there were two discrete relationships at long photoperiods. The curved arrows indicate the progression of time. 92 One classic measure of adaptive heterothermy is a positive correlation between the amplitude of body temperature rhythm and the amplitude of air temperature rhythm. I found no such correlation (Fig. 3.4E). On the other hand, the amplitude of body temperature rhythm was significantly correlated to photoperiod (Fig. 3.4F). Although the linear correlation between amplitude of body temperature rhythm and photoperiod was significant, the relationship was not straightforward. The amplitude of body temperature rhythm was ~ 2?C during the warm wet period, when photoperiod was ~ 13 h, and increased to ~ 6?C as conditions got progressively warmer and drier, which coincided with an increased photoperiod. However, the amplitude of body temperature rhythm did not return to the expected 2?C as day length shortened and conditions got cooler. Instead, during the warm dry conditions oryx continued to have a low minimum body temperature, which resulted in the amplitude of body temperature rhythm remaining high until conditions became cooler with shorter day lengths. 3.4.3 Activity The nychthemeral rhythm of activity also varied over the three seasonal periods analyzed (Fig. 3.5); because of a limited battery life of the activity loggers I did not obtain measurements in the cool dry period. During the warm wet period, when the amplitude of body temperature rhythm was smallest, the oryx displayed a continuous but biphasic activity pattern with crepuscular peaks (Fig. 3.5A). This pattern shifted to a predominantly nocturnal activity during the hot dry period (Fig. 3.5B), with minimal activity between 09:00 and 18:00. The pattern returned to biphasic activity as conditions got cooler in the warm dry period (Fig. 3.5C), even though the amplitude of body temperature rhythm still was greater during the warm dry period than during the warm wet period (Fig. 3.5C) 93 0:00 6:00 12:00 18:00 36 38 40 42 0 10 20 30 40C Time of day 36 38 40 42 0 10 20 30 40 B Bo dy te m pe ra tu re ( ?? ??C ) Pr o po rt io n o f m ax im u m ac tiv ity (% )36 38 40 42 0 10 20 30 40 A Figure 3.5. Mean ? SD of the nychthemeral rhythm of body temperature (black line) and activity (grey bars) of a representative oryx (Oryx 3) over the warm wet (panel A), hot dry (panel B) and warm dry (panel C) periods. The oryx shifted from a continuous but biphasic activity pattern with crepuscular peaks during the warm periods to nocturnal activity during hot periods. Activity was measured by a data logger, implanted subcutaneously on the left abdominal wall, as activity counts per 5 min and is shown as a percentage of the maximum count for each logger. 94 Mean 24-h activity did not differ across the three periods (F2,8 = 2.2, P = 0.17). Nevertheless, mean 24-h activity of all five oryx was significantly, although weakly, correlated to both maximum daily air temperature (r2 = 0.22, P < 0.0001, Fig. 3.6A), and photoperiod (r2 = 0.18, P < 0.0001, Fig. 3.6B). Mean activity level between 06:00 and 18:00, however, did differ significantly over the three periods (F2,8 = 22.4, P = 0.0005), with mean daytime activity being significantly lower during the hot dry period than during both the warm wet (P = 0.001) and the warm dry (P = 0.0007) periods. The proportion of total daily activity which occurred between 06:00 and 18:00 decreased from 48 ? 5% during warm wet periods to 25 ? 8% during the hot dry period, and returned to 48 ? 11% during the warm dry period (F2,8 = 24.0, P = 0.0004). The proportion of total daily activity occurring between 06:00 and 18:00 was correlated strongly with maximum 24-h air temperature (Fig. 3.6C), and with photoperiod (Fig. 3.6D). Maximum 24 h air temperature accounted for 61% (P < 0.0001) of the variability in diurnal activity, and photoperiod accounted for 47% (P < 0.0001). However, a partial correlation coefficient revealed that the original correlation of activity versus photoperiod was confounded by maximum 24-h air temperature (rxz.y = -0.05, t210 = -0.75, P = 0.45) and that maximum 24-h air temperature accounted for only 27% of the proportion of diurnal activity when the effect of photoperiod was eliminated (rxy.z = -0.52, t210 = -8.89, P < 0.0001). 3.4.4 Microclimate The periods of reduced activity during the hot dry period appeared to coincide with periods of shade-seeking behaviour. The slopes of the correlations between the miniglobe temperature on the collar of two female oryx and miniglobe temperature recorded at a nearby weather station and exposed to direct solar radiation, were significantly less than one (Fig. 3.7, ANCOVA, P < 0.0001). The regression lines intersected the line of identity, implying that oryx selected microclimates cooler than the exposed miniglobe at high environmental heat loads, across all four periods. The intersection of the regression lines of collar miniglobe temperature versus weather station miniglobe temperature with the line of identity, for the two oryx across all four periods, was 26.8 ? 3.8?C (mean ? SD, n = 2). 95 0 5 10 15 A M ea n da ily ac tiv ity (% m ax im u m ) B 25 35 45 0.0 0.5 1.0 C Maximum daily air temperature (?C) Di u rn al ac tiv ity (p ro po rt io n to ta l a ct iv ity ) 11 12 13 14 D Photoperiod (h) Figure 3.6. Mean 24-h activity, averaged for all five free-living oryx, plotted against maximum daily air temperature (panel A, mean, n = 5, r2 = 0.28), and photoperiod (panel B, mean, n = 5, r2 = 0.18), for 224 days, and mean diurnal activity, calculated as activity between 06:00 and 18:00 expressed as a proportion of total 24-h activity, plotted against maximum daily air temperature (panel C, mean, n = 5, r2 = 0.47), and photoperiod (panel D, mean, n = 5, r2 = 0.61). The slopes of regression lines of collar miniglobe temperature versus that of the weather station were significantly different between the periods (F3,8184 = 205.8, P < 0.0001), with the regression during the hot dry period showing the flattest slope and the cool dry period showing the steepest slope. These results imply that the oryx were selecting microclimates cooler than that in direct sunlight more at high environmental heat loads, during the hot dry periods, than they were during the cool dry period. Despite similar ambient conditions, the slope of the regression line was steeper during the warm wet period compared to the warm dry period, implying that the oryx prioritized behavioural thermoregulation during the dry periods. 96 0 20 40 60 warm wet 0 20 40 60 hot dry 0 20 40 60 warm dry Co lla r m in ig lo be te m pe ra tu re ( ?? ?? C) 0 20 40 60 0 20 40 60 cool dry Weather station miniglobe temperature (?C) Figure 3.7. Scatter diagram showing the relationship between collar miniglobe temperatures at the site chosen by a single representative female oryx (Oryx 1) and temperatures recorded by an identical miniglobe at a nearby weather station, which was exposed to the sun, during warm wet (panel A), hot dry (panel B), warm dry (panel C) and cool dry (panel D) periods. Measurements were made hourly. The dashed line is the line of identity and the solid line is the line of linear regression. The slopes of the regression lines were significantly different between the periods (P < 0.0001) with the hot dry period showing the lowest slope, indicating more frequent selection of microclimates cooler than the microclimates under full solar exposure during this period. Warm wet period: y = 0.87x + 3.5, r2 = 0.97, P < 0.0001, intersect 27.2?C. Hot dry period: y = 0.76x + 6.7, r2 = 0.97, P < 0.0001, intersect 28?C. Warm dry period: y = 0.81x + 4.7, r2 = 0.97, P < 0.0001, intersect 24.6?C. Cool dry periods: y = 0.90x +2.0, r2 = 0.97, P < 0.0001, intersect 19.3?C. 97 3.4.5 Selective brain cooling Figure 3.8 shows the typical pattern of brain and carotid blood temperature recorded for a single free-living male oryx (Oryx 2) over four-day epochs during the warm wet period (Fig. 3.8A), and the hot dry period (Fig. 3.8B). The brain temperature of the oryx generally was above that of the carotid blood at low body temperatures (Fig. 3.8A), however, oryx implemented selective brain cooling near the acrophase of the endogenous nychthemeral body temperature rhythm during the hot dry period (Fig. 3.8B). 36 38 40 42 brain carotid A 1 2 3 4 36 38 40 42 brain carotidB Days Te m pe ra tu re ( ?? ?? C) Figure 3.8. Brain (light line) and carotid blood (dark line) temperatures of a single free- living male oryx (Oryx 2) over four-day periods during the warm wet period (A) and the hot dry period (B). 98 Figure 3.9 shows hypothalamic temperature as a function of carotid blood temperature and the frequency distribution of blood temperature in a male oryx (Oryx 3, Fig. 3.9A) and female oryx (Oryx 1, Fig. 3.9B) under initial captive conditions when animals had free access to food and water, and in a single free- living male oryx (Oryx 2) over a month in the warm wet period (Fig. 3.9C) and in the hot dry period (Fig. 3.9D). The dashed line represents the line of identity, points below this line reflect selective brain cooling. The oryx exhibited selective brain cooling frequently under captive conditions, spending 88 ? 0.06% of their time with the hypothalamus cooler than carotid blood. On average, brain temperature was 0.25 ? 0.04?C (mean ? SD, n = 3) cooler than carotid blood temperature throughout the study. The same was not true when Oryx 2 was free-living. He spent 11% of the time during the warm wet period with the hypothalamus cooler than carotid blood, and that frequency increased to 38% of his time during the hot dry period. His brain temperature was on average 0.28 ? 0.25?C higher than carotid blood temperature during the warm wet month, and only 0.03 ? 0.32?C higher during the hot dry month. The threshold for selective brain cooling, defined as the point at which carotid blood and mean hypothalamic temperatures are equal, decreased from 39.6?C in the warm wet month to 39.0?C in the hot dry month. Above the threshold, the average difference between carotid blood and hypothalamic temperature was greater in the hot dry period (0.28 ? 0.24?C) than during the warm wet period (0.16 ? 0.20?C). Yet, maximum magnitude of selective brain cooling was similar between the hot dry period (1.03?C) and the warm wet period (0.93?C) and seldom exceeded 1?C at any time. 99 35 37 39 41 A H yp o th al am ic te m pe ra tu re (?C ) 35 37 39 41 B 35.0.135.2.335.4.535.6.735.8.9.0.136.236.3.436.5.636.736.8.97.0.13 .27.33 .47.537.6.737.8.98.0.13 .2.338.4.538.6.738.8.99.0.139.2.33 .49.53 .69.739.8.940.0.10.24 .3.44 .50.64 .740.8.941.0.11.24 .31.44 .54 .61.74 .81.942.0 0 200 400 35 37 39 41 Fr eq u en cy 35.0.135.2.335.4.535.6.735.8.936.0.1.26.33 .46.53 .66.73 .86.937.0.1.2.337.4.537.6.737.8.938.0.18.2.33 .48.53 .68.73 .88.939.0.19.2.33 .4.539.6.739.8.940.0.110.2.30.44 .5.640.7.840.91.04 .11.2.31.44 .5.64 .71.84 .92.0 0 200 400 35 7 9 35 37 39 41 C H yp o th al am ic te m pe ra tu re (?C ) 35 37 39 41 D 0 300 600 35 37 39 41 Carotid temperature (?C) Fr eq u en cy 0 300 600 35 37 39 41 Carotid temperature (?C) Figure 3.9. Hypothalamic temperature as a function of carotid blood temperature (top panels) and the frequency distribution of blood temperature (bottom panels), in a male oryx (Oryx 3, panel A), and a female oryx (Oryx 1, panel B), when both animals were captive and had ad libitium access to food and water, and a free-living male oryx (Oryx 2), in the warm wet period (May, panel C) and in the hot dry period (June, panel D). All 5-min recordings of carotid blood temperature were sorted into 0.1?C classes and the mean (solid line), minimum (lower grey line) and maximum (upper grey line) hypothalamic temperature calculated for each 0.1?C class of carotid blood temperature. The dashed line represents the line of identity; points below this line reflect selective brain cooling. Lower panels in each set show the absolute frequencies with which each 0.1?C class of blood temperature occurred; 6491 data points were obtained for the captive male oryx (panel A), 4393 data points for the captive female oryx (panel B), 8067 data points for the free-living male oryx in warm wet period (panel C) and 8355 data points in the hot dry period (panel D). 100 3.5 Discussion My study provides the first remote and continuous measurement of body temperature, brain temperature, activity and microclimate selection of free-living Arabian oryx (Oryx leucoryx) in their natural habitat, the deserts of Saudi Arabia. The data allowed me to examine physiological and behavioural responses of this species in the hyper-arid conditions of the Arabian deserts. I could detect these responses because, together with my colleagues, I instrumented free-living oryx with data loggers capable of producing a more complete and continuous record of the thermal status and activity of undisturbed free-living oryx than had been achieved previously. The temperature data loggers, which my colleagues and I implanted in the abdomen, revealed that my oryx displayed not only higher daily maximum body temperatures (41.1 ? 0.3?C vs. 39.7 ? 0.1?C), but also lower daily minimum body temperatures (36.1 ? 0.3?C vs. 36.8 ? 0.2?C), resulting in higher daily amplitudes of nychthemeral rhythm of body temperature (5.0 ? 0.5?C vs. 2.9 ? 0.2?C), during hot dry months than during cool dry months (Fig. 3.2A), as reported previously by Ostrowski et al. (2003). The maximum daily amplitude of the nychthemeral rhythm of body temperature reached 7.7?C for two of the oryx during the hot dry period, the highest ever recorded for a large mammal. There was a positive power relationship between maximum daily body temperature and maximum daily air temperature. Maximum daily air temperature accounted for 85% of the variability in maximum daily body temperature (Fig. 3.4B). These results imply that the oryx indeed did implement heterothermy, but, contrary to the conventional definition of adaptive heterothermy (Mitchell et al., 2002), my oryx had higher mean body temperatures at night (39.0 ? 0.2?C) than in the daytime (38.4 ? 0.2?C). In addition, the classic measure of adaptive heterothermy, a correlation between the amplitude of nychthemeral rhythm of body temperature and the amplitude of ambient temperature, was not evident in my oryx (Fig. 3.4E), implying that the heterothermy was induced by something other than the prevailing air temperature. I believe that the factor responsible for the appearance of heterothermy in my oryx was the lack of water in their environment, leading both to a reduction in water 101 intake and a decline in food quality and availability (Spalton, 1999). During the transitional warm wet and warm dry periods, when ambient temperatures were similar, the oryx displayed greater nychthemeral heterothermy (3.6 ? 0.6?C vs. 2.3 ? 0.3?C), mostly as a result of lower daily minimum body temperatures (37.8 ? 0.2?C vs. 36.5 ? 0.6?C) during the warm dry period than during the warm wet period (Fig. 3.2C). I believe that these low daily minimum body temperatures, which occurred in the early mornings, resulted from water and food limitations in the desert environment. In addition to heterothermy, selective brain cooling is considered to be an important adaptation to environmental heat stress. The conventional view of selective brain cooling would predict that the magnitude of brain cooling would have increased during hot dry months, when carotid blood temperature exceeded 41?C. Although the oryx used selective brain cooling more frequently during the hot dry period, and the threshold at which selective brain cooling was implemented was lower then, the magnitude of selective brain cooling actually was reduced in the hot dry month, compared to the warm wet month (Fig. 3.9), and brain temperature reached 41.2?C. It is possible that 1?C is the maximal capacity for selective brain cooling in Arabian oryx, but even this small magnitude of selective brain cooling would have provided both metabolic and water conservation advantages. My oryx appeared to have further conserved water by transferring daily activity during the hot dry period to the night, when metabolic heat could be dissipated non-evaporatively. Whereas my oryx showed continuous activity with crepuscular peaks during the warm periods, whether wet or dry, they shifted to nocturnal activity during the hot periods, without reducing total 24-h activity (Fig. 3.5). However, because the decline in forage quality during the dry season, maintaining total 24-h activity may not have been an adequate compensatory response for maintaining energy intake. Oryx have been observed to seek shade during the heat of the day and I detected the selection of cooler microclimates, without observers present, by using miniature globe thermometers attached to collars, a technique I 102 developed for recording microclimate selection (Appendix 1, Hetem et al., 2007). As expected oryx selected cooler microclimates more frequently during the hot dry period, and at times selected microclimates where the miniglobe temperature was 12?C lower than that in the sun (Fig. 3.7B). Selecting cooler microclimates would have further reduced evaporative heat loss, and therefore conserved body water. Although there are obvious advantages to obtaining data from free-living, undisturbed animals, there are disadvantages in that some variables cannot yet be measured. My proposals concerning the crucial role of water in determining body temperature profiles and the magnitude of selective brain cooling in the oryx would have been strengthened if I could have measured water loss, metabolic rate, or even osmolarity by biotelemetry. Also, because it is well known that the presence of humans disrupts normal thermoregulation in wild animals (Recarte et al., 1998), I stayed away from them, and so could not draw any conclusions on habitat selection, migration movements or change in body condition over the experimental period. Advancement in GPS technology combined with GIS now makes some of these measurements feasible. Hopefully it will be possible, in the future, to expand the suite of relevant thermoregulatory variables that can be measured by biotelemetry and remote observation. An additional limitation of this study was the small sample size with respect to the selective brain cooling data. Nevertheless, I reported the data because it has not previously been collected in such an arid-adapted species. 3.5.1 Heterothermy Despite the limitations, I believe that my study expands on and refines the one previous study that has reported body temperature in free-living Arabian oryx (Ostrowski et al., 2003). In that study, body temperature of six oryx was measured with temperature-sensitive radiotelemeters, which required the observer to be within 800 m of the individual animals, a difficult task in the vast terrain, especially without disturbing the animals. Although the study provided valuable insight into the adaptations of the Arabian oryx, and was the first to identify 103 heterothermy in a free-living antelope, the data appear to have been confined to less than one day of recordings per animal per season during the day and only about one hour of recordings per animal per season during the night, as compared to my continuous body temperature measurements, throughout the day and night, for 313 consecutive days. Not only have I recorded body temperature continuously, but I also recorded activity continuously, and, in some oryx, microclimate selection. Perhaps because conditions were drier during my study (31 mm vs. 74 ? 46 mm), water may have been less of a factor in the study of Ostrowski et al. (2003), and they attributed daily heat storage to the amplitude of air temperature variation. Nevertheless, my oryx displayed a larger daily amplitude of the nychthemeral rhythm of body temperature than Ostrowski?s oryx during the summer (5.0 ? 0.5?C vs. 4.1 ? 1.7?C), despite the lower maximum air temperature prevalent during my hot dry period (40.4 ? 1.3?C vs. 43.5 ? 1.7?C). I propose that the greater degree of heterothermy in my oryx was likely the result of the drier conditions prevalent during my study and possibly the larger data set sampling more days. The latent heat of evaporation of water is sufficiently high, and the water vapour pressure of desert air sufficiently low, for desert mammals to be able to dissipate resting metabolic heat and environmental heat (even in direct solar radiation) by evaporation and so avoid hyperthermia. However, the evaporation that is necessary depletes body water at a rate likely to induce potentially-lethal hyperosmolarity, unless the mammal has ready access to adequate dietary water. At my experimental site, the only drinking water to which the oryx had access was in the ephemeral pools that formed when the 31 mm of rain fell between March and May; at all other times they had to rely on the pre-formed water in the sparse desert vegetation, and so likely could not afford prolific evaporative cooling. Consequently, the oryx employed autonomic and behavioural processes which reduced metabolic and environmental heat loads, and transferred responsibility for heat dissipation to non-evaporative avenues. They did so without reducing foraging time, important because reducing foraging would have 104 compromised energy homeostasis, especially when forage quality deteriorated during the hot dry period. One of the processes for transferring heat loss to non-evaporative avenues is the employment of heterothermy. The capacity to store heat holds whether the heterothermy is the controlled thermoregulatory event that constitutes adaptive heterothermy, or results from failure of homeothermy (Mitchell et al., 2002). But failure of homeothermy places the animal at risk of damaging hyperthermia in the absence of the suppression of the temperature nadir (a hallmark of adaptive heterothermy). One cause of failure of homeothermy is dehydration itself. Several ungulate species reduce the rate of evaporative water loss as they become dehydrated (Taylor, 1970b; Maloiy, 1973b; Finch and Robertshaw, 1979; Baker, 1989; Nijland and Baker, 1992; Silanikove, 1994; Jessen et al., 1998), which results in an increased body temperature during heat exposure (Taylor, 1969a; Taylor, 1970a; Taylor, 1970b; Finch and Robertshaw, 1979; Jessen et al., 1998; Alamer, 2006). Whether that elevation in body temperature constitutes dehydration-induced hyperthermia or adaptive heterothermy is not self evident. Nevertheless, whether heterothermy is dehydration induced hyperthermia or adaptive heterothermy, one expects the heterothermy to increase with increasing dehydration. I had no way of quantifying dehydration in my free-living oryx, but previous studies have suggested that free-living Arabian oryx have a low rate of water influx of 1310 ml H2O.day-1 during summer (Williams et al., 2001; Ostrowski et al., 2002), and display higher haematocrit, plasma protein concentration and plasma osmolality during summer than during winter (Ostrowski et al., 2003). My animals were exposed to harsh environments without access to free-standing water for nine months. It therefore seems plausible that the high body temperatures of my oryx, displayed during the hot dry period, were the combined result of dehydration and extreme environmental temperatures. What distinguishes adaptive heterothermy from dehydration induced suppression of evaporative heat loss, though, is that in adaptive heterothermy the amplitude of 105 body temperature rhythm increases without the mean 24-h body temperature necessarily increasing, because minimum body temperature is depressed (Mitchell et al., 2002). Under captive conditions of water deprivation, the dromedary camel (Camelus dromedarius, Schmidt-Nielsen et al., 1957; Schroter et al., 1987), hartebeest (Alcelaphus buselaphus, Harthoorn et al., 1970; Maloiy and Hopcraft, 1971), Thomson?s gazelle (Gazella thomsonii) and zebu steer (Bos primigenius indicus, Taylor, 1970a) have been reported to display an increased body temperature amplitude with an associated lower body temperature in the morning. Taylor found that heterothermy was not dependent on hydration state for captive eland (Taylor and Lyman, 1967; Taylor, 1969a). However, unlike the seven degree increase in body temperature, which was observed on a single occasion, and is so often referred to, the body temperature of the eland (Tragelaphus oryx) usually only increased by three of four degrees during the 12-h day when the animals had water available and were exposed to 40?C (Taylor, 1969a). In fact, the single rectal temperature measurement of 33.9?C, on which the seven degree amplitude of body temperature rhythm is based, may have been a measurement error. In addition, these rectal temperatures were based on measurements obtained from immature, captive eland and the findings may have been artefacts of studying animals in captive conditions in which they were prevented from employing behavioural thermoregulation (Finch, 1972; Fuller et al., 1999; Mitchell et al., 2002). Subsequent studies on free-living eland showed a much lower amplitude of nychthemeral rhythm (Bligh and Harthoorn, 1965; Harthoorn et al., 1970; Finch, 1972; Finch and Robertshaw, 1979; Fuller et al., 1999; Fuller et al., 2004). More recently, Ostrowski and Williams (2006) argued that free-living Arabian sand gazelle (Gazella subgutturosa marica) displayed heterothermy at times when the animals were not dehydrated. However, this assumption was based on the finding that plasma osmolality was not different between October (319.2 ? 1.9?mOsm) and September (320.2 ? 2.3?mOsm), yet their estimates of heterothermy were made in winter (January and February) and summer (June and July). Therefore, to date, there is no substantiated evidence that heterothermy exists in free-living 106 animals with free access to water; several studies have shown a narrow range of body temperature in normally hydrated desert-adapted ungulates (Bligh and Harthoorn, 1965; Harthoorn et al., 1970; Hofmeyr and Louw, 1987; Jessen et al., 1994; Mitchell et al., 1997; Fuller et al., 1999; Maloney et al., 2002; Mitchell et al., 2002; Fuller et al., 2005). When given the opportunity, oryx increase their water consumption and consume an average of 5 l.day-1 during the summer months (Stanley Price, 1989) and preferentially select areas where new rain has fallen (Corp et al., 1998). Under captive conditions, with food and water available ad libitum, my single male oryx maintained an amplitude of nychthemeral rhythm of body temperature of 2.6 ? 0.5?C throughout the year (Fig. 3.1, upper right panel) despite being exposed to environmental heat loads very similar to those of the free-living animals. Previous studies have proposed that the morning drop in body temperature is the result of vasodilation, which mixes cool peripheral blood with that of the core (Schmidt-Nielsen et al., 1957; Zervanos and Hadley, 1973; Brown and Dawson, 1977; Fuller et al., 1999; Maloney et al., 2004). Such a drop in body temperature may allow an animal to anticipate hot ambient conditions (Maloney et al., 2004) and thus pre-emptively permit additional storage of heat. In my free-ranging oryx, minimum body temperatures were low not only during the hot dry period, but also throughout the other dry periods, even though ambient temperatures were not as high during the warm dry and cool dry periods (Table 3.1). Since dehydration has been shown to decrease metabolic rates of various species including Bedouin goats (Brosh et al., 1986; Dmi'el, 1986), Holstein cows (Seif et al., 1973), camels (Schmidt-Nielsen et al., 1967; Schroter et al., 1987), eland (Taylor, 1969b) and cats (Doris and Baker, 1981), I propose that in my oryx the low body temperature in the morning may reflect these low metabolic rates associated with dehydration. The low morning body temperatures may also reflect the so-called ?nocturnal hypometabolism?, observed during winter in both red deer (Cervus elaphus, Arnold et al., 2004) and the Przewalski horse (Equus ferus przewalskii, Arnold et al., 2006). Such seasonal fluctuations in metabolic rates occur not only in northern 107 ruminants, which display a winter nadir (Nilssen et al., 1984; Parker and Robbins, 1985; Mesteig et al., 2000; Arnold et al., 2004), but also desert ungulates, which display a summer nadir in metabolic rate (Nagy and Knight, 1994; Williams et al., 2001; Ostrowski et al., 2006a; Ostrowski et al., 2006b). The low metabolic rates associated with low quality and quantity of food could result in a lowered body temperature, as seen in food deprived rodents (Sakurada et al., 2000; Bae et al., 2003), sheep (Piccione et al., 2002) and desert goats (Choshniak et al., 1995; Ahmed and El Kheir, 2004). During drought some grasses provide an adequate source of crude protein and water for the oryx, but these levels are predicted to be close to minimum maintenance requirements (Spalton, 1999).It therefore seems plausible that limited food and water availability during the dry seasons in a desert environment would combine to reduce the metabolic rate of desert ungulates (Brosh et al., 1986; Merkt and Taylor, 1994; Williams et al., 2001; Ahmed and El Kheir, 2004; Ostrowski et al., 2006a; Ostrowski et al., 2006b), thereby reducing endogenous heat production and energy expenditure. Similarly, Dawson et al. (2007) suggested that a low morning body temperature in kangaroos was probably more related to energy balance than heat storage in hot conditions. So, though as Ostrowski et al. (2003) first demonstrated, Arabian oryx do exhibit heterothermy in the hot dry period, and that heterothermy potentially saves water, I am not convinced that it is necessarily a consequence of ambient heat stress. 3.5.2 Cathemerality I am more convinced that the second water-saving mechanism that my oryx employed, cathemerality, was a consequence of ambient heat stress. The term ?cathemerality?, coined originally for primates, described deliberate transfer of activity between the light and dark phases of the 24-h cycle in response to prevailing ecological conditions (Tattersall, 1987; Tattersall, 2006; Tattersall 2008). Chronoecological factors, such as temperature, light, competition for food resources and predation, appear to promote cathemerality by overriding endogenous clocks (Curtis and Rasmussen, 2006); the distribution of activity between day and night in my oryx was correlated with both maximum daily air temperature (Fig. 3.6C) and photoperiod (Fig. 3.6D). My oryx shifted from a 108 crepuscular activity pattern during warm wet periods to a nocturnal activity pattern during the hot dry period. Previous observations on the activity patterns of Arabian oryx within the Mahazat as-Sayd Protected Area also revealed that oryx were less active diurnally on warm (maximum air temperature >35?C) than on cool days (maximum air temperature <30?C, Seddon and Ismael, 2002), and high ambient temperatures attenuated the diurnal activity of other tropical ungulates (Leuthold and Leuthold, 1978; Belovsky and Slade, 1986; Klein and Fairall, 1986; Owen-Smith, 1998). However, any conclusions, based on diurnal observations only, that environmental heat load reduces foraging are likely to be spurious. Identification of cathemerality requires measurement of both diurnal and nocturnal activity (Curtis and Rasmussen, 2006), which I achieved objectively by using activity loggers. Activity loggers provided a long-term, remote and unbiased assessment of activity, over successive 24-h cycles, not otherwise feasible under field conditions. Because I used activity loggers, I discovered that mean 24-h activity was not strongly correlated to either photoperiod or maximum daily air temperature, nor was there any significant difference across the three seasonal periods analyzed. Similarly, neither bushbuck (Tragelaphus scriptus, Wronski et al., 2006) nor bighorn sheep (Ovis canadensis mexicana, Alderman et al., 1989) showed differences in mean 24-h activity across seasons, when activity was observed during both nocturnal and diurnal hours. My oryx, exposed to diurnal ambient temperatures often exceeding 35?C, compensated for a reduced diurnal activity during hot periods by increasing nocturnal activity so that the magnitude of mean 24-h activity did not change. Cathemerality without loss of overall activity allowed foraging time to be maintained. Similarly, other large herbivores, including ungulates (Zervanos and Hadley, 1973; Belovsky and Jordan, 1978; Grenot, 1992; Hayes and Krausman, 1993; Berger et al., 1999; Dussault et al., 2004; Maloney et al., 2005b), increased nocturnal activity in the face of high diurnal heat loads. Such an increase in nocturnal activity is believed to be a compensatory response, as grazing activities during the heat of the day would be limited as a result of shade-seeking behaviour, 109 in response to high ambient temperatures. Arabian oryx in Oman increased shade- seeking behaviour at ambient temperatures above 27?C (Stanley Price, 1989), a threshold remarkably similar the average air temperature above which my oryx selected cooler microclimates across all periods. The selection of cooler microclimates during the heat of the day has been observed in a variety of ungulates, particularly those inhabiting desert environments, including the desert mule deer (Odocoileus hemionus, Sargeant et al., 1994; Tull et al., 2001), desert bighorn sheep (McCutchen, 1981; Hansen, 1982), eland (Taylor and Lyman, 1967), dik-dik (Rhynchotragus kirkii, Kamau and Maloiy, 1985), collared peccary (Tayassu tajacu, Zervanos and Hadley, 1973) and dorcas gazelle (Gazella dorcas, Wilson, 1989). 3.5.3 Selective brain cooling The third process that I believe Arabian oryx used to facilitate homeostasis at high environmental heat loads was selective brain cooling. Arid-zone mammals possessing a carotid rete may employ selective brain cooling to attenuate thermal drive by reducing hypothalamic temperature, which reduces evaporative heat loss and ultimately conserves water by transferring cooling to non-evaporative means (Kuhnen, 1997; Jessen, 1998; Jessen, 2001; Mitchell et al., 2002). Animals capable of selective brain cooling increase the magnitude and/or frequency of brain cooling during conditions of water deprivation. Both laboratory-housed sheep (Fuller et al., 2007) and Bedouin goats (Jessen et al., 1997; Jessen et al., 1998) did so without changing the threshold body temperature at which selective brain cooling was implemented. Anomalously, captive camels showed no obvious difference in selective brain cooling between hydrated and dehydrated states at rest (Schroter et al., 1989). My single free-living male oryx did not change the magnitude of selective brain cooling during the hot dry month, compared to the warm wet month, and instead appeared to decrease the threshold, therefore increasing the range of body temperatures over which selective brain cooling was implemented. The differences between my findings and those in the sheep and goats may be the result of differences in the pattern of brain cooling between laboratory and wild animals. Laboratory sheep (Fuller et al., 2007) and goats 110 (Jessen et al., 1998) display the greatest degree of selective brain cooling at night, whereas my oryx, like free-living eland (Fuller et al., 1999), gemsbok (Oryx gazella, Maloney et al., 2002), black wildebeest (Connochaetes gnou, Jessen et al., 1994) and springbok (Antidorcas marsupialis, Mitchell et al., 1997), displayed the greatest degree of selective brain cooling when the animal was at rest and body temperature was rising in late afternoon (Fig. 3.8). Although my oryx employed selective brain cooling more frequently than has been reported previously for other artiodactyls, probably because its blood temperature was higher than that measured in other free-living artiodactyls, I could not conclude that selective brain cooling was enhanced in the arid-adapted oryx, since the magnitude of the difference between carotid blood and hypothalamic temperature was similar to that observed in other free-living artiodactyls (Jessen et al., 1994; Jessen and Kuhnen, 1996; Mitchell et al., 1997; Fuller et al., 1999; Maloney et al., 2002). My conclusions about selective brain cooling in the arid-adapted oryx, although noteworthy, must remain tentative, however, as I only obtained a complete data set from a single free-living male oryx. Since non-thermal factors affecting sympathetic tone may override the thermal regulation of brain temperature (Kuhnen and Jessen, 1994), particularly in free-living animals (Jessen et al., 1994; Maloney et al., 2002; Caputa, 2004), a high vigilance and sympathetic tone may have acted to decrease the magnitude of selective brain cooling in this individual. Although under captive conditions the magnitude and pattern of selective brain cooling was similar between male and female oryx (Fig. 3.8 upper panels), future studies aimed at assessing selective brain cooling should focus on free-living female individuals. Even though the magnitude of selective brain cooling in my single male oryx was much less than the anomalous 2.7?C originally reported for the exercising captive Thomson?s gazelle (Taylor and Lyman, 1972), it is believed that a reduction in brain temperature of even 0.5?C is enough to substantially reduce both the respiratory water loss (Kuhnen, 1997) and the metabolic cost of thermoregulation (Kuhnen and Jessen, 1991) in goats. 111 3.5.4 Conclusion The prosperity, and indeed the survival, of desert mammals depends on them managing successfully the often-competing demands of their homeostatic mechanisms. That management is particularly challenging for species too large to escape below the desert surface. I have shown how the Arabian oryx, a large artiodactyl native to one of the world?s hottest deserts, is likely to manage the competing demands of thermoregulatory homeostasis, body fluid homeostasis, and metabolic energy homeostasis, and have demonstrated, I believe, that Arabian oryx give the highest priority, among those three, to body fluid homeostasis. By studying an arid-adapted artiodactyl, inhabiting one of the hottest deserts in the world, I have established the kind of physiological mechanisms required if artiodactyls are to adapt phenotypically to climate change. I have shown how the oryx uses four different physiological strategies, namely heterothermy, cathemerality, selective brain cooling and shade-seeking behaviour, to counter dehydration, hyperthermia and starvation. Yet, their current dependence on behavioural thermoregulation, combined with the loss of genetic variability as a result of a historical population bottleneck (Marshall et al., 1999), may limit their capacity to adapt to future aridification and habitat transformation. The Arabian oryx already may be at the edge of its physiological limit and, with climate change predicted to increase summer temperatures in the southwest part of the Arabian Peninsula (Gitay et al., 1998), its thermoregulatory competence may not be sufficient to ensure long-term survival. 112 3.6 Acknowledgements I thank the National Commission for Wildlife Conservation and Development (NCWCD), Riyadh, Saudi Arabia, in particular the director His Royal Highness Prince Saud Al Faisal, the current secretary-general, His Highness prince Bander Bin Saud, and the secretary-general at the time the study was conducted, Professor AH Abuzinada, for supporting the research. From the National Wildlife Research Center (NWRC), I am grateful to Dr Saud Anagariyah for his support in capturing the oryx and the current director, Ahmad Al Bouq. In addition, I thank the Mahazat as-Sayd Protected Area rangers for monitoring the animals and the mammal keepers at NWRC for their help with animal handling and assistance during surgery. 113 ___________________________________________________________ CHAPTER 4 ___________________________________________________________ 4 Does size matter? Comparison of body temperature and activity of free-living Arabian oryx (Oryx leucoryx) and the smaller Arabian sand gazelle (Gazella subgutturosa marica) in the Saudi desert Data and ideas presented in this chapter have been written up as a scientific paper and will be submitted to Journal of Comparative Physiology B. Hetem R.S., Strauss W.M., Shobrak M., Fick L.G., Maloney S.K., Meyer L.C.R., Fuller A. and Mitchell D. Does size matter? Comparison of body temperature and activity of free-living Arabian oryx (Oryx leucoryx) and the smaller Arabian sand gazelle (Gazella subgutturosa marica) in the Saudi desert. 114 4.1 Abstract Heterothermy, a variability in body temperature, is widely viewed as a key adaptation of arid-adapted ungulates. However, ungulates with a small body mass, i.e. a relatively large surface area-to-volume ratio and a small thermal inertia, are proposed to be less likely to employ adaptive heterothermy than are larger ungulates. With the help of my colleagues, I measured body temperature and activity patterns, using implanted data loggers, in five free-ranging Arabian oryx (Oryx leucoryx, ? 70 kg) and six smaller Arabian sand gazelle (Gazella subgutturosa marica, ? 15 kg) inhabiting the same Arabian desert environment, at the same time. Compared to the oryx, the sand gazelle displayed higher (0.7?C) mean daily body temperature (F1,6 = 47.3, P = 0.0005), higher minimum daily body temperature (F1,6 = 42.6, P = 0.0006) and higher maximum daily body temperature (F1,6 = 11.0, P = 0.02). Despite these differences, both species responded similarly to changes in environmental conditions. As predicted for adaptive heterothermy, maximum daily body temperatures increased (F1,6 = 84.0, P < 0.0001), minimum daily body temperature decreased (F1,6 = 92.2, P < 0.0001), and daily body temperature amplitude increased (F1,6 = 97.6, P < 0.0001) as conditions got progressively hotter and drier. There were no species-specific differences in activity levels. Both gazelle and oryx showed continuous activity with crepuscular peaks during the warm wet period but shifted to nocturnal activity during the hot dry period. Activity was attenuated during the heat of the day at times when both species selected cool microclimates. Therefore, the two species of Arabian artiodactyls employ heterothermy, cathemerality and shade- seeking very similarly to survive the extreme, arid conditions of Arabian deserts, despite their difference in size. 4.2 Introduction Heterothermy reduces demands on both energy and water supply otherwise imposed by maintaining homeothermy. Theoretically, in hot arid environments, heterothermy allows animals to store heat during the day and to dissipate that heat 115 later non-evaporatively, so reducing evaporative cooling, and therefore waterloss. The potential water saving depends only on the degree to which the average temperature of the body tissue increases and on the animal?s thermal inertia (and therefore body mass). Heterothermy is proposed to be particularly effective as a water saving mechanism for large mammals, since, in addition to their larger thermal inertia, within a given time, larger mammals would gain relatively less heat from the hot environment (McNab, 1983; Phillips and Heath, 1995) and generate less metabolic heat per gram of tissue than would smaller mammals (Taylor, 1970a; Mitchell et al., 2002). The phenomenon of adaptive heterothermy was first demonstrated by Schmidt- Nielsen and his colleagues in camels and they also confirmed the logical prediction that the degree of heterothermy would increase in animals deprived of drinking water (Schmidt-Nielsen et al., 1957). Subsequently, the ability to employ adaptive heterothermy was described in several species of African antelope, but without the heterothermy always being accentuated by water deprivation (Taylor and Lyman, 1967; Taylor, 1969a). All the studies demonstrating adaptive heterothermy in antelope, however, were conducted in highly artificial circumstances, on individual or small groups of animals in climatic chambers or other confined spaces. Once biotelemetry allowed measurement of body temperature in free-living animals, it became apparent that, at the same ambient temperature at which confined mammals demonstrated adaptive heterothermy, five species of free-living African mammals did not do so (Jessen et al., 1994; Mitchell et al., 1997; Fuller et al., 1999; Fuller et al., 2000; Maloney et al., 2002; Fuller et al., 2005). A possible explanation for the discrepancy between confined and free-living mammals was that the low minimum body temperatures measured in confined mammals were artefacts of confinement, in animals deprived of their full suite of behavioural thermoregulation (Fuller et al., 1999). The African savannas are not the hottest and driest habitats occupied by antelope, and it remains possible that, in sufficiently stressful conditions, even free-living antelope would demonstrate adaptive heterothermy. Indeed, my colleagues and I 116 have demonstrated that Arabian oryx do display heterothermy (chapter 3). Yet, deserts provide a challenging habitat for not only large but also medium-sized antelope, since these antelope are too large to escape the heat by burrowing. Despite such challenges, the Arabian Peninsula supports seven species within the Bovidae family, more than half of which are gazelle (Groves, 1988). Using biotelemetry, Ostrowski and his colleagues, working in the Saudi Arabian desert, recently demonstrated heterothermy in free-living Arabian oryx (~ 70 kg, Ostrowski et al., 2003) and the much smaller Arabian sand gazelle (~ 15 kg, Ostrowski and Williams, 2006). They reported a smaller amplitude of nychthemeral rhythm of body temperature in the sand gazelle, which together with a lower thermal inertia, would result in a much lower heat storage, and therefore lower water saving. One cannot draw that conclusion, however, because the sand gazelle were studied in an environment cooler and wetter than that in which the oryx were studied. Also, there clearly were problems with the body temperature measurements in the oryx, for which radiotelemetry was used. The difficulties of being close enough to the animals to pick up signal, but not so close as to disturb the animals (Ostrowski et al., 2003), allowed only sporadic measurements of body temperature, especially at night, during which measurements were attained for less than 1% of the study time. The problems were evident in an unexplained change in body temperature of at least 0.5?C between 19:00 and 19:30 and 1?C between 05:30 and 06:00, making estimates of the amplitude of the oryx body temperature rhythm unreliable. Although the smaller gazelles may have the advantage of smaller resource requirements and greater access to refuge sites than do the larger oryx (Ostrowski and Williams, 2006), they are likely to be disadvantaged by a high mass-specific metabolic rate, high water turnover and less capacity to store heat (Bartholomew, 1964; Taylor, 1970a). With the help of my colleagues, I therefore set out to compare the body temperature and refuge selection of the sand gazelle and oryx living free in the same desert environment, at the same time. To contribute to the open question of whether time spend foraging increases (Peters, 1983; Schmidt- Nielsen, 1984; Owen-Smith, 1988; Owen-Smith, 1992; du Toit and Yetman, 2005) 117 or decreases (Belovsky and Slade, 1986; Mysterud, 1998; P?rez-Barber?a and Gordon, 1999) with increasing body mass, I also measure activity continuously. Ostrowski and colleagues reported activity only as observed behaviour in the oryx (Ostrowski et al., 2003) and not at all in the sand gazelle. I used data loggers to make all of my measurements so that the animals were free from the disturbance imposed by human observers. 4.3 Materials and methods 4.3.1 Animals and habitat The experiment took place between March and July 2006 within the 2200?km2 Mahazat as-Sayd Protected Area (28?15? N, 41?40? E) in west-central Saudi Arabia. This open steppe desert is the historical and current habitat for both the Arabian sand gazelle (Gazella subgutturosa marica) and the Arabian oryx (Oryx leucoryx). Six captive-bred male Arabian sand gazelle were transported from the King Khaled Wildlife Research Center, Thumamah, Saudi Arabia (25?20? N, 45?35? E) to Mahazat as-Sayd Protected Area in mid-March 2006. During the same period, five adult wild-born Arabian oryx were captured, under veterinary supervision, from the Mahazat as-Sayd Protected Area. The oryx were habituated in pens for two weeks to reduce potential peri-operative stress. The procedures were approved by the Animal Ethics Screening Committee of the University of the Witwatersrand (protocol no. 2005/87/5 and 2005/88/4). 4.3.2 Surgery Sand gazelle were hand caught and placed in individual crates. Anaesthesia was induced with 8% isoflurane (Aerrane, Astra Zeneca, Johannesburg, South Africa). The oryx were darted and anaesthetized in the holding pens with etorphine hydrochloride (2.5 mg intramuscularly (I.M.), M99, C-Vet, Leyland, UK), which later was reversed with diprenorphine hydrochloride (7.5 mg intravenously (I.V.), M5050, C-Vet, Leyland, UK), once an adequate plane of inhalation anaesthesia had been established. 118 Once recumbent, animals were transported to a temporary sterile operating theatre within 200 m of the pens, where they were placed in sternal recumbency, supported by sandbags, with their heads elevated. Anaesthesia was maintained with 2-6% isoflurane, administered in 100% oxygen. Respiratory rate, heart rate, arterial oxygen saturation (pulse oximeters, Nonin 9847V, Nonin Medical, North Plymouth, USA) and rectal temperature (thermocouple thermometer, BAT-12, Physitemp, Clifton, USA) were monitored throughout the surgery. For biotelemetry of body temperature and activity, miniature data loggers were implanted. The data loggers were covered in an inert wax (Sasol, Johannesburg, South Africa) and dry-sterilized in formaldehyde vapour before implantation. Incision sites were shaved and sterilized with povidone iodine antiseptic (Vetedine, Vetoquinol, Lure, France). A thermometric logger (see below) then was inserted, via an incision in the paralumber fossa, into the abdominal cavity. The muscle layer was sutured closed and an activity logger (Actical, Mini-Mitter Corporation, Bend, OR, USA) was tethered subcutaneously before the skin was sutured closed. The activity logger recorded at 5-min intervals, had dimensions of ~ 40 x 40 x 15 mm, and weighed ~ 40 g. Wounds were treated with a topical antiseptic spray (Necrospray, Centaur Labs, Johannesburg, South Africa). The animals received a long-acting antibiotic (4.5 mg.kg-1 I.M., penicillin, Norocillin La, Norbrook Laboratories Ltd., Newry, Northern Ireland), a non-steroidal anti-inflammatory analgesic (10 mg.kg-1 I.M., phenylbutazone, Dexaphenylarthrite injectable solution, Vetoquinol Veterinary Pharmaceuticals, Cedex, France), a long-acting parasiticide (0.02 ml.kg-1 subcutaneously (S.C.), Ivermectin, Noromectin, Norbrook Laboratories Ltd., Newry, Northern Ireland) and a multivitamin (0.1 ml.kg-1 I.M., Multivit injectable solution, Univet Ltd., Ireland). After surgery but before inhalation anaesthesia was terminated, a neck collar was fitted to each oryx (MOD-500 Telonics, Inc. Mesa, AZ, USA) and sand gazelle (African Wildlife Tracking, Pretoria, South Africa). In addition to a tracking radio 119 transmitter, the collar supported a miniature black globe thermometer (?miniglobe?), to allow for the dynamic ambulatory measurement of the microclimate that each animal chose to occupy. This technique has been validated previously on other ungulate species (Appendix 1, Hetem et al., 2007). Miniglobe temperature was measured by a small temperature-sensitive data logger (see below) inserted into the centre of a matt-black hollow copper sphere (30 mm diameter, Press Spinning & Stamping Co., Cape Town, South Africa). The globe was attached to the collar by a 10 mm diameter polyvinyl chloride rod. A weight on the ventral side of the collar ensured that the miniglobe remained over the dorsum of the neck and could not be shaded by the animal?s body. Following surgery, the animals were transported back to their pens, where they became ambulatory within ~ 10 min of termination of inhalation anaesthesia. After two weeks of recovery, and following veterinary inspection, the oryx and sand gazelle were released into adjacent, but separate, 2 km2 fenced enclosures with natural forage and water available ad libitum. Ten days later, the gate of the oryx?s enclosure was opened, and the oryx were allowed to range freely within the 2200?km2 Mahazat as-Sayd Protected Area where there was no permanent natural water and no artificial water sources. To improve the chance of recapturing the sand gazelle, they remained in their 2 km2 enclosure for the duration of the study. The artificial water source evaporated within one month, after which it was not replaced, to replicate the dry environment to which the animals would be subjected naturally, and which was occupied at that time by the oryx. Thus, the only drinking water to which the sand gazelle and oryx had access was in the ephemeral pools that formed when 31 mm of rain fell between March and May; at all other times both species had access only to the pre-formed water in the sparse desert vegetation, and to metabolic water. Unfortunately all six sand gazelle died in August 2006, five months after surgery, during an exceptionally hot and dry spell. I retrieved four body temperature loggers, four miniglobe temperature sensors and six activity loggers. I was unable to determine the cause of death, but saw no evidence of fever in the temperature 120 records. The oryx were recaptured and transported to the holding pens in April 2007. The animals were anaesthetised once again, and the data loggers were removed under a surgical procedure similar to that used for the original implantation. The animals? wounds had healed and there were no signs of infection from the initial surgery. Most of the body temperature loggers were found in the pelvic canal and were not encapsulated in adhesive tissue. Only two miniglobe sensors still were attached to the collars on the oryx. After removal of all monitoring equipment, the oryx were re-released into the Mahazat as-Sayd Protected Area. 4.3.3 Temperature measurements The miniature thermometric data loggers (StowAway XTI, Onset Computer, Pocasset, Massachusetts, USA), used to measure body temperature, had outside dimensions of ~ 50 ? 45 ? 20 mm and a mass of ~ 40 g when covered with wax. These loggers had a resolution of 0.04?C and measurement range from + 34 to + 46?C. The scan interval of the body temperature loggers was set at 20-min. Miniglobe temperatures were recorded every hour, with a smaller thermometric data logger (iButton DS1922T, Maxim, Dallas Semiconductor, Texas, USA), which weighed ~ 10 g. These loggers had a resolution of 0.5?C and a measurement range from 0 to 125?C. All temperature sensors and loggers were calibrated against a high-accuracy thermometer (Quat 100, Heraeus, Hanau, Germany) in an insulated water bath. After calibration, the loggers and their sensors measured body temperature to an accuracy of better than 0.05?C and miniglobe temperature to better than 0.5?C. 4.3.4 Meteorological data measurements I collected climatic data from a portable weather station erected near the Mahazat as-Sayd Protected Area, at the Saja/Umm ar-Rimth Protected Area (23?22? N, 42?45? E). I recorded wind speed (m.s-1), solar radiation (W.m-2), dry-bulb air temperature (?C) and relative humidity (%). I also recorded miniature (30 mm 121 diameter) black globe temperature (?C) and rainfall on site at the Mahazat as-Sayd Protected Area for the duration of the study period. 4.3.5 Data analysis I analysed data within two periods, April-May and June-July, which I designated ?warm wet? and ?hot dry? on the basis of the prevailing ambient temperature and rainfall. I performed an unpaired Student?s t-test to compare mean daily environmental conditions between the two periods. For successive 24-h periods, I calculated mean, minimum, maximum and amplitude of nychthemeral rhythm of body temperature. I averaged the body temperature parameters for each animal over each two-month period and performed a two-way repeated-measures analysis of variance (ANOVA) to test for differences in the temperature and activity profiles between the species and the two-month periods. Newman-Keuls multiple comparison tests were used to identify sources of differences. To further assess the influence of the environment on body temperature patterns, I correlated the various parameters of the body temperature profile and the corresponding prevailing environmental parameters. Since both species sought shade during the heat of the day, dry-bulb air temperature proved to be a better index of operative temperature than did the temperature of a black globe thermometer in the sun. I therefore correlated the amplitude of the nychthemeral rhythm of body temperature, averaged for all individuals within each species, against the daily amplitude of air temperature. I fitted and compared simple non- linear regressions for minimum and maximum daily body temperature against the maximum daily air temperature. I also correlated collar miniglobe temperatures against weather station miniglobe temperatures, using linear Pearson procedures. To assess whether the animals were conforming to ambient conditions or were selecting micro-environments, I tested whether the slope of the regression equation was significantly different from one (the slope of the line of identity). In addition, I tested whether the slope and elevation of the regression equation was significantly different between the oryx and sand gazelle across the two periods, using a two-way repeated measures ANOVA. 122 I used Statistica (kernel release 5.5 for Windows, StatSoft, Inc. (1999), Tulsa Oklahoma, USA) and GraphPad Prism (version 4.00 for Windows, GraphPad Software, San Diego California USA) for statistical analyses. Values are expressed as mean ? SD and ? < 0.05 was considered to be significant. 4.4 Results 4.4.1 Climate Air and black globe temperature varied as a function of time of day, peaking just after solar noon and reaching a minimum just before sunrise (Fig. 4.1). Mean 24-h air temperature increased from 27.7 ? 4.0?C during the warm wet period to 33.6 ? 1.5?C during the hot dry period (t120 = 10.9, P < 0.0001). Similarly, mean 24-h black globe temperature increased from 33.8 ? 3.4?C during the warm wet period to 37.8 ? 1.5?C during the hot dry period (t111 = 8.1, P < 0.0001). Solar radiation showed the expected bell-shaped distribution, with a peak around solar noon, and wind speed increased in the late afternoon (Table 4.1). Rainfall totalled 31 mm during the study period, substantially lower than the preceding 10 year average of 100 ? 60 mm. April was the wettest month (17 mm) but some rain fell during May (7 mm) and March (7mm). March was not included in my analysis of temperature data. No rain fell in the hot dry period. The climatic conditions prevailing during the hot dry period of my study were similar in temperature to the summer period in which Ostrowski and Williams (2006) studied sand gazelle, but somewhat cooler than the summer conditions of their oryx study (Ostrowski et al., 2003). My study period was much drier, though, than either of those that Ostrowski and colleagues encountered. 123 Figure 4.1. Standard black globe (mean ? SD, black line) and air dry-bulb temperature (mean ? SD, grey line) as a function of time of day, for the warm wet (left panel) and hot dry (right panel) periods, at the Saja/Umm ar-Rimth Protected Area, near the Mahazat as- Sayd Protected Area. Black bars represent night periods. Table 4.1. Environmental conditions (mean ? SD) prevalent during the warm wet and hot dry periods. Warm wet Hot dry Air temperature (?C) Mean 24 hour Minimum Maximum 27.7 ? 4.0 19.9 ? 4.2 34.1 ? 4.0 33.6 ? 1.5 25.3 ? 2.1 40.1 ? 1.4 Mean 24-hour wind speed (m.s-1) 4.4 ? 1.4 4.1 ? 1.0 Mean daytime radiation (W.m-2) 561 ? 93 604 ? 32 Total rainfall (mm) 24 0 4.4.2 Body temperature Figure 4.2 shows the nychthemeral rhythm of body temperature for the five oryx and four sand gazelle over the warm wet and hot dry periods. Sand gazelle displayed a higher mean 24-h body temperature compared to the oryx (F1,6 = 47.3, P = 0.0005), a higher 24-h minimum (F1,6 = 42.6, P = 0.0006) and a higher 24-h maximum (F1,6 = 11.0, P = 0.02) body temperature, than did the oryx (Table 4.2). However, there was no difference in the amplitude of 24-h nychthemeral rhythm of body temperature between the species (F1,6 = 4.7, P = 0.07). Also, the environmental effects on body temperature were similar for the two species. 0:00 6:00 12:00 18:00 10 20 30 40 50 60 warm wet air black globe Time of day Te m pe ra tu re (o C ) 0:00 6:00 12:00 18:00 10 20 30 40 50 60 hot dry air black globe Time of day 124 Compared to the warm wet period, during the hot dry period maximum 24-h body temperature increased (F1,6 = 84.0, P < 0.0001) and minimum 24-h body temperature decreased (F1,6 = 92.2, P < 0.0001), thus resulting in a larger amplitude of daily nychthemeral rhythm of body temperature (F1,6 = 97.6, P < 0.0001). Mean 24-h body temperature also decreased (F1,6 = 14.3, P = 0.009) during the hot dry period, compared to the warm wet period. However, there was no significant statistical interaction between the species and the period analysed for any body temperature parameter. Figure 4.2. Nychthemeral rhythm of body temperature for five free-living oryx (mean ? SD, grey line) and four free-living sand gazelle (mean ? SD, black line) over both the warm wet (left panel) and hot dry (right panel) periods. Black bars represent night periods. Table 4.2. Body temperature and daytime activity (mean ? SD) for five Arabian oryx and four Arabian sand gazelle during the warm wet and hot dry period (see text for statistical analyses). Warm wet Hot dry Sand gazelle Oryx Sand gazelle Oryx Body temperature (?C) 24-h mean 24-h minimum 24-h maximum 24-h amplitude 39.6 ? 0.1 38.3 ? 0.1 40.8 ? 0.2 2.5 ? 0.3 38.9 ? 0.2 37.8 ? 0.2 40.0 ? 0.2 2.3 ? 0.3 39.4 ? 0.1 37.4 ? 0.3 41.5 ? 0.2 4.1 ? 0.5 38.7 ? 0.2 36.1 ? 0.3 41.1 ? 0.3 5.0 ? 0.5 Mean daytime activity (% maximum) 7.4 ? 1.8 5.7 ? 1.5 2.7 ? 0.8 1.9 ? 0.9 0:00 6:00 12:00 18:00 36 38 40 42 warm wet gazelle oryx Time of day B o dy te m pe ra tu re (?C ) 0:00 6:00 12:00 18:00 36 38 40 42 hot dry gazelle oryx Time of day 125 To further assess the influence of ambient temperature on body temperature, I correlated the amplitude of the nychthemeral rhythm of body temperature with the 24-h amplitude of air temperature for the sand gazelle (Fig. 4.3A) and the oryx (Fig. 4.3B). The correlation was poor and, statistically, the amplitude of nychthemeral rhythm of body temperature was independent of the 24-h amplitude of air temperature. However, the amplitude of the nychthemeral rhythm of body temperature showed a positive power relationship with the concurrent 24-h maximum air temperature for the sand gazelle (r2 = 0.75, y = 0.87 + 6.3x10-10 x6.1) and the oryx (r2 = 0.81, y = 0.17 + 1.8x10-8 x5.2) and 24-h mean air temperature for the sand gazelle (r2 = 0.73, y = 0.86 + 1.3x10-7 x4.8) and the oryx (r2 = 0.81, y = 0.35 + 8.7x10-7 x4.4). A power curve also proved to be the most appropriate simple non-linear equation for the correlation of maximum 24-h body temperature against maximum 24-h air temperature for both the sand gazelle (Fig. 4.3C) and the oryx (Fig. 4.3D). Comparison of power curves revealed significant differences between the two species (F3,262 = 112.1, P < 0.0001). The regression line for the sand gazelle displayed a greater elevation than did the regression line for the oryx, as a result of the higher body temperatures displayed by the sand gazelle. Maximum 24-h air temperature not only predicted at least 75% of the variability in maximum 24-h body temperature, but also predicted minimum daily body temperature. Minimum 24-h body temperature was associated negatively with maximum 24-h air temperature for both the sand gazelle (Fig. 4.3E) and oryx (Fig. 4.3F). Comparison of the best-fit simple non-linear equation revealed significant differences between the two species (F2,264 = 102.6, P < 0.0001). Maximum 24-h body temperature was also correlated negatively with minimum 24-h body temperature for both the sand gazelle (r2 = 0.55, P < 0.0001) and oryx (r2 = 0.69, P < 0.0001). A partial correlation coefficient revealed that minimum 24-h body temperature of the oryx was more dependant on maximum 24-h body temperature than on maximum 24-h air temperature, and the latter correlation was spurious for oryx (rxy.z = -0.17, t134 = -1.93, P = 0.06), but not for the sand gazelle (rxy.z = -0.39, t134 = -4.86, P < 0.0001). Such findings raise the possibility that, at least for the oryx, the nadir of body temperature rhythm is set by maximum body temperature, which, in turn, is set by maximum air temperature and water stress. 126 5 10 15 20 25 0 2 4 6 A Air temperature amplitude ( ?C) Bo dy te m pe ra tu re am pl itu de ( ?? ??C ) 5 10 15 20 25 0 2 4 6 B 25 30 35 40 45 39 40 41 42 43 C Maximum air temperature ( ?C) M ax im u m bo dy te m pe ra tu re ( ?? ??C ) 25 30 35 40 45 39 40 41 42 43 D 25 30 35 40 45 35 36 37 38 39 E Maximum air temperature ( ?C) M in im u m bo dy te m pe ra tu re ( ?? ??C ) 25 30 35 40 45 35 36 37 38 39 F sand gazelle oryx Figure 4.3. The top two panels show mean 24-h amplitude of the nychthemeral rhythm of body temperature, averaged for four sand gazelle (panel A, r2 = 0.008, P = 0.30) and for five oryx (panel B, r2 = 0.03, P = 0.07) plotted against the concurrent 24-h amplitude of air temperature. The bottom four panels show the mean maximum 24-h body temperature for sand gazelle (panel C, r2 = 0.76, y = 40.3 + 1.2x10-14 x8.7) and oryx (panel D, r2 = 0.86, y = 38.8 + 1.3x10-7 x4.5), and mean minimum 24-h body temperature for sand gazelle (panel E, r2 = 0.58, y2 = 1612 ? 0.003 x3) and oryx (panel F, r2 = 0.66, y2 = 1599 ? 0.004 x 3), plotted against concurrent maximum 24-hour air temperature. All panels show 134 days of data. 127 4.4.3 Activity Low minimum body temperatures were not related to a reduced activity. Minimum 24-h body temperature was not correlated with mean 24-h activity of the previous day, for either the sand gazelle (Fig. 4.4A, r2 = 0.005, P = 0.44) or oryx (Fig. 4.4B, r2 = 0.03, P = 0.05). The fraction of total activity which took place during daylight hours was correlated negatively with the concurrent maximum 24-h air temperature for both sand gazelle (Fig. 4.4C, r2 = 0.71, P < 0.0001), and oryx (Fig. 4.4D, r2 = 0.65, P < 0.0001). There was no difference between the slope (F1,219 = 6.8, P = 0.14) nor the elevation (F1,220 = 0.01, P = 0.91) of the regression line between species. Figure 4.4. The top two panels show minimum 24-h body temperature, averaged for four sand gazelle (panel A, r2 = 0.005, P = 0.44) and for five oryx (panel B, r2 = 0.03, P = 0.05) plotted against the mean 24-h activity for the preceding day. The bottom two panels show the fraction of total activity which took place during daylight hours, for six sand gazelle (panel C, r2 = 0.71, P < 0.0001) and five oryx (panel D, r2 = 0.65, P < 0.0001), plotted against concurrent maximum 24-h air temperature. 0 5 10 15 35 36 37 38 39 40 A Mean daily activity (% maximum) M in im u m bo dy te m pe ra tu re (?C ) 0 5 10 15 35 36 37 38 39 40 B 25 30 35 40 45 0.0 0.5 1.0 C Maximum air temperature ( ?C) D iu rn al ac tiv ity (p ro po rt io n to ta l a ct iv ity ) 25 30 35 40 45 0.0 0.5 1.0 D sand gazelle oryx 128 Although the sand gazelle appeared to be more active than the oryx at all times (Fig. 4.5), the mean 24-h activity was not significantly different between the species (F1,8 = 2.1, P = 0.19). There also were no differences between species in mean activity during daylight hours (F1,8 = 2.3, P = 0.16), nor in the fraction of total activity which took place during daylight hours (F1,8 = 0.3, P = 0.59). The activity patterns of both species changed between the periods. Both species shifted from a continuous activity pattern with crepuscular peaks at sunrise and sunset during the warm wet period to nocturnal activity during the hot dry period (F1,8 = 171.8, P < 0.0001). Both species became inactive three hours after sunrise and remained inactive until an hour before sunset (09:00 and 18:00). The fraction of total activity which occurred during daylight hours decreased from 48 ? 5% to 25 ? 8% for the oryx and similarly from 47 ? 3% to 23 ? 2% for the gazelle (F1,8 = 236.1, P < 0.0001). There was no statistical interaction between the species and the period for any of the activity parameters analysed. Figure 4.5. Nychthemeral rhythm of activity for six sand gazelle (white bars) and five oryx (black bars) over both the warm wet (left panel) and hot dry (right panel) periods. Both species shifted from a continuous 24-h activity with crepuscular peaks during the warm wet period to nocturnal activity during hot dry period. Activity was measured over the abdomen with a subcutaneous activity data logger, in activity counts expressed as a percentage of maximum counts for that logger. Black bars represent night periods. See text for statistical analyses. 0 5 10 15 20 25 hot dry 0:00 6:00 12:00 18:00 Time of day 0 5 10 15 20 25 warm wet 0:00 6:00 12:00 18:00 Time of day Pr o po rt io n o f m ax im u m ac tiv ity (% ) 129 4.4.4 Microclimate Inactivity during the hot dry period was accompanied by increased shade-seeking behaviour (Fig. 4.6). During the heat of the day both oryx and sand gazelle selected microclimates which were cooler than the temperature recorded by an identical miniglobe exposed to the sun. There was no difference in the microclimates selected by the oryx and sand gazelle (F1,3 = 1.9, P = 0.26), as indexed by the mean daily maximum difference between the temperature at the site chosen by the animal and the temperature of an identical miniglobe exposed to the sun at a nearby weather station. The microclimates selected by the sand gazelle and oryx were cooler during the hot dry period (6.6 ? 0.5?C) than during the warm wet period (5.0 ? 0.3?C, F1,3 = 32.7, P = 0.01). There was no statistical interaction between the species and the period for microclimate selection. Figure 4.6. Nychthemeral rhythm of microclimate selection, expressed as the difference between miniglobe temperature at the site chosen by four free-living sand gazelle (mean ? SD, black line) and two oryx (mean ? SD, grey line) and the temperature of an identical miniglobe exposed to the sun at a nearby weather station, over both the warm wet (left panel) and hot dry (right panel) periods. Black bars represent night periods. Figure 4.7 shows the correlation between the miniglobe temperatures on the collar of a single sand gazelle and a single oryx, and temperatures recorded by an identical miniglobe in the sun at a nearby weather station, during the two periods analyzed. The slopes for all the animals, over both periods, were significantly less than one (ANCOVA, P < 0.0001), and the regression lines intersected the line of identity, implying that all animals selected microclimates cooler than the exposed microclimates at higher environmental heat loads, across all four periods. There was no species difference in the slope of the regression line between sand gazelle 0:00 6:00 12:00 18:00 -6 -4 -2 0 2 4 oryx gazelle warm wet Time of day Co lla r - w ea th er st at io n m in ig lo be te m pe ra tu re ( ?? ??C ) 0:00 6:00 12:00 18:00 -6 -4 -2 0 2 4 oryx gazelle hot dry Time of day 130 and oryx (F1,3 = 0.4, P = 0.57), nor the elevation of the regression line (F1,3 = 0.1, P = 0.75). However, the slope of the regression lines for both species decreased from 0.84 ? 0.02 during the warm wet period to 0.77 ? 0.03 during the hot dry period (F1,3 = 34.2, P = 0.01). These results imply that both sand gazelle and oryx were selecting cooler microclimates at high environmental heat loads during the hot dry periods and were less selective of microclimates during the warm wet period. There was no significant difference between the intercept of the line of identity and regression slope between the species (F1,3 = 0.4, P = 0.59) nor between periods (F1,3 = 3.3, P = 0.17), implying that both sand gazelle and oryx selected cooler microclimates when the ambient air temperature exceeded 27.9 ? 1.5?C during both periods. 10 20 30 40 50 60 10 20 30 40 50 60 A 10 20 30 40 50 60 10 20 30 40 50 60 C 10 20 30 40 50 60 10 20 30 40 50 60 B Weather station miniglobe temperature ( ?C) Co lla r m in ig lo be te m pe ra tu re ( ?? ?? C) 10 20 30 40 50 60 10 20 30 40 50 60 D sand gazelle oryx Figure 4.7. Scatter diagram showing the relationship between miniglobe temperatures at the site chosen by a male sand gazelle (left panel) and a male oryx (right panel) and the miniglobe temperatures recorded at a nearby weather station, during warm wet (upper panel) and hot dry (lower panel). Measurements were made at hourly intervals. The dashed line is the line of identity, the solid grey line is the linear regression. The slope of regression lines were not significantly different between the species, but were different between the periods. During the hot dry period both sand gazelle (panel B, y = 0.78x + 6.3) and oryx (panel D, y = 0.77x + 6.2) showed lower slopes, compared to the warm wet period for both sand gazelle (panel A, y = 0.84x + 4.3) and oryx (panel C, y = 0.85x + 3.5), indicating the selection of cooler microclimates during the hot dry period. 131 4.5 Discussion My study provides the first comparison of body temperature, activity and microclimate selection of a large and a small desert-adapted antelope species, while the animals were living free at the same time and under the same hot and hyper-arid environmental conditions. Despite the oryx having a body mass more than four-fold that of the sand gazelle, the species responded remarkably similarly to changes within those environmental conditions. If anything, the heavier oryx appeared to be more influenced by environmental temperatures and maximum air temperature accounted for 86% of the variability in maximum daily body temperature of the oryx (Fig. 4.3D), compared to only 76% for the sand gazelle (Fig. 4.3C). Maximum daily body temperatures increased and minimum daily body temperature decreased as conditions got progressively hotter and drier (Fig. 4.3). The decreasing minimum body temperature with increasing environmental heat stress may seem counter-intuitive, but is exactly what is predicted by the concept of adaptive heterothermy, namely the consequence of active suppression of heat conservation and generation at night and in the early hours of the morning, in conditions in which heat storage during the rest of the day will be high, and distinguishes adaptive heterothermy from dehydration hyperthermia (Mitchell et al., 2002). Theory predicts that species with more thermal inertia are more likely to employ adaptive heterothermy than those with less thermal inertia (Taylor, 1970a; Mitchell et al., 2002), and indeed, the heavier oryx showed a greater decrease in minimum daily body temperature (0.9?C vs. 1.7?C, Table 4.2) when maximum air temperature exceeded 35?C (Fig. 4.3F). Contrary to the conventional view of adaptive heterothermy, though, there was no association between the amplitude of the nychthemeral rhythm of body temperature, and therefore body heat storage, and the amplitude of the nychthemeral rhythm of air temperature. Thus, the nychthemeral swings in body temperature were not caused by nychthemeral swings in air temperature. Compared to that of the oryx, the thermoregulatory system of the sand gazelle appeared to operate around a higher set-point, with the sand gazelle displaying a mean 24-h body temperature an average 0.7?C higher, a higher minimum 24-h body temperature and a higher maximum 24-h body temperature (see Fig. 4.2 and Table 4.2). 132 There were no significant species? differences in activity. Both gazelle and oryx were active over the full 24 h, but with crepuscular peaks, during the warm wet period, and shifted to nocturnal activity during the hot dry period (Fig. 4.5). Activity virtually ceased during the heat of the day, when both species are known to seek shade. I was able to detect the selection of cooler microclimates without observers present, by using miniature globe thermometers attached to collars, a novel technique I developed (Appendix 1, Hetem et al., 2007) for recording microclimate selection by free-living ungulates. There was no difference in the pattern of microclimate selection by the two species, consistent with them being stationary, in the shade, during the hottest times of the day (Fig. 4.6). Both species selected cooler microclimates more readily during the hot dry period than during the warm wet period, and at times selected a microclimates with temperatures 12?C below that of an identical miniglobe in the sun (Fig. 4.7). My hot dry study period was not only significantly hotter than my warm wet period, with air temperatures being about 6?C higher at all times of day (Table 4.1), but it also was significantly drier. Indeed no rain at all fell over the two months of the hot dry period, and rain would have been the only source of drinking water available to the animals. A lack of rain would have reduced the pre-formed water in plants, and fodder quality generally. I believe that these data support the same conclusions that I drew in chapter three, namely that the important environmental variable inducing adaptive heterothermy was the lack of water availability in conditions of heat stress, rather than the heat itself. As I have shown, the degree of adaptive heterothermy employed was not related to the daily swings in air temperature (see Fig. 4.3). Also, at the same ambient temperature, the antelope selected cooler microclimates during the hot dry period than during the warm wet period (see Fig. 4.7). I believe that, if they can access sufficient water, the antelope prefer to maintain homeothermy, which will require evaporative cooling, even in the intense desert heat, rather than implementing heterothermy, and storing heat. 133 The similarities in the body temperature and activity patterns of the gazelle and oryx were remarkable, given that the gazelle were captive bred and remained in a 2 km2 enclosure throughout the study. Although the gazelle were exposed to the same environmental heat load as the oryx were, and the habitat inside their enclosure was the same as that outside it, the gazelle were not as able to seek appropriate microhabitats, including water sources, because their fenced enclosure restricted long-distance movements. Even though sand gazelle are reputed to be able to survive independent of surface water (Ostrowski et al., 2006a), the scarcity of resources in the enclosure, after an exceptionally hot and dry period, may have contributed to the death of the gazelle five months after implant surgery. If so, that resource scarcity was not confined to my gazelle?s enclosure, because the high mortality of sand gazelle was not specific to the gazelle in my enclosure; nearly 700 sand gazelle (50% of the estimated population) died in the Mahazat as- Sayd during 2006 (Cunningham et al., 2008). The body temperatures of sand gazelle and oryx inhabiting the Mahazat as-Sayd Protected Area have been measured previously in two studies separated by 3-6 years (Ostrowski and Williams, 2006). The body temperatures of the gazelle were measured with data loggers similar to mine, but those of the oyrx were measured by radiotelemetry, which allowed only sporadic measurements spread over three years, and in different animals, especially at night. Similar to the results of my study, sand gazelle had a higher mean daily body temperature of 39.5 ? 0.3?C (Ostrowski and Williams, 2006), compared to 38.4 ? 1.3?C for the oryx (Ostrowski et al., 2003), but unlike my gazelle, they had a smaller amplitude of nychthemeral body temperature rhythm of 2.6 ? 0.8?C (Ostrowski and Williams, 2006), compared to 4.1 ? 1.7?C for the oryx (Ostrowski et al., 2003), during summer. Since conditions were milder during the sand gazelle study, the differences in body temperature were attributed to differences in environmental conditions prevailing during the two studies (Ostrowski and Williams, 2006), and indeed, as I have shown (Table 4.2), sand gazelle can achieve a rhythm amplitude the same as that of the oryx. My study had the advantage of comparing these species in the same environmental conditions, so that any differences I observed 134 between the species were inherent, and not the result of different environmental conditions. Even given their constraints, though, what the studies of Ostrowski and colleagues did establish was that free-living antelope in the Saudi Arabian desert can employ heterothermy. On the basis of studies of African antelope my colleagues had concluded previously that adaptive heterothermy may have been an artefact of confinement (Mitchell et al., 2002). I now have confirmed the ability of free-living sand gazelle and oryx to use heterothermy, but I differ from Ostrowski et al. (2003), in that I found no association between the degree of heterothermy (and therefore heat storage) and the amplitude of variation in air temperature in the oryx. Ostrowski and his colleagues, like myself, found no association between the degree of adaptive heterothermy and amplitude of air temperature rhythm in sand gazelle (Ostrowski and Williams, 2006). I do not believe that the degree of heterothermy is driven by swings in ambient temperature, but rather by water availability, and consequently that the large amplitudes that I found, in both species, at ambient temperatures similar to those which Ostrowski studied sand gazelle, but cooler than those in which they studied oryx, were the result of water deprivation. Although Ostrowski and colleagues contradicted themselves later (Ostrowski and Williams, 2006), they stated that their oryx, in summer when they showed a large amplitude in rhythm of body temperature, could have been ?somewhat dehydrated?, showing ?higher haematocrit, plasma protein concentration and plasma osmolality? (Ostrowski et al., 2003). Similarly, their sand gazelle, with a plasma osmolality of about 320 mOsm in September and October (Ostrowski and Williams, 2006), that is in autumn, were likely to have been severely dehydrated in the summer (Alamer, 2006), the time at which they showed appreciable heterothermy. The main body temperature change inducing the increase in body temperature amplitude in the hot dry period, in my oryx and sand gazelle, was the drop in minimum body temperature. While the original definition of adaptive heterothermy, and subsequent studies, imply that the drop in minimum body temperature was under active physiological control (Schmidt-Nielsen et al., 1957; 135 Ostrowski et al., 2003; Maloney et al., 2004), it could result from thermoregulatory failure resulting from undernutrition. At that time, of heat without any rainfall, the forage quality available to the animals would have been poor, and, in other artiodactyl species, energy restriction leads to a lower nadir in the nychthemeral rhythm of body temperature (Choshniak et al., 1995; Piccione et al., 2002; Ahmed and El Kheir, 2004). However, since the maximum 24-h body temperature was a stronger determinant of the nadir of body temperature rhythm it seems more likely that, in circumstances in which animals otherwise would use evaporative cooling but don?t have access to enough water, the nadir of body temperature is set by dehydration. Dehydrated animals would benefit most from the water savings associated with a lower nadir of body temperature rhythm, but such an hypothesis that the nadir of body temperature rhythm is set by ambient temperature and dehydration state, perhaps via the animal?s osmolality, would need to be established through laboratory experimentation. Irrespective of the origin of the lower nadir, though, it led to increased capacity to store heat during the day, in both species. The lower minimum body temperature displayed by my oryx potentially would permit additional storage of heat during the day, compared to the sand gazelle. Such a capacity for increased heat storage combined with a greater thermal inertia may add support for Taylor?s (1970a) theory that antelope with different body masses employ different strategies of heterothermy. Under simulated desert conditions in a laboratory, small gazelle allowed their rectal temperature to increase and exceed air temperature, whereas the larger antelope displayed low morning rectal temperatures, which allowed heat storage during the day negating the need for excessively high maximum body temperatures (Taylor, 1970a). However, since there was no difference in the nychthemeral rhythm of body temperature between my species, such differences in body temperature may be inherent. Since I have established that the difference observed earlier in thermoregulatory function of sand gazelle and oryx are not only the consequence of the species previously having been exposed to different environmental conditions, I can seek an intrinsic origin for the differences, and the parsimonious explanation would be 136 that they relate to body size. However, the scaling of thermoregulatory function with body size is neither simple nor universally agreed. I assert caution about drawing any phylogenetic conclusions from a comparison of just two species. Given those complexities and that caution, the higher set-point for body temperature regulation displayed by the sand gazelle, both in my study and that of Ostrowski and Williams (2006), relative to the oryx, conforms to the negative scaling of body temperature with body mass reported by Clarke and Rothery (2008), particularly for the artiodactyls, and for mammals generally (Morrison and Ryser, 1952 ; McNab, 1970; Aschoff, 1982), and not with the positive scaling proposed originally by Rodbard (1950) and more recently by White and Seymour (2003). That the amplitude of the nychthemeral rhythm of body temperature of the two species, when exposed to the same environment, was not significantly different conforms to the prediction that the amplitude will be mass-independent in larger mammals (Aschoff, 1982). That the amplitude was the same for the two species, and also that the amplitude increased by the same magnitude between the warm wet and hot dry periods, does not imply that the species were equally dependent on heterothermy. With a mass approximately four times greater, the oryx would store about four times as much heat as the gazelle as body temperature rises, and therefore save about four times as much water, that otherwise would be spent on evaporative cooling (Mitchell et al., 2002). The relative benefit of that water saving would depend on factors such as metabolic rate, apportioning of cooling to evaporation, and radiant and connective heat transfer between the animal and the environment in the sun and shade, factors unknown for the gazelle and oryx. It was not just the temperature rhythm amplitudes that did not differ between the gazelle and the oryx, despite the differences in mass, but also the nychthemeral activity pattern and its response to increasing heat and aridity (Fig. 4.5). Since time spent active was likely to correlate with time spent foraging, my data do not conform to the prediction that larger species would spend proportionally more time foraging, as a result of increased metabolic (Peters, 1983; Schmidt-Nielsen, 137 1984; Owen-Smith, 1988; Owen-Smith, 1992; du Toit and Yetman, 2005) and other nutritional (Demment and Van Soest, 1985) requirements. Nor do my data conform to the contrary view of the allometric scaling, namely that that time spent active, particularly time spent foraging, tends to decrease with increasing body mass (Belovsky and Slade, 1986; Mysterud, 1998; P?rez-Barber?a and Gordon, 1999). Further, my data do not conform to the principle that mixed feeders which consume high quality forage, such as sand gazelle (Groves, 1988), are more active are than grazers, such as oryx (Bunnell and Gillingham, 1985; Klein and Fairall, 1986; Owen-Smith, 1992; Mysterud, 1998; Taylor et al., 2006). Instead, my activity data fit best with the proposition that neither feeding style (P?rez-Barber?a and Gordon, 1999) nor body mass (Jeschke and Tollrian, 2005) significantly influence activity. Body mass also has been proposed to influence the apportioning of feeding between day and night (Owen-Smith, 1988), but, once again, there is no agreement about even the slope of a relationship. Some authors have proposed that smaller species would be more vulnerable to predation and therefore forced to be more active at night (P?rez-Barber?a and Gordon, 1999), but others have concluded that larger species spend less time foraging during the daylight hours than do smaller species with similar food habits (Bunnell and Gillingham, 1985). Though the difference in activity patterns between my sand gazelle and oryx were small, and perhaps surprisingly so, there were very significant differences in the changes in pattern for both species, between the warm wet and hot dry periods. Both species shifted from continuous activity with crepuscular peaks in warm wet months to nocturnal activity during hot dry months (Fig. 4.5). The biological phenomenon of changing the timing of activity, in a way that depends on the environmental conditions, has been termed ?cathemerality? (Tattersall, 2006; Tattersall, 2008). Pressure towards cathemerality is proposed to be greater for large animals, because of their greater absolute energy requirements, and all Indonesian mammals studied with a body mass above 100 kg, and most above 10 kg, were cathemeral (van Schaik and Griffiths, 1996). However, whereas in the Indonesian study the pressure towards cathemerality was likely related to foraging 138 requirements, my sand gazelle and oryx appeared to shift activity patterns based on prevailing ambient heat and aridity. If thermal stress is the driving force for cathemerality, one would expect smaller species to make the greater relative adjustments to diurnal activity in hot environments, as a result of their lower thermal inertia, greater surface area to mass ratio, and smaller surface area for evaporative cooling. Several studies have reported the attenuation of diurnal activity patterns of small ungulates at high ambient temperatures (Zervanos and Hadley, 1973; Kamau and Maloiy, 1985; Wilson, 1989; Roberts and Dunbar, 1991; du Toit, 1993), commonly as a result of shade-seeking sedentary behaviour. Although high ambient temperatures also reduce the activity of large species (Jarman, 1977; Belovsky and Jordan, 1978; Leuthold and Leuthold, 1978; Bunnell and Gillingham, 1985; Belovsky and Slade, 1986; Klein and Fairall, 1986; Owen-Smith, 1998), that reduced activity was not necessarily associated with shade-seeking behaviour. Because nocturnal activity is difficult to measure without biotelemetry, which has become available only recently, it is not clear whether their adjustments of diurnal activity were accompanied by an increase in the proportion of nocturnal activity, as required if the pattern change is cathemeral, rather than just lethargy. In my oryx and gazelle, in which activity was measured by biotelemetry, the decrease in diurnal activity in the hot dry period indeed was compensated by an increase in nocturnal activity (Fig. 4.5), but also with disappearance of the crepuscular peaks. Because my study design required observers not to be near the animals, I did not observe what they were doing while they were inactive during the day. Under observation, oryx abandon foraging and seek refuge in shade on hot days (Ostrowski et al., 2003), and anecdotal evidence claims that sand gazelle do too (Ostrowski and Williams, 2006). In general, smaller diurnal animals may be forced to take refuge from temperature extremes more frequently (Peters, 1983) and are better able to seek refuge from the sun than their larger counterparts (Louw and Seely, 1982). Yet, the selection of cool microclimates during the heat of the day has been observed in a variety of desert ungulates, both large (Taylor 139 and Lyman, 1967; McCutchen, 1981; Hansen, 1982; Sargeant et al., 1994; Tull et al., 2001; Ostrowski et al., 2003) and small (Zervanos and Hadley, 1973; Kamau and Maloiy, 1985; Wilson, 1989). What I have shown, again using biotelemetry, is that, in the absence of human interference, the oryx and gazelle occupied microclimates cooler than the exposed thermal environment, presumably shaded microclimates, when environmental heat load exceeded 28?C. At any particular ambient temperature, the degree to which they avoided exposed microclimates increased when the habitat was drier. Although sand gazelle have been observed to retreat to small rock crevices during the heat of the day (Ostrowski and Williams, 2006), a microclimate proposed to provide greater thermal advantages than vegetation (Cain et al., 2008), the similarity in the microclimate selection (Fig. 4.6) implies that the oryx had access to microclimates as cool as those available to the much smaller sand gazelle. In summary, using biotelemetric techniques which free the animals from disturbance by human observers, I have confirmed that both Arabian oryx and sand gazelle employ adaptive heterothermy, and here show for the first time that, in the same environment, the amplitudes of their nychthemeral rhythm are the same, despite their different masses. I also have shown cathemerality in the activity of both species. I believe that, in very hot environments, the degree of adaptive heterothermy employed, and the degree of avoidance of exposed microclimates, are determined not by ambient temperatures alone, but by the aridity, consonant with those thermoregulatory processes saving body water. Although both sand gazelle (Ostrowski et al., 2006a) and oryx (Ostrowski et al., 2006b) are reported to have a low total evaporative water loss, and are believed to survive independent of surface water, the year of my study was particularly dry. Under conditions of extreme aridity in the Negev desert, the dorcas gazelle (Gazella dorcas), also reputed to survive independent of surface water, preferentially chose habitats with access to drinking water (Henley et al., 2007). With many areas of the Africa predicted to get both hotter and drier under climate change scenarios, medium and large desert artiodactyls are likely to have to cope with increasing physiological stress. Although numerous small endotherms have 140 exhibited a linear decrease in body mass with recent warming trends (Smith et al., 1995; Smith et al., 1998; Yom-Tov, 2001; Smith and Betancourt, 2006; Yom-Tov et al., 2006), such an adaptation strategy may not be appropriate for medium and large desert artiodactyls. Since water requirements scale allometrically with body mass (MacFarlane et al., 1971), small ungulates may be particularly stressed in the future. 4.6 Acknowledgements I thank the National Commission for Wildlife Conservation and Development (NCWCD), Riyadh, Saudi Arabia, in particular the director His Royal Highness Prince Saud Al Faisal, the current secretary-general, His Highness prince Bander Bin Saud, and the secretary-general at the time the study was conducted, Professor AH Abuzinada, for supporting the research. From the National Wildlife Research Center (NWRC), I am grateful to Dr Saud Anagariyah for his support in capturing the oryx and the current director, Ahmad Al Bouq. In addition, I thank the Mahazat as-Sayd Protected Area rangers for monitoring the animals and the mammal keepers at NWRC for their help with animal handling and assistance during surgery. 141 ___________________________________________________________ CHAPTER 5 ___________________________________________________________ 5 Effects of desertification on the physiology of Angora goats: testing global change predictions Data and ideas presented in this chapter have been written up as a scientific paper and will be submitted to Journal of Arid Environments. Hetem R.S., de Witt B.A., Fick L.G., Fuller A., Maloney S.K., Meyer L.C.R., Mitchell D. and Kerley G.I.H. Effects of habitat transformation on the physiology of Angora goats: testing global change predictions. 142 5.1 Abstract Globally, pastoralism has led to the transformation of habitat, which often leads to desertification. With climate change predicted to exacerbate desertification, phentoypic plasticity provides the best survival strategy for agriculturally important herbivores, which are long-lived and, for economic reasons, cannot be translocated. While changes in land use and regional climate have been assessed, there is little information on the plasticity of herbivores in responding to these changes. With the help of my colleagues, I investigated the physiology of Angora goats inhabiting transformed and intact sites in the Eastern Cape Province, South Africa. Although goats on both sites responded similarly under most environmental conditions, when goats were subjected to a physiological stress, imposed by shearing, goats inhabiting the transformed site had a faster rate of rise in abdominal temperature in the morning (0.38 vs. 0.31?C.h-1, P = 0.0009), displayed an increased 24-h abdominal temperature amplitude (1.8 vs. 1.6?C, P = 0.01) and were generally less active (3.9 vs. 5.2 activity units) compared to goats inhabiting the intact site. Post-shearing, goats inhabiting the transformed site had higher water turnover rates (P < 0.0001) and selected more variable microclimates (P < 0.0001), than did goats inhabiting the intact site, even though they obtained less water from their diet (P = 0.02). Goats that inhabited the transformed site were more water dependent and appeared more susceptible to thermal stresses in their environment than were those that inhabited the intact site. Coping with such thermal challenges will be essential if livestock farming is to thrive under climate change scenarios. 5.2 Introduction Climate change is likely to affect livestock production both because high ambient temperatures compromise the reproductive efficiency and performance of livestock (Scholes et al., 1999; Nardone et al., 2006), and because the predicted decrease in the quantity and quality of forage (Topp and Doyle, 1996) is likely to reduce the carrying capacity of rangelands (Hanson et al., 1993; Milton and Dean, 1995; Richardson et al., 2005). Since agriculture is inherently sensitive to extreme 143 and variable climatic conditions (Katz and Brown, 1992), it is predicted to be the economic sector most vulnerable to the risks and impact of climate change (Parry and Carter, 1989; Reilly, 1995). Agricultural production in Africa, in particular, may be compromised severely by climate change because of its heavy dependence on natural resources and a low adaptive capacity (Boko et al., 2007). In advance of the potential threats from climate change, in the subtropical thickets of the Eastern Cape of South Africa, domestic ungulates have been replacing indigenous herbivores in this area since the mid 1800s (Downing, 1978), resulting in severe habitat transformation (Aucamp et al., 1980; Kerley et al., 1995; Palmer et al., 2004). In many places, heavy browsing by goats has transformed the indigenous thicket vegetation from a dense closed-canopy shrubland, frequently dominated by the highly palatable and nutritious forage plant Portulacaria afra, into an open savanna-like system (Hoffman and Cowling, 1990; Stuart-Hill, 1992; Moolman and Cowling, 1994; Kerley et al., 1995; Robertson and Palmer, 2002; Sigwela et al., 2003; Lechmere-Oertel et al., 2005b). Such transformation of the natural flora, classified as desertification or dryland degradation, is the result of overgrazing due to human mismanagement of agricultural and pastoral resources (Dean et al., 1995; Kerley et al., 1995). These severely degraded areas of South Africa are likely to become even more vulnerable under predicted climate change scenarios (Meadows and Hoffman, 2003). However, such habitat transformation is not unique to Africa and changes in the structure and dynamics of the plant community and ecosystem stability as a result of unsustainable livestock production are a world-wide phenomenon (Mattison and Norris, 2005). Climate change will exacerbate desertification (Le Houerou, 1996; Robertson and Palmer, 2002; Sivakumar, 2007), particularly because of the increased variability of climatic conditions and the frequency of extreme events, such as droughts (Mason and Joubert, 1997; Boko et al., 2007). In order to meet agricultural requirements in the future it will become necessary to make use of all available pastureland (Bianca and Kunz, 1978). Domestic ungulates will have to cope with, and adapt to, climatically-unfavourable regions and an increased mean 144 annual temperature (Boko et al., 2007), which is likely to increase energy demand (Canas et al., 2003). Compared to other domestic ungulates, goats are well adapted to harsh, dry environments, as they have a relatively small body mass, low metabolic requirements, are able to use low-quality forage, are disease resistant and can survive long periods of water deprivation (Lu, 1988; Erasmus, 2000; Silanikove, 2000a; Silanikove 2000b; Lachica and Aguilera, 2003; Salem et al., 2006; Tovar-Luna et al., 2007). A proposed management response to climate change is the selection of heat-tolerant breeds (Nardone et al., 2006), which would allow pastoralism to persist even in very hot and dry environments. Additional management responses to climate change include changes in stocking rate, forage supplementation and alternative land uses such as game ranching (Scholes et al., 1999), which already is occurring in the Eastern Cape Province, South Africa (Smith and Wilson, 2002). Nevertheless, stock farming remains the primary land use in the Eastern Cape. It is doubtful whether the transformed landscapes can be farmed profitably with goats at the low stocking rates, which are likely to result (Stuart-Hill, 1992; Stuart-Hill and Aucamp, 1993). How domestic ungulates cope with habitat transformation will become increasingly important in the face of global climate change. I therefore investigated the physiology of Angora goats inhabiting transformed and intact thicket sites in the Eastern Cape. Although Angora goats are susceptible to inanition because of their high nutritional requirement for fibre production (Hart et al., 1993), they are particular well suited to low (230-560mm) rainfall areas (Norton and Deery, 1985), making them a good domestic ungulate model to investigate such responses. Most previous climate change assessments do not consider pre- adaptation to the expected conditions (Smit and Skinner, 2002), and especially not the plasticity of herbivore responses to these changes. 145 5.3 Materials and methods 5.3.1 Animals and habitat Twenty-four neutered male Angora goats (Capra aegagrus), at Blaauwkrantz farm (33?32?S 25?23?E), near Port Elizabeth in the Eastern Cape Province, South Africa, underwent surgery (detailed below) in July 2005 before being randomly assigned to one of two habitats on the farm. Twelve goats (body mass 39.2 ? 3.5 kg, mean ? SD) were released into a 50 ha camp (intact site, comprising dense thicket vegetation) classified as Sundays Spekboomveld (Vlok et al., 2003) and 12 goats (body mass 39.7 ? 5.9 kg) were released into an adjacent 40 ha camp (transformed site), transformed by historical heavy browsing to form a savanna- like vegetation with a cover of ephemeral grasses and forbs and few remnant small trees (Lechmere-Oertel et al., 2005a). Goats had access to water ad libitum and foraged on the natural vegetation. Experimental procedures were approved by the Animal Ethics Screening Committee of the University of the Witwatersrand (protocol no. 2005/45/4). 5.3.2 Surgery For implantation of data loggers, the goats were anaesthetised by intramuscular (I.M.) injection of 2.5 mg.kg-1 ketamine hydrochloride (Anaket-V, Bayer Animal Health Pty, Isando, South Africa) and 0.04 mg.kg-1 medetomidine hydrochloride (Domitor, Novartis, Kempton Park, South Africa). Anaesthesia was maintained with 1-4% halothane (Fluothane, Astra Zeneca, Johannesburg, South Africa) administered in 100% oxygen via a facemask. After ~ 10 min of halothane administration, the action of medetomidine was reversed with 0.2 mg.kg-1 (I.M.) atipamezole hydrochloride (Antisedan, Novartis, Kempton Park, South Africa). Respiratory rate, heart rate, oxygen saturation (pulse oximeters, Nonin 9847V, Nonin Medical, North Plymouth, USA) and rectal temperature (thermocouple thermometer, BAT-12, Physitemp, Clifton, USA) were monitored throughout the surgical procedure, which lasted ~ 30 min. 146 All implanted data loggers were covered in an inert wax (Sasol wax EXP986, Sasol, Johannesburg, South Africa) and dry-sterilized in formaldehyde vapour before implantation. Incision sites were shaved and sterilized with chlorhexidine gluconate (Hibitane, Astra Zeneca, Johannesburg, South Africa). A 70 mm cranial-caudal incision was made through the skin and linea alba, within the midline area on the ventral abdominal surface, and a miniature temperature- sensitive data logger (see description below) was placed into the abdominal cavity, where it floated freely. A single goat from each group also received an implantable tracking transmitter (African Wildlife Tracking, Pretoria, South Africa). The skin and muscle layers then were sutured closed. In six of the goats from each group, after administration of a local anaesthetic (0.04 g lignocaine hydrochloride, S.C., Bayer Animal Health Pty, Isando, South Africa), an additional incision was made on the upper hind leg. A smaller temperature- sensitive data logger (see below) was inserted subcutaneously. At the same subcutaneous site, a single goat from each group also had an activity logger (Actical, Mini-Mitter Corporation, Bend, OR, USA) implanted, which was tethered in a small pocket. The activity logger recorded at 10-min intervals, had dimensions of 40 x 40 x 15 mm, and weighed ~ 40 g when covered in wax. Activity counts were normalised for different logger sensitivities by expressing activity counts as a percentage of maximum counts for that logger. Wounds were sprayed with a topical antiseptic spray (Necrospray, Centaur Labs, Johannesburg, South Africa) and coated with a topical tick repellent (Tickgrease, Cypermethrin 0.025% m/m, Bayer Animal Health Pty, Isando, South Africa). Each goat received a long-acting antibiotic (500 mg I.M. penicillin, Peni LA Phenix, Virbac Animal Health, Centurion, South Africa), an analgesic and anti- inflammatory medication (420 mg I.M. ramiphenazone, Dexa-Tomanol, Centaur Labs, Johannesburg, South Africa), and a long-acting parasiticide (5 mg S.C., doramectin, Dectomax, Pfizer Laboratories, Sandton, South Africa). Animals were marked individually with different coloured plastic ear tags. 147 Before halothane administration was terminated, a collar was fitted to a single goat from each group. The collar supported a miniature (diameter 30 mm) black globe thermometer (?miniglobe?) to allow for the dynamic measurement of the microclimate the goat chose to occupy. This technique has proven successful on other ungulate species (Appendix 1, Hetem et al., 2007). Miniglobe temperature was measured with a small temperature sensitive data logger, which was inserted into the centre of the matt-black hollow bronze sphere (Press Spinning & Stamping co., Cape Town, South Africa). The globe was attached to a 10-mm diameter, polyvinyl chloride rod, which in turn was attached to the apex of the collar. After recovery from surgery, the goats were released into their allocated habitats where they roamed freely for a year. Within each site, goats behaved similarly and remained as a herd, so I believe that the activity and microclimate selection of the index goat reflected that of all goats in that habitat. They were caught and weighed once a month and subjected to standard husbandry practices of parasite control. In September 2005 and March 2006, the goats were shorn according to the standard management practices of the farm. The diet composition and quality, and fibre production of these animals were measured in a parallel study (Milne, 2008). In August 2006, the goats once again were anaesthetised and the data loggers were removed under a surgical procedure similar to that used for the original implantation. The data loggers all were in perfect order, the animals? wounds had healed and there were no signs of infection from the initial surgery. Most of the abdominal loggers were found in the pelvic canal and were not encapsulated in adhesive tissue. After surgery the goats were returned to the resident herd on Blaauwkrantz farm. 5.3.3 Temperature measurements The miniature data loggers (StowAway XTI, Onset Computer, Pocasset, USA) used to measure abdominal temperature had outside dimensions of ~ 50 ? 45 ? 20 148 mm and a mass of ~ 40 g when covered in wax. The data loggers had a storage capacity of 32 kb and measured temperatures within the range of + 34 to + 46 ?C, at a resolution of 0.04 ?C. The loggers measuring abdominal temperature were set to record at 20-min intervals. Subcutaneous and miniglobe temperatures were recorded every hour with a smaller thermometric data logger (iButton DS1922T, Maxim, Dallas Semiconductor, Texas USA), which had a diameter of ~ 25 mm, a height of 15 mm, and weighed about 10 g. These loggers had a resolution of 0.5?C and a measurement range from 0 to 125?C. All the loggers were calibrated, in an insulated water bath, against a high-accuracy thermometer (Quat 100, Heraeus, Hanau, Germany). 5.3.4 Meteorological data measurements I collected climatic data by erecting a portable weather station at the field site (Hobo Weather Station, Onset Computer Corporation, Pocasset, Mass., USA). I recorded wind speed (m.s-1), solar radiation (W.m-2), dry-bulb temperature (?C), relative humidity (%), standard (150 mm diameter) and miniature (30 mm diameter) black globe temperature (?C) hourly. 5.3.5 Isotope analysis and water balance I assessed water influx and water turnover rates, by dilution of the stable isotope, deuterium oxide, 10 days before and 10 days after the summer (March) shearing. Before the deuterium oxide was administered, animals were weighed and 10 ml jugular blood samples were taken to determine the background concentration of deuterium oxide. Animals then received a dose of 0.04 ml.kg-1 deuterium oxide (D2O, I.M., 99.8 atom %, Merck & Co. Ltd. Rahway, NJ, USA). A second 10 ml blood sample was taken four hours after injection, when I assumed that the administered deuterium oxide had equilibrated with body water. The goats? drinking water source was removed, and they were weighed and 10 ml blood samples were taken three, five and seven days later. Water was returned after the blood samples had been taken on the seventh day. The blood samples were collected by jugular venipuncture with Vacutainers (BD Diagnostics-preanalytical 149 systems, Plymouth, UK), placed in ice and centrifuged (Wifug Ltd., Bradford, England) at 6000 G for 10 min. Serum and plasma were separated and stored in 2 ml vials (cryogenic vials 430489, Corning Inc., NY, USA) and frozen at -20?C for later analysis. Deuterium concentration of the serum was measured using a high-temperature elemental analyzer (Flinnigan elemental analyser, Thermo electron corporation, Bremen, Germany), normalized against an international references, namely Vienna-Standard Mean Ocean Water (V-SMOW). A multi-point method was used to determine water turnover in the goats (Fusch, 2000) and a log-linear regression line to establish rate constants for deuterium dilution. The measure of total body water was derived from the known dose of injected deuterium oxide divided by the difference between the y-intercept of the curve at time zero and the background concentration of deuterium oxide. Water influx was determined from the average rate of deuterium oxide dilution over the seven-day period, when goats did not have access to water, multiplied by the total body water. Water turnover was determined as the sum of water influx and the total weight loss, as all weight loss was assumed to be water loss. To estimate the type of food consumed, the 13C values (?0.1 ? VPDB) of the goat serum samples were determined using a Thermo Delta V Plus mass spectrometer (Thermo Electron Corporation, Waltham, MA, USA) plumbed inline with a Flash EA 112 elemental analyser (Thermo Electron Corporation, Waltham, MA, USA), at the Council for Scientific and Industrial Research (CSIR) in Pretoria, South Africa. Laboratory standards with 13C values of -25.0 ? 0.1 ? VPDB and -11.4 ? 0.1 ? VPDB were used to correct for equipment drift. Carbon isotope ratios can be used as an indication of diet selection, since plants which fix CO2 by way of the Calvin C3 cycle pathway differ in their natural abundance of 13C from plants which fix CO2 through the C4-dicarboxylic acid pathway (Smith and Epsten, 1971). Portulacaria afra has a Crassulacean Acid Metabolism (CAM) photosynthetic pathway during summer and therefore has a 150 carbon isotope ratio of approximately -17.4 (Mills et al., 2005), whereas the C4 grasses approximate -12 and C3 trees approximate -25. 5.3.6 Blood variables Serum albumin, globulin and glucose concentrations were analyzed by the Department of Companion Animal Clinical Studies section of Clinical Pathology, University of Pretoria, Pretoria, South Africa. Total plasma lipids were extracted using the technique described by Bligh and Dyer (1959). Briefly, methanol and chloroform (50 ml, 2:1) were added to the sample and allowed to homogenize overnight. The sample was then filtered and rehomogenized in 15 ml of weak salt solution (0.85% NaCl). After the final biphasic system was separated into two layers, the lower (chloroform) phase was collected for evaporation and subsequent lipid estimation. 5.3.7 Data analysis To test how the goats in the two habitats responded to climatic variation, before being shorn, I analysed data from the five hottest days and the five coldest days of my study. The five hottest days were non-continuous and determined by the highest maximum globe temperatures, which averaged 47.7 ? 3.1?C, within the two-month period before the March shearing. The five coldest days also were non-continuous and determined by the lowest night-time air temperature, which averaged 0.4 ? 0.5?C, within the month before the September shearing. The mean ? SD of the mean, minimum, maximum and amplitude of nychthemeral rhythm of abdominal temperature of the goats that inhabited the two treatments were calculated for each of these periods. I used a repeated measures two-way analysis of variance (ANOVA) to test for differences between the goats that inhabited the two habitats across hot and cold days. I also tested for differences in the time of minimum and maximum abdominal temperature, as well as differences in the maximum rate of abdominal temperature rise, between the goats that inhabited the two habitat types. 151 To assess how the goats that inhabited the two habitat types responded to shearing stress I analysed 10 days of data after the summer (March) shearing, hereforth termed ?post-shearing?. I used an unpaired Student?s t-test to investigate differences in the temperature profiles between the goats that inhabited the two habitat types post-shearing. Over seven of those days I also performed a repeated measures two-way ANOVA on each of the blood variables to test for an interaction between the goats that inhabited the two habitat types across time. Newman-Keuls multiple comparisons tests were used to identify sources of significant differences in ANOVAs. To assess whether the animals were conforming to ambient conditions or were selecting microclimates, I tested whether the slope of the regression equation, fitted to the correlation between miniglobe temperature of the microclimate selected by the index goat and weather station miniglobe temperature, was significantly different from one (the slope of the line of identity). In addition, I tested whether the slope and elevation of the regression equation was significantly different between pre-shearing and post- shearing periods using an analysis of co-variance (ANCOVA). I used Statistica (kernel release 5.5 for Windows, StatSoft, Inc. (1999), Tulsa Oklahoma, USA) and GraphPad Prism (version 4.00 for Windows, GraphPad Software, San Diego California USA) for statistical analyses. Data are expressed as mean ? SD and ? < 0.05 was considered to be statistically significant. 5.4 Results 5.4.1 Climate Black globe temperature peaked just after solar noon and reached a minimum just before sunrise. Solar radiation peaked between 12:00 and 13:00 and wind speed increased in the late afternoon. Environmental conditions over the three periods analyzed, namely cold days, hot days and post-shearing, are presented in Table 5.1. Both solar radiation and wind speed were lower on the cold than on the hot days, but both maintained a similar rhythm. Rainfall totalled 402 mm over the 152 one-year study period, which was 29% higher than the long term (1958 - 2005) average annual rainfall of 313 mm per annum (A. Rudman, pers. com.). Table 5.1. Environmental conditions prevailing at the field site for the five hottest days, the five coldest days, and the 10-day period after summer shearing. Hot days Cold days Post-shearing Black globe temperature (?C) 24-h mean 24-h minimum 24-h maximum 29.2 ? 1.1 18.3 ? 1.7 47.7 ? 3.1 14.6 ? 1.4 0.4 ? 0.5 31.1 ? 2.4 22.2 ? 1.4 12.7 ? 3.2 35.9 ? 3.5 Mean daytime radiation (W.m-2) 564 ? 66 375 ? 26 379 ? 160 Maximum wind speed (m.s-1) 4.5 ? 0.5 3.6 ? 1.1 4.0 ? 1.0 5.4.2 Hot and cold days The abdominal temperature of the goats showed a nychthemeral (24-h) rhythm with a nadir shortly after sunrise and peak near sunset. On cold days, daily mean (F1,20 = 129.5, P < 0.0001), minimum (F1,20 = 91.9, P < 0.0001) and maximum (F1,20 = 54.8, P < 0.0001) abdominal temperatures of the goats were lower, but the 24-h amplitude (F1,20 = 6.1, P = 0.02) of abdominal temperature rhythm was higher, than on hot days (Fig. 5.1A and 5.1B). On hot days (Fig. 5.1B), goats that inhabited the intact site reached a slightly higher absolute maximum abdominal temperature (40.8 ? 0.3?C versus 40.6 ? 0.3?C, P = 0.01) than did goats that inhabited the transformed site. However, on cold days (Fig. 5.1A), goats that inhabited the transformed site displayed a higher absolute maximum abdominal temperature (40.4 ? 0.2?C versus 40.2 ? 0.2?C, P = 0.02), a greater variability of abdominal temperature, as represented by a larger standard deviation of abdominal temperature (0.47 ? 0.05?C versus 0.38 ? 0.04?C, P = 0.004), and a higher amplitude of abdominal temperature rhythm (1.98 ? 0.24?C versus 1.61 ? 0.19?C, P = 0.001) than did goats that inhabited the intact site. Goats that inhabited the transformed site also reached their minimum abdominal temperatures later in the day (09:07 ? 0:39 versus 07:39 ? 1:17, P = 0.003), which 153 resulted in faster maximum rate of abdominal temperature rise during the day (0.36 ? 0.07?C.h-1 versus 0.24 ? 0.05?C.h-1, P = 0.0002) than goats that inhabited the intact site. On cold days, the daily mean (F1,9 = 150.9, P < 0.0001), minimum (F1,9 = 152.4, P < 0.0001) and maximum (F1,9 = 11.8, P = 0.007) subcutaneous temperatures of the goats were lower, and had a greater amplitude (F1,9 = 72.1, P < 0.0001), than on the hot days (Fig. 5.1C and 5.1D). Daily mean subcutaneous temperatures were lower in the goats that inhabited the transformed site than in the goats that inhabited the intact site, on both hot (39.0 ? 0.5?C versus 39.4 ? 0.4?C, P = 0.008, Fig. 5.1D) and cold (38.0 ? 0.4?C versus 38.4 ? 0.6?C, P = 0.007, Fig. 5.1C) days. In addition, goats that inhabited the transformed site had a lower daily minimum (38.0 ? 0.7?C versus 38.7 ? 0.5?C, P = 0.04) and a lower daily maximum (40.0 ? 0.5?C versus 40.4 ? 0.4?C, P = 0.01) subcutaneous temperature than did goats in the intact site on hot days (Fig. 5.1D). These low subcutaneous temperatures resulted in a larger difference between abdominal and subcutaneous temperature in goats that inhabited the transformed site compared to those in the intact site, on both hot (0.79 ? 0.62?C versus 0.40 ? 0.35?C, P = 0.03, Fig. 5.1F) and cold (1.3 ? 0.4?C versus 1.0 ? 0.5?C, P = 0.04, Fig. 5.1E) days. This greater difference between abdominal and subcutaneous temperature was likely to reflect greater vasoconstriction in goats that inhabited the transformed site. On hot days the goats remain vasodilated throughout the day and night (Fig. 5.1F), but on cold days goats were vasoconstricted during the cold night (Fig. 5.1E). Shortly after sunrise on cold days there was a distinct rise in the subcutaneous temperature, likely to correspond to a period of vasodilation, which reduced the difference between abdominal and subcutaneous temperature. 154 39 40 41 A A bd o m in al te m pe ra tu re (?C ) B 36 38 40 C Su bc u ta n eo u s te m pe ra tu re (?C ) D 0 1 2 3 E Ab do m in a l-s u bc u ta n eo u s te m pe ra tu re (?C ) F 0 10 20 30 G 0:00 6:00 12:00 18:00 Time of day Pr o po rt io n o f m ax im u m ac tiv ity (% ) H 0:00 6:00 12:00 18:00 Time of day Cold days Hot days Figure 5.1. Nychthemeral rhythm (mean ? SD) of abdominal temperature, subcutaneous temperature and the difference between abdominal and subcutaneous temperature of 12 goats that inhabited the intact site (solid line) and 12 goats that inhabited the transformed site (dotted line), as a function of time of day, over the five coldest days in winter (average minimum globe temperature of 0.4 ? 0.5?C, left panel) and the five hottest days in summer (average maximum globe temperature of 47.7 ? 3.1?C, right panel) of the one- year study period. The lowest panel represents the nychthemeral activity rhythm (mean ? SD) of a single goat that inhabited the transformed site (white bars) and another goat that inhabited the intact site (black bars) over the same periods. The goats displayed a biphasic activity pattern with crepuscular peaks, with less activity between peaks on hot days than on cold days. 155 5.4.3 Shearing Shearing imposed a thermal stress on the goats (Appendix 2, Hetem et al., 2009). After summer shearing, goats that inhabited the transformed site displayed a higher variability of abdominal temperature, as represented by a greater standard deviation of abdominal temperature (0.46 ? 0.06?C versus 0.38 ? 0.04?C, t18 = 3.7, P = 0.002), and had a higher amplitude of abdominal temperature rhythm (1.8 ? 0.2?C versus 1.6 ? 0.1?C, t18 = 2.8, P = 0.01) than did goats inhabiting the intact site (Fig. 5.2A). To assess the rate of abdominal temperature rise, I calculated the rate of rise in abdominal temperature over each four hour interval and analysed the maximum rate. Shorn goats that inhabited the transformed site showed a faster maximum rate of change of abdominal temperature over the 4h interval than did the goats that inhabited the intact site (0.38 ? 0.05?C.h-1 versus 0.31 ? 0.03?C.h-1, t18 = 3.9, P = 0.0009, Fig. 5.2A). Figure 5.1G and 5.1H shows the nychthemeral rhythm of activity for an index goat that inhabited the transformed site and another that inhabited the intact site. On hot days (Fig. 5.1H), the goat, and by inference, others in the herd, displayed a biphasic activity pattern with crepuscular peaks, but there was more activity between the peaks on cold days (Fig. 5.1G), so the goats were active throughout the day. The index goat in the transformed site generally was less active than the index goat in the intact site on both hot (mean activity of 3.9 ? 3.6 versus 5.2 ? 5.6 relative activity units) and cold (4.1 ? 4.2 versus 5.3 ? 5.5 relative activity units) days. 156 38 39 40 A A bd o m in al te m pe ra tu re (?C ) 0 1 2 3 4 C Ab do m in al - su bc u ta n eo u s te m pe ra tu re ( ?? ??C ) 35 37 39 B Su bc u ta n eo u s te m pe ra tu re (?C ) 0 10 20 D 0:00 6:00 12:00 18:00 Time of day Pr o po rt io n o f m ax im u m ac tiv ity (% ) Post-shearing Figure 5.2. Nychthemeral rhythm (mean ? SD) of abdominal temperature, subcutaneous temperature and the difference between abdominal and subcutaneous temperature for 12 goats that inhabited the transformed site (dotted line) and 12 goats that inhabited intact site (solid line) over 10 days after summer shearing. The lowest panel shows the nychthemeral rhythm of activity for a single goat that inhabited the transformed site (white bars) and another goat that inhabited the intact site (black bars) over the same 10 day period. 157 There was no significant difference between subcutaneous temperatures (Fig. 5.2B), nor the difference between abdominal and subcutaneous temperature (Fig. 5.2C), for goats that inhabited the transformed site and goats that inhabited the intact site, after shearing. However, goats in both groups vasoconstricted at night (Fig. 5.2C), which they did not do on hot days when they were not shorn (Fig. 5.1F). Goats in both groups also appeared to be inactive during the night and displayed a reasonably high level of activity throughout the day (Fig. 5.2D), which was similar to that of the unshorn goats on cold days (Fig. 5.1G). In general, the shorn index goat inhabiting the transformed site was less active than was the shorn index goat inhabiting the intact site (mean activity 4.6 ? 2.9 versus 5.6 ? 3.7 relative activity units). To assess differences in microclimate selection, after summer shearing, of goats that inhabited the two land types, I correlated miniglobe temperature at the site chosen by an index goat that inhabited the transformed site, and another index goat that inhabited the intact site with miniglobe temperature recorded at a nearby weather station, for 10 days after shearing (Fig. 5.3). The slope of the regression lines were significantly less than one, both for the goat that inhabited the intact (F1,414 = 223.9, P < 0.0001) and the goat that inhabited the transformed (F1,414 = 73.3, P < 0.0001) site. The regression lines both intersected the line of identity within the range of observations, implying that the goats selected microclimates cooler than the prevailing environmental conditions at high environmental heat loads on the intact site (threshold miniglobe temperature 24?C) and on the transformed site (threshold miniglobe temperature 17?C). The slope of the regression line was significantly lower (F1,414 = 37.7, P < 0.0001) for the goat that inhabited the intact site (0.74 ? 0.02) than that for the goat that inhabited the transformed site (0.89 ? 0.01). The combination of lower slope and higher threshold implied that the goat that inhabited the intact site selected warmer microclimates at low environmental heat loads than did goats that inhabited the transformed site. Thus, Angora goats on the intact site appeared to select more stable microclimates than did goats that inhabited the transformed site, post- shearing. 158 0 10 20 30 40 50 0 10 20 30 40 50 Weather station miniglobe temperature (?C) G o at co lla r m in ig lo be te m pe ra tu re (?C ) Figure 5.3. Scatter diagram showing the relationship between miniglobe temperatures recorded on a single Angora goat that inhabited the transformed site (grey dots and regression line) and another goat that inhabited the intact site (black dots and regression line) plotted against miniglobe temperatures recorded at a nearby weather station, during a 10 day period post-shearing in summer. Measurements were made hourly. The dashed line is the line of identity. Both goats exhibited microclimate selection, with the slope of both regression lines being less than one, the slope of the line of identity. The goat, and, by inference, the herd, that inhabited the intact site selected more stable microclimates as indicated by the lower slope of the regression line (F1,414 = 37.7, P < 0.0001), than did goats that inhabited the transformed site. Over the post-shearing period, I assessed water influx and water turnover rates of goats that inhabited the two sites. Although there was no difference in body mass between the two groups (47.1 ? 3.6 kg versus 48.5 ? 6.5 kg, t22 = 0.68, P = 0.50), goats that inhabited the intact site had a significantly higher initial total body water than did goats that inhabited the transformed site (34.6 ? 4.9 L versus 29.9 ? 2.9 L, t20 = 2.8, P = 0.01). When total body water was expressed relative to body mass, goats that inhabited the intact site had a significantly higher relative total body water than did goats that inhabited the transformed site (77.2 ? 8.2% versus 66.8 ? 8.5%, t20 = 2.9, P = 0.009).On both sites, water influx (F2,40 = 47.98, P < 0.0001) and water turnover rates (F3,51 = 203.1, P < 0.0001) of goats decreased with increasing level of dehydration (Fig. 5.4). Goats in the transformed site had a higher water turnover rate (Fig. 5.4B, F1,17 = 37.1, P < 0.0001) despite their lower water influx (Fig. 5.4A, F1,20 = 6.21, P = 0.02). It therefore appears that goats in the transformed site were more dependent on 159 evaporative cooling to maintain their abdominal temperature in what would appear to be an environment more challenging thermally. 0 1 2 A W at er in flu x ra te (l.d ay - 1 ) day3 day5 day7 0 1 2 3 B Time W at er tu rn o v er ra te (l.d ay - 1 ) Figure 5.4. Water influx (upper panel) and water turnover rates (lower panel) of 12 goats that inhabited the transformed (white bars) and intact (black bars) site, after summer shearing, over a seven-day period during which the goats were denied access to water. Since the goats were denied access to drinking water over the period of the water flux study, water influx could only come from pre-formed dietary water and metabolic water. I hypothesized that the difference in water influx may have arose from a difference in diet, with goats in the intact treatment obtaining a high water influx by feeding on the dominant tree-succulent in their habitat, Portulacaria afra (Milne, 2008). Portulacaria afra has a Crassulacean Acid Metabolism (CAM) photosynthetic pathway during summer and therefore has a carbon isotope ratio of approximately -17.4 (Mills et al., 2005), whereas the C4 grasses approximate -12 and C3 trees approximate -25. Figure 5.5A shows the carbon ratio of the serum of goats that inhabited the two sites. Although the data from goats that inhabited the 160 intact site were consistent with them consuming a higher proportion of P. afra compared to goats that inhabited the transformed site before shearing, the carbon isotope ratios of goats from the two sites appeared to converge after shearing. Since dietary differences, based on carbon isotope ratios, therefore could not account for differences in water influx after shearing, I propose that goats that inhabited the intact site may have had a higher metabolic rate or metabolised a higher proportion of fats. The total plasma lipid concentrations should provide an indication of the proportion of fats metabolised. Figure 5.5B shows the total plasma lipids at various times over the pre- and post-shearing period. Total plasma lipids changed over time (F4,16 = 8.1, P < 0.0009), decreasing with increasing level of dehydration. This effect was likely the result of a greater fat metabolism during dehydration, which would increase the metabolic water contribution. The effect was similar, however, between the goats from the two sites as, overall, total plasma lipids did not differ between the goats that inhabited the two land types (F1,4 = 0.32, P = 0.60), nor was there any statistical interaction in lipid concentrations between the goats that inhabited the two land types across time (F4,16 = 2.4, P = 0.09). One of the blood variables that did turn out to be significantly different between the goats inhabiting the two sites was the albumin to globulin ratio (Fig. 5.5C), with goats inhabiting the intact site showing a lower albumin to globulin ratio than goats inhabiting the transformed site (F1,11 = 6.0, P = 0.03), a difference which was likely to be the result of a slightly higher globulin for goats inhabiting the intact site. Although, in general, there was no difference in serum glucose concentration between goats inhabiting the two sites, immediately after shearing the serum glucose concentrations spiked, and this spike was more pronounced for goats inhabiting the transformed site (Fig. 5.5D, P = 0.04). 161 -18 -16 -14 -12 A shearing d1 3 C (? VP DB ) 100 300 500 B shearing To ta l l ip id s (m g. dl - 1 ) 0.4 0.5 0.6 0.7 0.8 C shearing Al bu m in :G lo bu lin 25 Feb 7 Mar 16 Mar 2.5 5.0 7.5 10.0 12.5 D shearing Time G lu co se (m m o l.l - 1 ) Figure 5.5. Angora goat serum carbon isotope ratios (A), total plasma lipid (B), serum albumin to globulin ratio (C) and serum glucose concentration (D) of goats that inhabited the transformed (open circles) and intact (closed circles) site. The black bars represent periods of water deprivation, and the black arrow represents time of summer shearing. 162 5.5 Discussion I have shown that habitat does influence the thermoregulatory responses of artiodactyls, under conditions of thermal stress. Angora goats that inhabited my transformed site were more water dependent and appeared more susceptible to environmental thermal stresses than were those that inhabited my intact site. Post- shearing, goats that inhabited the transformed site had higher water turnover rates, even though they had a lower water influx and lower total body water, than did goats that inhabited the intact site. The goats that inhabited the transformed site had the potential advantage of greater glucose mobilization, immediately after shearing, and lower albumin to globulin ratio, than goats that inhabited the intact site. Both after summer shearing and on cold days, goats that inhabited the transformed had a higher amplitude of 24-h abdominal temperature rhythm and a faster maximum rate of abdominal temperature rise, than did the goats that inhabited the intact site. In addition, goats that inhabited the transformed site were more vasoconstricted and generally less active than were goats that inhabited the intact site, probably as a result of shade-seeking behaviour when environmental heat loads exceeded 17?C. Although statistically significant, the differences I observed were small and, under most environmental conditions, the Angora goats responded similarly, irrespective of which site they inhabited. The thermoregulatory differences between the goats that inhabited the intact and the transformed sites may well have been anomalously small as a result of the benign conditions which prevailed at the study site during my study. The rainfall during the study was unusually high, 30% higher than the long-term average, and the goats that inhibited the transformed treatment accessed better quality diet items as indexed by faecal lignin content (Milne, 2008). Such an hypothesis is supported by the fact that I found no differences in mass between the goats that inhabited the two sites, and the animals did not exhibit any differences in mohair production (Milne, 2008), so that any physiological differences that I did find were not the result of nutritional stress. Since forage availability varies by orders of magnitude between wet and dry periods (Stuart-Hill and Aucamp, 1993), the 163 thermoregulatory differences that I observed between goats in the two habitats are likely to be exacerbated with further desertification predicted with climate change. 5.5.1 Body temperature The high maximum abdominal temperature of the goats that inhabited the intact site on hot days may be a result of the higher specific dynamic action associated with the high proportion of browse in their diet (Milne, 2008). Similarly, desert goats maintained on a lucerne hay diet have been shown to have higher rectal temperatures than goats maintained on a grass hay diet (Ahmed and El Kheir, 2004). However, since abdominal temperatures of the goats that inhabited the intact site were not consistently higher than those of the goats that inhabited the transformed site, it seems more likely that these differences were the result of differences in the thermal environment between the two sites. Thicket vegetation may restrict wind and create a more humid environment (Henley, 2001), resulting in a thermally more-stressful environment by reducing an animals? access to convective and evaporative cooling on hot days. The higher maximum abdominal temperature of the goats that inhabited the intact site, compared to goats that inhabited the transformed site, on hot days, may reflect those effects. The soil of the transformed habitat reached higher maximum and lower minimum temperatures than did those of the intact thicket (Lechmere- Oertel, 2003), implying more thermal variability in the transformed environment. A variable proposed to be sensitive to environmental variability is blood glucose. Immediately after shearing my goats showed a spike in blood glucose concentrations, either as a result of stress or an increased metabolism (Wentzel et al., 1979), and this increase was more pronounced for the goats that inhabited the transformed site (Fig. 5.5D). Supplemental feeding and the selection for high levels of mohair production are believed to have lead to adrenal insufficiency and an inability to mobilize glucose in Angora goats (Wentzel et al., 1979; Cronj?, 1992). However, Angora goats which are exposed to environments characterized by seasonal fluctuations in nutrient supply, such as my goats that inhabited the transformed site, show a superior ability to mobilize glucose precursors, at the 164 expense of mohair production (Cronj?, 1995). Despite the higher glucose mobilisation of goats that inhabited the transformed site, compared to goats that inhabited the intact site, after summer shearing, such a response was transient and did not result in a difference in length or diameter of hair produced by the goats in the two treatments (Milne, 2008). Goats that inhabited the transformed site displayed higher maximum, greater variability and faster rates of rise of abdominal temperature than did goats that inhabited the intact site on cold days. In other species, the amplitude of daily body temperature rhythm depends on both habitat features and environmental temperatures, with species from more stable temperate habitats displaying smaller amplitude than species from desert and tropical habitats, which are exposed to a much greater variability in environmental temperature (Lovegrove and Heldmaier, 1994; Refinetti, 1999a; Lowe et al., 2001). Although the climatic variability and habitat differences in my study were less extreme, those influences may have played out in the significant differences in the 24-h amplitude of abdominal temperature rhythm that I found between the goats that inhabited the transformed site and those that inhabited the intact site. Despite the habitat effects, the amplitude of abdominal temperature rhythm of my goats remained within the 2?C limits of variability traditionally associated with homeothermy (Bligh and Johnson, 1973). Goats typically regulate their core body temperature within narrow limits (Appleman and Delouche, 1958; Piccione et al., 2007), but Angora goats in South Africa are more labile thermally than are other livestock species in the same environment (Hofmeyr et al., 1965). Such breed differences may result from differences in the insulative properties of the hair coat (Bianca and Kunz, 1978), so that the removal of the insulation by shearing makes the animal more susceptible to changes in environmental conditions, and the presence of a full coat may further stress the animal on hot days. It was around summer shearing that thermoregulatory differences between the goats that inhabited the two land types were most apparent. 165 Shorn Angora goats in poor body condition are more sensitive to changes in the ambient thermal conditions than are goats in good body condition (Fourie, 1984). Similarly, shorn sheep in poor condition show a lower rectal temperature (Morris et al., 1962; Parer, 1963; Hopkins et al., 1978), a greater rate of change and a greater daily amplitude in rectal temperature (Parer, 1963) than do shorn sheep in good condition. These differences in body temperature are similar to those reported here for shorn goats that inhabited the transformed site compared to the intact site, despite there being no evidence for a loss of body condition of goats that inhabited the transformed site. Such differences in the abdominal temperature profile are therefore likely to be the result of differences in environmental conditions between the two sites. 5.5.2 Autonomic thermoregulation While they were fleeced, my goats remained vasodilated over both day and night, as evidenced by small differences between abdominal and subcutaneous temperature (Fig. 5.1F). After summer shearing, however, the goats vasoconstricted at night, and the degree of vasoconstriction (or at least the difference between abdominal and subcutaneous temperature) was greater for goats that inhabited the transformed site than in the goats that inhabited the intact site. Since high stocking densities decrease the availability of pasture and alter pasture composition and structure (McGregor, 1985; McGregor, 1998), and since transformed thicket has a decreased carrying capacity (Stuart-Hill and Aucamp, 1993), one would expect stocking densities to affect vasoconstriction responses in a way similar to site transformation. However, Morris et al. (1962) found winter skin temperature were up to 1.4?C higher in shorn sheep stocked at the high densities than in shorn sheep stocked at low densities. These low subcutaneous temperatures resulted in a difference between rectal and skin temperature of nearly 6?C for sheep stocked at low densities, compared to only 3?C for those stocked at high densities (Morris et al., 1962), with the latter value similar to the difference for my shorn goats on summer nights. Similarly, Sykes and Slee (1969) found skin temperatures of sheep maintained on a low plane of nutrition to be an average 1.8?C higher than sheep maintained on a high plane of nutrition, during 166 cold exposure. These authors attributed the failure of heat conservation in poorly nourished sheep to a loss of insulating fat, thinning of the skin, and reduced vasomotor tone of animals on a low plane of nutrition (Morris et al., 1962; Sykes and Slee, 1969). Since even my goats that inhabited the transformed site were not stressed nutritionally (Milne, 2008), and goats generally have less subcutaneous fat than do sheep (Owen et al., 1977), the greater vasoconstriction of my goats that inhabited the transformed site was likely the result of heat loss to the environment potentially being greater at the transformed site, particularly at night. The reduced vegetation cover of the transformed site could have increased both the goat?s radiative heat loss to the night sky and its convective cooling. Indeed, transformed sites, in general, have greater air movement than do intact thicket vegetation (Henley, 2001). 5.5.3 Activity In addition to the environmental and autonomic influences, activity has been proposed to influence body temperature rhythm (Lowe et al., 2001). However, unlike previous laboratory studies, which found synchronous rhythms in body temperature and activity (Refinetti and Menaker, 1992; Refinetti, 1999b), my index goats displayed a dissociated rhythm and were active throughout most of the day, with peaks of activity in the early morning and late afternoon. Similar patterns of activity have been shown for Angora goats on west Texas range (Askins and Turner, 1972) and may support the conclusion that climatic factors are important determinants affecting the foraging behaviour of goats (Askins and Turner, 1972), as they are with sheep (Dudzinski and Arnold, 1979; Arnold, 1982). Since time spent foraging can be used as an index of pasture quality (Arnold and Dudzinski, 1978), I predicted that goats on the transformed site would have to walk further for an adequate supply of food and water, as found for grazing ruminants in arid areas (Young and Corbett, 1972; Allison, 1985; Manteca and Smith, 1994; Lachica and Aguilera, 2003), thus conforming to the ?time minimizing? model predicted for bison (Bergman et al., 2001). On the contrary, I 167 found that the index goat, and, by inference, the herd, that inhabited the intact site was generally more active than the counterpart that inhabited the transformed site. Stocking rates influence animal performance (Huston et al., 1993a), production (Davies and Southey, 2001), and grazing time and number of steps taken have been shown to increase at high stocking densities of both sheep and goats (Animut et al., 2005b). However, rather than such differences being the result of forage availability, these differences may be the effect of social feeding, with goats in larger groups displaying higher intake rates with an increased number of competitors (Shrader et al., 2007; Van et al., 2007). My study had the advantage of a transformed environment without altering the social structure of goats. The higher activity level of my index goat that inhabited the intact site may be explained by foraging theory, which predicts that animals should feed more selectively when high quality foods are more abundant (Stephens and Krebs, 1986). Sheep increase their forage intake at low levels of energy supplementation (Caton and Dhuyvetter, 1997) and high levels of protein supplementation (Huston et al., 1993a; Huston et al., 1993b). Angora goats fed more selectively on natural forage, thus increasing their foraging effort and movement rates by nearly 70%, when given a high-quality supplement (Murden and Risenhoover, 1993). Conversely, Herselman et al. (1999) found that Angora goats grazed on native pasture for longer and took more steps per unit of time spent grazing than did goats grazed on improved pasture. Such differences in activity may be confounded by vegetation type. The ?native pasture? provided a diet with a larger proportion of browse than the diet of animals on ?improved pasture?. Since goats show a preference for browse (Huston, 1994; Rodriguez Iglesias and Kothmann, 1998; Ngwa et al., 2000; Animut et al., 2005a; Sanon et al., 2007), the higher proportion of grass consumed by both the goats on the ?improved pasture? and my goats that inhabited the transformed site and may have resulted in goats foraging less selectively. Indeed, my goats that inhabited the transformed site consumed grass in relation to its abundance (Milne, 2008), potentially resulting in a lower foraging distance. 168 Since energy expenditure of locomotion and grazing are a significant contributor to the energy requirements of goats in free-living conditions (Lachica and Aguilera, 2005a; Lachica and Aguilera, 2005b), one may have expected that the higher activity level of my goats that inhabited the intact site would have resulted in a higher energy expenditure and heat production (Graham, 1962; Young and Corbett, 1972; Osuji, 1974), compared to the goats that inhabited the transformed site. However, other studies have shown that heart rate, an index of energy expenditure (Puchala et al., 2007), increased linearly with increased stocking rates (Animut et al., 2006). Although I did not measure heat production or metabolic rate directly, since my goats were denied access to water during my assessment of water influx, the higher water influx of my goats that inhabited the intact site could have been the result of a higher metabolic rate. 5.5.4 Water influx The water influx of my goats that inhabited the transformed site was lower that that of the goats that inhabited the intact site. Morris et al. (1962) found that the water intake of sheep grazing at a high densities were nearly half that of sheep grazed at low densities, a difference which they attributed to the low-density sheep eating more food with a higher water content. Similarly, desert goats both acquire more water when maintained on alfalfa hay than on wheat straw (Brosh et al., 1986) and gain more water from preformed and metabolic water when maintained on a lucerne hay diet than on a grass hay diet (Ahmed and El Kheir, 2004). My goats that inhabited the intact site had access to browse of high moisture content, such as P. afra, whereas the goats that inhabited the transformed site had a lower availability of browse and succulents (Milne, 2008), potentially making the goats that inhabited the transformed site more dependent on free water, as found for other grazing ruminants (Maloiy, 1973b; Kay, 1997). The goats that inhabited the transformed site were therefore worse-off than goats that inhabited the intact site when water was taken away. Although the 13C ratio of my goats that inhabited the two sites appeared to converge over the post-shearing period (Fig. 5.5A,), such a change in diet was not the result of an increase in the proportion of succulents consumed, but rather an increase in the proportion of trees consumed 169 by the goats that inhabited the transformed site (Milne, 2008). Since the water content of succulents is higher than that of browse, the higher portion of succulents consumed by goats that inhabited the intact site compared to the transformed site post-shearing, 41% versus 7% (Milne, 2008), may have contributed to their higher water influx. However, differences in diet are not the only variable which influence 13C enrichment. A shift in the relative rates of oxidation of metabolic substances for thermogenesis may change post-shearing (McGraham et al., 1958; Symonds et al., 1986), which has been associated with a change in the background enrichment of 13C (Lachica and Aguilera, 2003). Although the total plasma lipid concentration of my goats changed over time (Fig 5.5B), these changes were not consistent with changes in the background enrichment of 13C. Although the total lipid concentration of the goats that inhabited the transformed site, at times, appeared slightly higher than the goats that inhabited the intact site, in general, these differences were not significant. The water contribution from fat metabolism was more than three times higher in sheep grazed at high stocking densities than at low stocking densities (Morris et al., 1962), which has been attributed to the high fat metabolism associated with mild starvation. Since there is no evidence that my goats were under nutritional stress on either site (Milne, 2008), differences in total plasma lipid concentrations before shearing were likely the result of difference in the amount and type of fat in the diet (Beynen et al., 2000). Since changes in fat metabolism could not account for differences in the 13C enrichment, I propose that the higher water influx of goats that inhabited the intact site were the result of both the higher portion of succulents in their diet and potentially a higher metabolic rate, as indexed by the high level of activity, than goats that inhabited the transformed site. 5.5.5 Water turnover rate Although my goats that inhabited the intact site had higher water influx, they had lower water turnover rates than did the goats that inhabited the transformed site. Morris et al. (1962) estimated summer water turnover rates for Merino sheep 170 stocked at low densities to be 26% greater than those of sheep stocked at high densities. The authors suggested that differences in the composition of the two pastures may have accounted for the difference in water turnover rates, but did not mention what these differences may have been. I believe that the higher water turnover rates of my goats that inhabited the transformed site were the result of differences in the microclimates selected by my goats. Goats which inhabited the transformed site were exposed to higher environmental heat loads than were the goats which inhabited the intact site, particularly during the heat of the day (Fig. 5.3), and would have had less access to microclimate niches. Habitat transformation of the succulent thicket is proposed to have homogenised the land surface, resulting in fewer microhabitats and greater environmental fluctuations (Lechmere-Oertel, 2003). Since maintaining homeothermy during heat stress demands water (Beede and Collier, 1986), and water turnover rates increase with increasing ambient temperature (Maloiy, 1973a), I propose that my goats that inhabited the transformed site were more reliant on evaporative cooling to dissipate their heat load, resulting in water turnover rates higher than those of the goats that inhabited the intact site. 5.5.6 Parasite load Although the transformed site appeared to be a more stressful environment than the intact site, both in terms of thermal microclimates and nutrient availability, one potential advantage of the more open habitat is a reduced parasite load (Mooring et al., 2004). My goats that inhabited the transformed site had higher albumin globulin ratio (Fig. 5.5C), resulting from lower globulin concentrations, than goats that inhabited the intact site. Although, a low albumin globulin ratio can result from a protein deficiency (Anderson et al., 1962), such a deduction is unlikely since there was no difference in total serum protein concentration between the goats that inhabited the two sites (unpublished data). A low albumin to globulin ratio is also associated with inflammation and bacterial invasion (Hurwitz and Whipple, 1917), decreased immune function in heifers (Piccinini et al., 2004) and low social status in pigs (Hicks et al., 1998). I propose that the 171 goats that inhabited the transformed site were less immunologically stressed than the goats that inhabited the intact site. 5.5.7 Conclusion Despite the potential advantage of a high albumin to globulin ratio and greater mobilisation of glucose, goats that inhabited the transformed site were more stressed thermally and more water dependent, than were goats that inhabited the intact site. I concede that the differences the differences that I observed were small. The year of my study had above-average rainfall year and even goats on the transformed site maintained good body condition. I expect that, in years of average, and particularly low, rainfall, the same trends will prevail but will have a much greater magnitude. Nevertheless, my colleagues and I are the first to demonstrate physiological changes in response to desertification, despite the relatively benign conditions of high rainfall prevalent during my study. Given that climate change is already exacerbating drought extremes (Chamaill?-Jammes et al., 2007) and that current models predict a further 20-30% transformation of the succulent thicket within the next 20 years (Robertson and Palmer, 2002; Rouget et al., 2003), understanding the effects of these extremes will become increasingly important. The long term viability and sustainability of agricultural herbivores will depend on maintaining appropriate stocking rates under changing climatic conditions (Richardson et al., 2005). The use of sophisticated physiological measurements and biologging devices, as I have demonstrated in my goats, will allow a better understanding of phenotypic plasticity of herbivores to both the direct thermal effects of climate change and the consequences of changes in habitat. 172 5.6 Acknowledgements: I gratefully acknowledge Arthur and Trinette Rudman for their hospitality and for allowing the study to take place on their farm Blaauwkrantz, and Stephan Woodborne at the CSIR for use of the high temperature elemental analyser for the determination of deuterium concentrations, Ms Elsbe Myburgh at the Department of Companion Animal Clinical Studies Section of Clinical Pathology, University of Pretoria, for serum total protein, glucose and osmolality analysis and Prof Bruce Davidson for taking time to teaching me the lipid extraction technique. 173 ___________________________________________________________ CHAPTER 6 ___________________________________________________________ 6 Adapting to the spread of pathogens: fever and sickness behaviour during an opportunistic infection in a free-living antelope, the greater kudu (Tragelaphus strepsiceros) Data and ideas presented in this chapter have been published: Hetem R.S., Mitchell D., Maloney S.K., Meyer L.C.R., Fick L.G., Kerley G.I.H. and Fuller A. (2008) Fever and sickness behavior during an opportunistic infection in a free-living antelope, the greater kudu (Tragelaphus strepsiceros). American Journal of Physiology - Regulatory, Integrative and Comparative Physiology 294: R246?R254. 174 6.1 Abstract To study their thermal responses to climatic stress, my colleagues and I implanted seven greater kudu (Tragelaphus strepsiceros) with intra-abdominal, brain, carotid and subcutaneous temperature data loggers as well as an activity logger. Each animal also was equipped with a collar holding a miniature black globe thermometer, which I used to assess thermoregulatory behaviour. The kudu ranged freely within succulent thicket vegetation of the Eastern Cape Province, South Africa. The kudu spontaneously developed a bacterial pneumonia, and consequent fever that lasted between six and ten days. The fever was characterized by a significant increase in mean 24-h abdominal temperature from 38.9 ? 0.2?C to 40.2 ? 0.4?C (mean ? SD, t6 = 11.01, P < 0.0001), although the amplitude of body temperature rhythm remained unchanged (t6 = 1.18, P = 0.28). Six of the kudu chose warmer microclimates during the fever than when afebrile (P < 0.0001). Despite the selection of a warmer environment, on the first day of fever, the abdominal-subcutaneous temperature difference was significantly higher than on afebrile days (t5 = 3.06, P = 0.03), indicating vasoconstriction. Some kudu displayed increased frequency of selective brain cooling during the fever, which would have inhibited evaporative heat loss and increased febrile body temperatures, without increasing the metabolic maintenance costs of high body temperatures. Average daily activity during the fever decreased to 60% of afebrile activity (t6 = 3.46, P = 0.014). I therefore have recorded quantitative evidence for autonomic and behavioural fever, as well as sickness behaviour, in the form of decreased activity, in a free-living ungulate species, and demonstrate the kinds of changes in physiological function that sick animals will display as they face the other consequences of climate change. 175 6.2 Introduction Since climate change is predicted to bring emergent pathogens, particularly with the relocation of arthropod vectors (Rogers and Randolph, 2000; Olwoch et al., 2003; Cumming and van Vuuren, 2006), we need to understand the mechanisms employed by free-ranging antelope in the face of infection. Animals employ a suite of autonomic and behavioural mechanisms during an infection. These mechanisms are regulated by pro-inflammatory cytokines which act centrally to inhibit the firing rate of warm-sensitive neurons in the hypothalamus, resulting in a rise in the thermoregulatory set point (Banet, 1979; Kluger, 1991; Aubert, 1999). In addition to an increased body temperature during infection, animals consistently exhibit symptoms such as lethargy, reduced engagement in social activities and reduced food intake (Aubert, 1999; Dantzer, 2004), collectively referred to as sickness behaviour. Sickness behaviour originally was thought to be the result of a weakened physiological state, but now is viewed as an adaptive reorganization of the host?s priorities to facilitate recovery from an infection (Ewald, 1980; Hart, 1988; Johnson, 2002; Konsman et al., 2002). Indeed, given their ubiquity among both endothermic and ectothermic vertebrates, both fever and sickness behaviour are postulated to offer a survival benefit to the host (Kluger, 1978; Kluger, 1979; Kluger, 1986; Wingfield, 2003). Although both fever and sickness behaviour have been studied extensively in laboratory-housed animals given purified pyrogens experimentally, little is known about how free-living animals respond to opportunistic infection. Unlike naturally- occurring fevers, which last several days at least, most experimentally-induced fevers last less than 24 h and are not superimposed on the nychthemeral rhythm of body temperature (Kluger, 1991; Mphahlele et al., 2004). However, in free-living impala, prolonged natural fevers did not disrupt the nychthemeral rhythm of body temperature (Kamerman et al., 2001). But neither these antelope nor laboratory- housed goats with chronic fevers displayed any obvious signs of sickness behaviour 176 (Kamerman et al., 2001; Mphahlele et al., 2004). It is therefore unknown whether sickness behaviour is a component of chronic fever in free-living animals. Laboratory-housed animals, although they may be physically active, often do not have access to the full range of natural behaviours, thermoregulatory or social. Changes in behaviour in laboratory-housed animals during fever, therefore, may not be similar to behavioural responses of a free-living animal, since wild animals would have to sacrifice parts of their usual behavioural repertoire if they employ sickness behaviour. For example, both free-living, male, north-western song sparrows and white-crowned sparrows decreased their territorial aggressive behaviour and song when exposed to an exogenous pathogen (Owen-Ashley and Wingfield, 2006; Owen- Ashley et al., 2006). Such behavioural changes may be costly for wild animals, particularly if they result in an increased susceptibility to predators, decreased reproduction and parental care, loss of social position or removal from territories (Hart, 1988; Yirmiya et al., 1995; Aubert et al., 1997; Konsman et al., 2002; Owen- Ashley and Wingfield, 2006). Sickness behaviour and immune responses to infection also are metabolically costly (Hart, 1988; Kluger et al., 1998; Lochmiller and Deerenberg, 2000). Such metabolic costs may be maladaptive, particularly during times of low energy availability. Weight loss following exogenous pyrogen administration in captive male white-crowned sparrows was greater when they had high energy reserves than when they had lower reserves (Owen-Ashley et al., 2006). Free-living animals therefore have to trade off any potential benefit of sickness behaviour, in supporting metabolic and physiological changes associated with infection, against associated energetic and social costs. It is difficult to study fever and sickness behaviour in wild animals as sickness occurs opportunistically and observations of behaviour usually require human presence, which in turn may disrupt normal behaviour. Serendipitously for us, a herd of free- living antelope (kudu, Tragelaphus strepsiceros), which my colleagues and I had instrumented with temperature and activity data loggers for other purposes, acquired 177 a spontaneous infection, which lasted between six and ten days, two to five weeks after implantation of the loggers. I obtained remote and continuous measurements of temperatures at various body sites, including the brain, carotid artery, abdomen and beneath the skin, both before and during the infection. I was able later to extract quantitative data reflecting the kudu?s autonomic and behavioural responses to the infection. I show that the free-living kudu exhibited sickness behaviour and implemented a suite of coordinated thermoregulatory responses that elevated body temperature during a spontaneous fever. 6.3 Materials and methods 6.3.1 Animals and habitat The experiment took place between December 2004 and March 2005 (austral summer) at Blaauwkrantz farm (33?32?S 25?23?E), near Port Elizabeth, South Africa. The vegetation in the area has been classified as Sundays Spekboomveld (Vlok et al., 2003), and the habitat is both historical and current habitat for greater kudu (Tragelaphus strepsiceros, Skinner and Chimimba, 2005). Seven adult female free-living kudu (body mass 121 ? 8 kg) were captured from a large camp on Blaauwkrantz farm in November 2004. The kudu were immobilized from a helicopter, by a professional capture team under veterinary supervision, and given a long acting tranquilizer, zuclopenthixol acetate (100 mg I.M. Clopixol Acuphase, H. Lundbeck (Pty) Ltd., Randburg, South Africa), before being transported to nearby holding pens at River Bend Lodge (~ 20-min drive). The kudu were housed in these pens for a two-week habituation period to reduce potential peri- operative stress. All experimental procedures were approved by the Animal Ethics Screening Committee of the University of the Witwatersrand (protocol no. 2004/84/4). 178 6.3.2 Surgery For surgery to implant the data loggers, each animal was immobilized with etorphine hydrochloride (6 mg intramuscularly (I.M.), M99, Novartis, Johannesburg, South Africa), together with azaperone (100 mg I.M., Stresnil, Kyron Laboratories, Johannesburg, South Africa) and transported to a veterinary surgery within 200 m of the pens. At the surgery, the animals were placed in sternal recumbency, supported by sandbags, with their heads elevated. Anaesthesia was maintained with 1 - 3% halothane (Fluothane, Zeneca, Johannesburg, South Africa), administered in oxygen via a face mask. Once an adequate plane of anaesthesia was established, the effect of the opioid, etorphine, was reversed using diprenorphine hydrochloride (15 mg intravenously (I.V.), M5050, Novartis, Johannesburg, South Africa). Respiratory rate, heart rate, arterial oxygen saturation (pulse oximeters, Nonin 9847V, Nonin Medical, North Plymouth, USA) and rectal temperature (thermocouple thermometer, BAT-12, Physitemp, Clifton, USA) were monitored throughout the surgery lasting ~ 2 h. Under sterile surgical conditions, a suite of miniature data loggers was implanted. All data loggers were covered in an inert wax (Sasol, Johannesburg, South Africa) and dry-sterilized in formaldehyde vapour before implantation. Incision sites were shaved and sterilized with chlorhexidine gluconate (Hibitane, Zeneca, Johannesburg, South Africa). Each animal was fitted with data loggers connected to thermistor sensors for temperature measurement in the carotid artery and brain. A thermistor, inserted in a blind-ended and thin-walled polytetrafluoroethylene (PTFE) tube (o.d 1.35 mm, i.d. 0.97 mm; Straight Aortic Flush 4F Catheter, Cordis, The Netherlands), was advanced 80 mm into the left common carotid artery towards the heart, at a position midway along the length of the neck, and secured in position with a purse-string suture in the artery wall. Outside the artery, the PTFE tube was sealed on a PTFE-coated co-axial cable (150 mm long, o.d. 3 mm, Belden, Richmond, USA) connecting the thermistor to the temperature-sensitive data logger (see below). The data logger was positioned subcutaneously, dorsal to the artery. A second data logger, connected to the thermistor positioned in the brain, was positioned subcutaneously caudal to the base 179 of the left ear. Its PTFE-coated cable was advanced subcutaneously over the skull, where it was connected to a head plate and guide tube. The guide tube, constructed from cellulose acetate butyrate tubing (o.d. 3.2mm, i.d. 1.6 mm; World Precision Instruments, Sarasota, Fla., USA) sealed at the tip by a steel cap, was 58 mm long and was inserted through a small 2 mm diameter hole, which was drilled through the cranium, at appropriate co-ordinates, pre-determined from head sections of dead kudu of similar size, so that the probe tip would be positioned near the hypothalamus. The brain guide tube was connected to a subcutaneously implanted plastic head plate (20 ? 10 ? 5 mm) that was secured to the skull by two bone screws. A third data logger, with an internal temperature sensor, was inserted, via an incision in the paralumbar fossa, into the abdominal cavity. The muscle layer was sutured closed and a smaller temperature-sensitive data logger (see below) was inserted subcutaneously before the skin was sutured closed. An additional incision was made on the upper hind limb, where an activity logger (Actical, Mini-Mitter Corporation, Bend, OR, USA) was implanted subcutaneously. The activity logger recorded at 10- min intervals, had dimensions of 40 x 40 x 15 mm, and weighed ~ 40 g. Wounds were treated with a topical antiseptic spray (Necrospray, Centaur Labs, Johannesburg, South Africa) and coated with a tick repellent grease (cypermethrin 0.025% m/m, Bayer Animal Health Pty, Isando, South Africa). Each of the kudu received a long-acting antibiotic (22 ml I.M., penicillin, Peni La Phenix, Virbac Animal Health, South Africa), a non-steroidal anti-inflammatory analgesic (5 ml I.M., phenylbutazone, Phenylarthrite injectable solution, Bayer (Pty) Ltd. Animal Health, Isando, South Africa), a long-acting parasiticide (2.5 ml subcutaneously (S.C.), doramectin, Dectomax?, Pfizer Laboratories, Sandton, South Africa), Vitamin E and Selenium (5 ml I.M., vitESe injectable solution, Kyron Laboratories, South Africa), and perphenazine (100 mg I.M., Kyron Laboratories, South Africa). 180 Before halothane administration was terminated, a neck collar (African Wildlife Tracking, Pretoria, South Africa) was fitted to each kudu. In addition to a tracking radio transmitter, each collar supported a miniature black globe thermometer (?miniglobe?), to allow for the dynamic measurement of the microclimate that the kudu chose to occupy. This technique previously has proven successful on other ungulate species (Appendix 1, Hetem et al., 2007). Miniglobe temperature was measured with a thermistor inserted into the centre of a matt-black hollow bronze sphere (30 mm diameter, Press Spinning & Stamping Co., Cape Town, South Africa). The thermistor leads were housed within a spring that was soldered to the miniglobe and filled with flowable silicon and covered in heat-shrink tubing to add flexibility and strength. The base of the spring was attached to the collar so that the miniglobe stood 20 mm above the collar. The miniglobe temperatures were recorded by a temperature-sensitive data logger (see below), which was attached to the collar and waterproofed with dental acrylic. A weight on the ventral side of the collar ensured that the miniglobe remained over the dorsum of the neck and could not be shaded by the animal?s body. Following surgery, the kudu were transported back to their pens, where they became ambulatory within ~ 10 min. After a three-day recovery period, and following veterinary inspection, they were transported back to Blaauwkrantz farm where they were released into a 50 ha fenced enclosure with natural forage and water available ad libitum. Two to five weeks post-release, the kudu spontaneously developed a lethal pneumonia, and a consequent fever that lasted between six and ten days before the kudu died. Unfortunately, my experimental design, which required that the animals be disturbed by humans as little as possible, and the data logging technology did not allow us to determine that the kudu were febrile until after they had died. Where possible I conducted gross-macroscopic and histopathological post-mortem examinations. These revealed a severe necrotizing bronchopneumonia and emphysema, with extensive neutrophil infiltrates and bacterial colonies. While it is known that wild animals do not inhabit sterile environments (Gannon et al., 2007), 181 the deaths were unexpected; my colleagues have performed similar procedures on a variety of mammalian species without ill effects. However, since there are no reports of recovery from surgery in wild kudu, I do not know whether the species is particularly susceptible to delayed post-operative infection, or whether they were the victims of an opportunistic infection acquired at the study site. 6.3.3 Temperature measurements The miniature thermometric data loggers (StowAway XTI, Onset Computer, Pocasset, Massachusetts, USA), used to measure brain, carotid artery and abdominal temperature, had outside dimensions of ~ 50 ? 45 ? 20 mm and a mass of ~ 40 g when covered in wax. These loggers had a resolution of 0.04?C and measurement range from + 34 to + 46?C. Temperature sensors used to measure brain and carotid blood temperatures were constructed from ruggedized glass-coated bead thermistors with insulated extension leads (bead diameter 0.3 mm; AB0E3-BR11KA103N, Thermometrics, Edison, New Jersey, USA). The scan interval of the brain and carotid blood loggers was set at 5 min and that of the abdominal logger was set at 20 min. Subcutaneous temperatures were recorded every hour using a smaller thermometric data logger (iButton DS1922T, Maxim, Dallas Semiconductor, Texas, USA), which had a radius of ~ 25 mm, a height of 15 mm, and weighed about 10 g when covered with wax. These loggers had a resolution of 0.5?C and a measurement range from 0 to 125?C. The temperature of each miniglobe was measured with an uncoated bead thermistor (bead diameter 1.5 mm; 27-10K4A801 Onset Computer Corporation, Pocasset, Mass., USA). The thermistor was connected to a miniature thermometric data logger (Hobo U12-013, Onset Computer Corporation, Pocasset, Mass., USA), with a temperature range of - 20 to + 70?C and an intrinsic resolution of 0.35?C. Miniglobe temperatures were recorded instantaneously every 15 min. 182 All temperature sensors and loggers were calibrated against a high-accuracy thermometer (Quat 100, Heraeus, Hanau, Germany) in an insulated water bath. After calibration, the loggers and their sensors measured blood, brain and abdominal temperature to an accuracy of better than 0.05?C, subcutaneous temperature to better than 0.5?C and miniglobe temperature to better than 0.4?C. 6.3.4 Meteorological data measurements I collected climatic data from a portable weather station in the enclosure into which the kudu were released (Hobo Weather Station, Onset Computer Corporation, Pocasset, Mass., USA). I monitored wind speed (m.s-1), solar radiation (W.m-2), dry- bulb temperature (?C), relative humidity (%) and standard (150 mm diameter) black globe temperature (?C) for the duration of the study period. 6.3.5 Data analysis I considered the kudu to be febrile if their mean daily body temperature was at least 0.5?C above normal. Paired Student?s t-tests were used to confirm differences in body temperature between febrile and afebrile states. I obtained a complete set of brain temperature and carotid blood temperature data for four febrile kudu for analysis of selective brain cooling. The relationship between brain temperature and carotid blood temperature in each animal was analyzed by sorting all 5-min measurements of carotid blood temperature into 0.1?C classes, and determining the mean, standard deviation, maximum and minimum brain temperature at each class of carotid blood temperature. After converting miniglobe temperature to their equivalent standard black globe temperatures (Appendix 1, Hetem et al., 2007), I correlated these converted animal miniglobe temperatures against weather station standard black globe temperatures using linear Pearson procedures. To assess whether the animals were conforming to ambient conditions or were selecting micro-environments, I tested whether the slope 183 of the regression equation was significantly different from one (the slope of the line of identity). In addition, I tested whether the slope and elevation of the regression equation was significantly different between febrile and afebrile states using an analysis of co-variance (ANCOVA). To compare changes in activity levels during the fever, I calculated the febrile activity levels as a proportion of afebrile activity levels and compared these to a value of one using a Paired Student?s t-test. I used GraphPad Prism (version 4.00 for Windows, GraphPad Software, San Diego California USA) for statistical analyses. Values are expressed as mean ? SD. 6.4 Results 6.4.1 Climate Over the six-week study period during which the kudu were both febrile and afebrile, average dry-bulb temperature, standard black globe temperature, wind speed and solar radiation varied as a function of time of day (Fig. 6.1). Average dry-bulb temperature over 24 h was 22.5 ? 2.4?C with a mean daily minimum of 16.8 ? 2.2?C and mean daily maximum of 30.4 ? 5.4?C. Standard black globe temperature was 25.0 ? 3.1?C on average, reaching a mean daily minimum of 16.7 ? 2.3?C and a mean daily maximum of 36.5 ? 6.0?C (Fig. 6.1A). Solar radiation showed a bell-shaped distribution, with a mean peak of ~ 750 W.m-2 around solar noon, with variability resulting from a combination of cloudy and clear days. Wind speed increased in the late afternoon, reaching a peak of about 4 m.s-1 around 16:00 (Fig. 6.1B). Rainfall averaged 24 mm per month in 2004, which was slightly lower than the 50 year rainfall average of 26 mm per month. However, prevailing weather conditions were not unusual during the study period. 184 15 25 35 45 dry-bulb air standard black globe A Te m pe ra tu re (?C ) 0:00 6:00 12:00 18:00 0 250 500 750 1000 0 2 4 6 radiation wind speed B Time of day Ra di at io n (W . m - 2 ) W in d sp ee d (m . s - 1 ) Figure 6.1. Field site weather station data, showing the standard black globe (black line) and dry-bulb air (grey line) temperatures (A), as well as radiation (black line) and wind speed (grey line) (B), as a function of time of day, as mean ? SD for the six-week study period. 6.4.2 Fever All seven kudu developed a spontaneous fever two to five weeks after surgery. These fevers lasted between six and ten days and were characterized by an upward displacement of the nychthemeral rhythm of body temperature (Fig. 6.2). The mean 24- h abdominal temperature, for the seven animals, increased significantly from 38.9 ? 0.2?C when the kudu were afebrile to 40.2 ? 0.4?C (t6 = 11.01, P < 0.0001) when they were febrile (Fig. 6.3). Minimum daily abdominal temperatures increased from 38.0 ? 0.4?C to 39.2 ? 0.5?C (t6 = 5.97, P = 0.001) and maximum daily abdominal temperatures from 39.7 ? 0.3?C to 41.2 ? 0.4?C (t6 = 19.02, P < 0.0001) during the fever. However, the amplitude of nychthemeral rhythm remained unchanged at 1.7 ? 0.5?C when the animals were afebrile and 1.9 ? 0.6?C during fever (t6 = 1.18, P = 0.28). 185 0 5 10 37 39 41 Days B o dy te m pe ra tu re (?C ) febrileafebrile Figure 6.2. Abdominal temperature for a single free-living female kudu (kudu 6) over 11 days, including 5 days of fever. The bars at the base of the figure represent the periods I considered the kudu to be ?afebrile? and ?febrile?. Tick marks on the time axis represent midnight. 0:00 6:00 12:00 18:00 37 39 41 afebrile febrile Time of day Te m pe ra tu re (?C ) Figure 6.3. Mean ? SD of the nychthemeral rhythm of abdominal temperature for seven female free-living kudu during febrile (grey line) and afebrile (black line) states. 186 During febrile days, the upward shift in the nychthemeral rhythm of body temperature appeared to be associated with a change in the thermal environment that the kudu chose to occupy. Figure 6.4 shows the correlation between the standard black globe temperatures at the sites chosen by a single animal (kudu 5), converted from miniglobe temperatures recorded on the collar, and standard black globe temperatures recorded at a nearby weather station, during both febrile and afebrile states in kudu 5. The slopes for all the animals were significantly less than one (ANCOVA, P < 0.0001), and the regression lines intersected the line of identity, implying that kudu selected microclimates cooler than the prevailing environmental conditions at higher environmental heat loads and microclimates warmer than the prevailing environment conditions at lower environmental heat loads. Six of the seven kudu showed no difference in the slope of the regression lines between febrile and afebrile states (ANCOVA, P > 0.17). The remaining animal (kudu 3) showed a significantly increased slope (F1,812 = 4.8, P = 0.03) under the febrile state (0.41 ? 0.04) compared to the afebrile state (0.32 ? 0.01), implying that this kudu selected warmer microclimates when febrile compared to afebrile, particularly at high environmental heat loads. One animal (kudu 7) showed no change in the slope or elevation of the regression line between the febrile and afebrile state. Five of the other six kudu, which showed no difference in the slope during afebrile and febrile states, showed an average elevation in points during febrile states, with the y-intercept being significantly (ANCOVA, P < 0.0001) higher in the febrile state (24.0 ? 3.5?C, n = 5) than in the afebrile state (22.8 ? 4.1?C, n = 5). These results imply that these five kudu were selecting warmer microclimates during their fevers, irrespective of prevailing ambient temperatures. 187 10 20 30 40 50 10 20 30 40 50 afebrile febrile Weather station globe temperature (?C) Co lla r gl o be te m pe ra tu re ( ?? ??C ) Figure 6.4. Scatter diagram showing the relationship between standard black globe temperatures at the site chosen by a single animal (kudu 5), converted from the miniglobe temperatures recorded on the collar, and standard black globe temperatures recorded at a nearby weather station, during febrile (grey) and afebrile (black) states, for a single kudu. Measurements were made at intervals of 15 min. Miniglobe temperatures were converted to equivalent standard black globe temperatures using an algorithm based on heat-transfer equations. The dashed line is the line of identity, the solid black line is the linear regression during afebrile states and the solid grey line is the linear regression during febrile states. The y-intercept of the line fitted to data obtained during fever was significantly higher (F1,693 = 25.58, P < 0.0001) than that of the line fitted to data from afebrile states. Febrile data: y = 0.31x + 24.1. Afebrile data: y = 0.31x + 23.8. Despite the selection of warmer environments in the febrile state, there was a larger temperature difference between the abdominal cavity and subcutaneous tissue, indicating a more vasoconstricted periphery, at all times of day (t5 = 2.74, P = 0.04), compared to the afebrile state. The largest temperature differences between the abdominal cavity and subcutaneous tissue occurred on the first day of the fever. The abdominal-subcutaneous temperature differences were significantly higher (t5 = 3.06, P = 0.03) on the first day of fever (1.8 ? 0.9?C) than during the afebrile state (1.3 ? 0.4?C). The average 24-h differences between abdominal and subcutaneous temperatures for the first day of fever, and for an average afebrile day, for all seven kudu, are represented in Figure 6.5. 188 0:00 6:00 12:00 18:00 0 2 4 febrile afebrile Time of day A bd o m in al - su bc u ta n eo u s te m pe ra tu re di ff er en ce (?C ) Figure 6.5. Difference between abdominal and subcutaneous temperature over the first day of fever (grey line), and for an average day of non-febrile states (black line), as a function of time of day for seven kudu (mean ? SD). 6.4.3 Sickness behaviour In addition to peripheral vasoconstriction and selecting warmer microclimates when they were febrile, all the kudu displayed sickness behaviour in the form of decreased activity. Figure 6.6 illustrates the average 24-h activity pattern, as detected by movement of the upper hind limb, for a single kudu (kudu 2). Although all the kudu maintained a biphasic activity pattern with crepuscular peaks throughout both febrile and afebrile states, they showed a decrease in activity at all times of day when febrile. On average, daily activity of the seven kudu decreased by 40% during febrile compared to afebrile states (t6 = 3.46, P = 0.01). 189 0 6 12 18 0 10 20 afebrile febrile Time of day Pr o po rt io n o f m ax im u m ac tiv ity (% ) Figure 6.6. Nychthemeral activity rhythm of a single kudu (kudu 2) during febrile (white bars) and afebrile (black bars) states (mean ? SD). The kudu maintained a biphasic activity pattern with crepuscular peaks throughout their fevers, but with reduced amplitude. Activity was measured on the upper hind limb with a subcutaneous activity data logger, as activity counts as a percentage of maximum counts for that logger. 6.4.4 Selective brain cooling Selective brain cooling is defined as a brain temperature lower than arterial blood temperature (IUPS Thermal Commission, 2003). Figure 6.7 shows the typical pattern of brain and carotid blood temperature recorded for a single kudu (kudu 3) over four days before fever was detected and for four days of fever. This kudu showed the largest magnitude of selective brain cooling at the highest carotid blood temperatures, which were recorded at the acrophase of the endogenous nychthemeral body temperature rhythm during fever (Fig. 6.7). 190 37 38 39 40 afebrile carotid blood brain 1 2 3 4 37 39 41 carotid blood brain febrile Days Te m pe ra tu re ( ?? ??C ) Figure 6.7. The pattern of brain (black line) and carotid blood (grey line) temperatures of a single kudu (kudu 3) over a four-day febrile period, and for the same kudu over a four-day afebrile period. I obtained complete records of brain and carotid blood temperatures for four animals. Figure 6.8 shows mean, SD, minimum and maximum hypothalamic temperature for each 0.1?C class of carotid blood temperature during both febrile and afebrile states, for those four kudu. The dotted line represents the line of identity; points below this line reflect selective brain cooling. Three of these four kudu displayed selective brain cooling during both the afebrile and the febrile states, and one animal never displayed selective brain cooling (Fig. 6.8, kudu 1). The threshold for selective brain cooling, defined as the carotid blood temperature at which carotid blood and mean hypothalamic temperatures are equal, was not statistically different (t2 = 1.30, P = 0.32) during afebrile (38.8 ? 0.12?C) and febrile (39.3 ? 0.68?C) states. However, 191 kudu increased the frequency with which they employed selective brain cooling during the fever (t2 = 4.6, P = 0.04), spending 84.4 ? 7.9% of their time with the hypothalamus cooler than carotid blood during the febrile state compared to only 60.5 ? 11.3% during the afebrile state. This higher frequency of selective brain cooling resulted in significantly higher mean selective brain cooling (t4 = 3.2, P = 0.03), as defined by the difference between carotid blood and hypothalamic temperature, during febrile (0.38 ? 0.14?C) compared to afebrile (0.10 ? 0.07?C) states. Although selective brain cooling was used more often during fever, the mean brain temperature was higher for a given blood temperature when febrile than when afebrile. The slope of selective brain cooling, defined as the difference between carotid blood and hypothalamic temperature as a function of carotid blood temperature (Kuhnen and Jessen, 1991), was significantly reduced in febrile compared to afebrile states for each of the individuals (P < 0.0001). In addition, the mean slope of lines of regression of brain temperature on carotid blood temperature, above the threshold for selective brain cooling, was lower in afebrile (0.49 ? 0.05) than in febrile (0.63 ? 0.07, t4 = 2.7, P = 0.05) states, indicating that the magnitude of selective brain cooling at any given carotid temperature was reduced during the febrile compared to the afebrile state. 192 37 39 41 kudu 3 H yp o th al am ic te m pe ra tu re (?C ) 37 39 41 kudu 1 37 39 41 kudu 2 37 38 39 40 41 42 37 39 41 37 39 41 kudu 4 Arterial blood temperature (?C) 37 39 41 Afebrile Febrile Figure 6.8. The mean, minimum, maximum and standard deviation of hypothalamic temperature for each 0.1?C class of carotid blood temperature during both febrile and afebrile states for four kudu. The dotted line represents the line of identity; points below this line reflect selective brain cooling. Kudu which display selective brain cooling when afebrile also did so when febrile, with no apparent shift of threshold temperature at which brain cooling was initiated. 193 6.5 Discussion My results provide the first demonstration, in free-living animals, of certain measurable physiological processes that bring about the elevation of body temperature that is characteristic of fever. I also present the first quantitative evidence for sickness behaviour in free-living animals. I could detect these responses because my colleagues and I had instrumented free-living kudu with data loggers capable of producing a more complete continuous record of the thermal status of undisturbed free-living animals than has been achieved previously. The data loggers, which my colleagues and I implanted in the abdomen, revealed that my kudu had an average 1.3?C increase in mean abdominal temperature and reached maximum temperatures ranging between 40.8?C and 41.6?C while febrile. The elevated body temperature was achieved partially through appropriate thermoregulatory behaviour, which is less metabolically costly than is implementing autonomic mechanisms. I was able to detect such behaviour, without observers present, by using miniature globe thermometers attached to collars, a novel technique I developed (Appendix 1, Hetem et al., 2007) for recording microclimate selection by animals. Irrespective of climatic conditions, or time of day, the kudu selected warmer microclimates when they were febrile (Fig. 6.4) than when they were afebrile. There was a parallel shift in the linear regression lines of the relationship between the selected microclimate and the prevailing ambient conditions, consistent with a change in set-point for behavioural thermoregulation. My technique of measuring selected microclimate also revealed that the kudu used thermoregulatory behaviour, in the form of choice of microclimate, when they were both febrile and afebrile; the slope of those regression lines were less than one, showing that the kudu selected appropriate microclimates to reduce the impact of environmental thermal stress to less than a third of what it would have been without employing thermoregulatory behaviour, over the standard black globe temperature range of 10?C to 50?C. 194 If behavioural processes alone are insufficient to increase body temperature to the elevated set-point during fever, autonomic processes are invoked. Peripheral vasoconstriction provides the least metabolically costly autonomic response in a cool environment. Using the difference between abdominal and subcutaneous temperatures as an index of peripheral blood flow, I provide evidence that the kudu implemented peripheral vasoconstriction during fever. Irrespective of the time of day, the difference between abdominal and subcutaneous temperature was higher when the kudu were febrile, particularly on the first day of the fever (Fig. 6.5), than when they were afebrile. The kudu, when febrile, therefore appeared to be more vasoconstricted at all times of day, even though standard black globe temperature reached 35?C, on average, around solar noon (Fig. 6.1A). An additional autonomic response which may be invoked to supplement the febrile rise in body temperature is the inhibition of evaporative heat loss (Stitt, 1973). Although I had no means of measuring evaporative heat loss continuously in the free- living animals, I do have indirect evidence that my kudu indeed did inhibit evaporation when febrile. At least three of the animals implemented selective brain cooling more frequently, and displayed a greater mean selective brain cooling, when they were febrile (Fig. 6.8). By lowering brain temperature below carotid blood temperature, selective brain cooling reduces the stimulus from hypothalamic warm receptors for evaporative heat loss. Research emanating from my research group (Mitchell et al., 2002; Fuller et al., 2007) and other laboratories (Kuhnen and Jessen, 1992; Kuhnen, 1997; Robertshaw, 2006) concluded that the main thermoregulatory function served by selective brain cooling is inhibition of evaporative heat loss. Even though the magnitude of selective brain cooling at any given carotid blood temperature decreased during fever, kudu spent more time using selective brain cooling. These results indicate that the control of selective brain cooling was altered during fever, in a manner which would suppress evaporative water loss and hence evaporative heat loss. 195 In response to the increase in set-point that occurs at the start of fever, animals increase body temperature by inhibiting evaporative heat loss, instituting peripheral vasoconstriction and selecting warmer microclimates. Once that set-point is reached, body temperature is regulated actively around the new set-point and the nychthemeral rhythm of body temperature is preserved during the ?plateau? phase of fever (Fig. 6.2 and 6.3). In addition to an increase body temperature, my free-living kudu also displayed a decreased activity, characteristic of sickness behaviour. When they were febrile, my kudu reduced activity by 40% compared to that when afebrile. Because I recorded activity continuously, by implanted data loggers, I also could show that, just as they preserved the nychthemeral rhythm of body temperature when febrile, the kudu preserved the nychthemeral rhythm of activity, and specifically its crepuscular peaks and midday trough (Fig. 6.6), as if there were a downward shift in the set-point of a ?regulator of activity?. Because the presence of human observers distorts the thermoregulation and behaviour of animals (Recarte et al., 1998), human presence in the vicinity of the animals was kept to a minimum. Consequently, I do not know what the kudu did to reduce activity when febrile, nor how they set about selecting microclimates, however, kudu have been reported to display clear habitat preference for patches with dense cover (Fabricius and Mentis, 1992; Henley, 2001). In addition, since it was not the original intention of my study to investigate fever patterns, I cannot verify the nature of the pathogen or the cause of death. I also do not have records of other sickness behaviours, for example feeding, nor did I have any way of monitoring heat production and heat loss, in these free-living animals. I assume that the increased difference between abdominal and subcutaneous tissue temperatures, during fever, reflected peripheral vasoconstriction. But a decrease in environmental temperature would lead to the same result without a change in skin blood flow. However, given that the kudu selected warmer microenvironments during fever, the former explanation is more likely. 196 My data provides the first evidence for the proposition that free-living animals elevate body temperature during fever by using the same mechanisms as do human patients and laboratory animals, namely, a coordinated suite of behavioural and autonomic responses. The shape and magnitude of the body temperature increases during fever in the kudu are similar to those seen in free-living impala (Kamerman et al., 2001) and springbok (Fuller et al., 2005). The maintenance of the nychthemeral rhythm of body temperature also has been reported during prolonged fevers in goats (Mphahlele et al., 2004) and humans (Musher et al., 1979; Lell et al., 2000). However, because body temperature during prolonged fever in small laboratory mammals typically is raised much more during the inactive phase than the active phase of their nychthemeral activity patterns, the nychthemeral rhythm of body temperature tends to be reduced or obliterated in small species (O'Reilly et al., 1988; Gourine et al., 1998; Luker et al., 2000; Szelenyi et al., 2004; Morrow and Opp, 2005). Whether or not the body temperature rhythms are maintained, in environments warm enough for evaporative heat loss to be stimulated, body temperature could increase as a result of vasoconstriction or inhibition of evaporative heat loss rather than because of an increase in metabolic heat production. It is likely that evaporative heat loss would be inhibited during the genesis of fever simply because the set-point for its initiation would be raised. However, I propose that, in animals capable of selective brain cooling, the implementation of selective brain cooling during fever further inhibits evaporative heat loss. If that were the case, one would expect selective brain cooling to be implemented at the initiation of the fever, which appeared to be the case in my kudu (Fig. 6.7). However, in oxen (Chesy et al., 1980) and goats (Kuhnen, 1994), selective brain cooling was most conspicuous during the defervescence phase of fever. If sheep are prevented, by upper respiratory bypass, from implementing selective brain cooling during fever, so that brain temperature rises, the elevation in rectal temperature is markedly attenuated, as would be expected if the high brain temperature stimulated evaporative heat loss (Laburn et al., 1988). In species that do 197 not have a carotid rete, and in which there is no selective brain cooling during fever (Tegowska and Narebski, 1985), artificial cooling of the brain during fever leads to an exaggerated increase in body temperature (Hori and Harada, 1974; Banet, 1979), without increasing the metabolic cost of fever. While I could not measure metabolic rate in my kudu, fever is metabolically costly and has been estimated to increase the metabolic heat production of humans by ~ 13% with each degree centigrade rise in body temperature (Hart, 1988; Kluger, 1991). This metabolic cost is species-specific and dependent on habitat, severity and duration of infection, nutritional and metabolic status, and the ambient temperature (Baracos et al., 1987; Hart, 1988; Demas et al., 1997; Elliot et al., 2002). At low ambient temperatures, the increase in body temperature is achieved mainly through heat production, such as shivering, however, at higher ambient temperatures heat is stored through inhibition of heat loss (for example cutaneous vasoconstriction and the inhibition of evaporative heat loss), with little change in heat production (Hales et al., 1973; Stitt, 1973; Hori and Harada, 1974; Baracos et al., 1987). The selection of warm microclimates by my kudu not only facilitated an increase in body temperature directly, but may also have resulted in a preference for the employment of heat conservation mechanisms, such as increased peripheral vasoconstriction and the inhibition of evaporative heat loss. In addition to the autonomic changes I have outlined and the selection of warmer microclimates, an additional behavioural mechanism proposed to reduce the metabolic maintenance costs of fever is a decreased activity. In theory, reduced activity acts as an energy conservation mechanism, by reducing wasteful wandering activity and convective heat loss associated with moving (Hart, 1988; Konsman et al., 2002; Wingfield, 2003). Decreased activity associated with fever has been reported in various laboratory studies in both rats (Gordon, 1994; Murakami et al., 1995; Luker et al., 2000; Szelenyi et al., 2004) and mice (Kozak et al., 1994). Conversely, a decreased activity was not observed during sustained experimental fevers in penned 198 goats (Mphahlele et al., 2004), but unlike rodents, which can be highly active in cages, penned goats have low activity, perhaps too low for a further decrease to be detected during fever. My study is the first to demonstrate reduced activity in free- living febrile animals. However, inactivity in free-living animals is not without risk as it could make them more susceptible to predation, and reduce their foraging capacity (Hart, 1988; Yirmiya et al., 1995; Aubert et al., 1997; Konsman et al., 2002; Owen- Ashley and Wingfield, 2006). As a regulated physiological phenomenon, one would expect both fever and sickness behaviour to have survival value (Hart, 1988; Kluger et al., 1996; Dantzer, 2004). Fever has evolved, however, during the so-called host-pathogen evolutionary arms race (Ewald, 1980; LeGrand and Brown, 2002), and could equally benefit the micro- organism causing the infection, as it could the host. Banet (1981) reported that the survival of rats was directly proportional to the metabolic cost during the rising phase of fever but inversely proportional to the magnitude of fever. My kudu not only appeared to reduce their metabolic costs, through vasoconstriction, warmer microclimate selection and reduced activity, but also displayed high mean maximum body temperatures of 41.2?C. I do not know whether the fever and sickness behaviour displayed by my kudu were beneficial to them. Nevertheless, I believe that a proper assessment of the natural role of fever in thermoregulation can be made only by studying fever in free-living animals, just as the roles of other thermoregulatory strategies are revealed only by studies of free-living animals away from human observers (Mitchell et al., 2002). By demonstrating, for the first time, coordinated behavioural and autonomic thermoregulatory effectors, and sickness behaviour, in free-living kudu that became infected, I believe that I have made an important first step in that direction. With the spread of pathogens predicted to increase as the arthropod vectors increase their distribution range with climate change, it seems likely that the pathogens will have an advantage over their hosts. I have demonstrated the kinds of physiological 199 mechanisms that sick artiodactyls will display as they develop immunity to these novel pathogens. However, the development of such immunity is not without costs, as the elevation in body temperature during a febrile response would increase the risk of lethal hyperthermia during hot periods, which are predicted to become more frequent with climate change. The decreased activity associated with the febrile state may also compromise an animal?s ability to employ cathemerality and the resultant reduction in foraging time may be particularly costly for species which are already nutritionally stressed as a consequence of the increased desertification predicted to occur in many regions throughout Africa. The loss of body condition which is likely to develop from both the reduced foraging time and anorexia that accompanies fever may further limit an artiodactyl?s capacity for normal behavioural patterns. If such behavioural changes result in an increased predation risk or decrease reproductive success, they may ultimately reduce an individual?s fitness. The cost of immunity and the associated sickness behaviours are therefore likely to reduce an animal?s ability to employ physiological plasticity to cope with climate change. 6.6 Acknowledgements I thank Arthur and Trinette Rudman for their hospitality and for allowing the study to take place on their farm Blaauwkrantz, Malcolm Rutherford of River Bend Lodge for use of their holding pens and surgical facilities, Dr Kennedy Erlwanger for assisting with surgical procedures, Andr? Matthee for project assistance and game management expertise, and Sophie and Martin Haupt from African Wildlife Tracking for their help in designing and making the animal collars. 200 ___________________________________________________________ CHAPTER 7 ___________________________________________________________ 7 Conclusion 201 The general aim of my thesis was to contribute to the understanding of phenotypic plasticity of long-lived mammals, likely to be essential to cope with climate change. More specifically, I investigate the physiological strategies employed by free-ranging artiodactyls to adapt to arid regions. Since many arid regions of Africa are predicted to get both hotter and drier with climate change (Boko et al., 2007), understanding the physiological mechanisms exhibited by arid-adapted artiodactyls would assist future studies in monitoring physiological adaptations to desertification. A secondary aim of my thesis was to investigate whether remote sensing technology is sensitive enough to detect physiological plasticity, such as changes in body temperature and activity. To achieve these aims, with the help of my colleagues, I implanted biologging devices and compared thermoregulatory strategies of different colour morphs of a single species (chapter two), of two different-sized artiodactyls inhabiting the same desert environment (chapter four) and of the same artiodactyl species inhabiting two different habitats (chapter five). 7.1 Morphological adaptations Morphological changes are believed to have constituted the first evolutionary response of mammals to climate change events of the past (Barnosky et al., 2003) and numerous studies have documented changes in morphology in response to recent changes in climate (Kingsolver et al., 2001). One such morphological variable, which is likely to have thermoregulatory consequences, is an animal?s colour. Yet, recent studies have concluded that concealment appears to be the driving force behind the evolution of colouration in ungulates, with communication and thermoregulation playing less of a role (Stoner et al., 2003). Nevertheless, in chapter two, I showed the thermoregulatory consequences of different colour morphs of springbok in the arid Karoo. I concluded that there is thermoregulatory significance of pelt colour and there probably are selective pressures acting against both the white and the black springbok, that is the less successful morphs, at different times of the year. The black springbok seemed able to reduce energy expenditure in winter, but experienced 202 higher solar heat load in hot conditions. On the other hand, the white springbok apparently lived closer to the energetic edge in winter, possibly because its lower heat load from solar radiation required a higher metabolic cost of homeothermy. From a physiological perspective, the normal springbok appeared to occupy a compromise position, with better energy balance than the white springbok in winter and less overheating than the black springbok in summer, a compromise which may explain why the black and white springbok rarely occur naturally, despite the genes for those colour morphs being present. Although it was beyond the scope of my study to compare the relative roles of colour in thermoregulation and in crypsis, the aposematic function of the normal springbok?s colouration may further explain the preponderance of its prevalence (Caro, 2009). We may find, however, that selection pressures change over time and that the white morph may be selected for in the future as conditions get progressively hotter and drier with climate change. Yet, such selection pressures would have to override any predatory pressure, which, because the conspicuousness of the white colour morph, may counter any thermoregulatory advantage. The prediction that lighter coloured individuals would be selected for as conditions get progressively drier with climate change would be consistent with Gloger?s ecogeographical rule (Millien et al., 2006). Gloger?s rule states that species of endotherms found in humid regions tend to be more heavily pigmented than those found in dry regions. Indeed, the Arabian sand gazelle, which inhabits one of the driest deserts in the world, is lighter in colour than the springbok and most gazelle species found throughout the tropics of Africa. Similarly, the Arabian oryx, reputed to be the most arid-adapted of all artiodactyls, is famed for its white colouration. Nevertheless, such an adaptation option would be feasible only for a species with sufficient genetic diversity for a trait that adapts individual animals to different thermal environments, and a sufficiently fast generation time to allow for that trait to be selected. Artificial selection pressures, which recently have increased the prevalence of the colour morphs of several species within South Africa, combined 203 with decades of predator control removing predation on conspicuous colour morphs, may be advantageous for increasing the genetic variability of populations and may ultimately provide a pre-adaptation to changing climatic conditions. Unlike pelt colouration, body mass need not be genetically determined and can be influenced by environmental factors, making it a phenotypically plastic trait. In addition to their light colouration, endotherms of the Arabian Peninsula are generally smaller than their counterparts which inhabit more temperature regions, conforming to yet another ecogeographical rule, namely Bergmann?s rule. Bergmann?s rule predicts a positive correlation between body mass of terrestrial endotherms and latitude, and, by inference, an inverse correlation between body mass and environmental temperature. Thus, the higher ambient temperatures predicted to occur with climate change may provide a selective advantage for smaller individuals (Teplitsky et al., 2008). However, rather than climate change favouring the selection of smaller species (genotypic adaptation) it may favour the selection of smaller individuals within a species (phenotypic plasticity), which would allow species to adapt to changing climatic conditions over a much shorter time interval. Yet temperature is not the only environmental factor which is likely to influence body mass; both reduced water availability and reduced plant productivity also would favour small individuals (Rosenzweig, 1968; Wigginton and Dobson, 1999). Thus, the primary increase in ambient temperature and the secondary reduction in water availability, and resultant lower plant productivity, predicted to occur in the future may act to reduce the body size of endotherms. Such changes in body size are observable already in small endotherms, such as birds (Yom-Tov, 2001; Yom-Tov and Yom-Tov, 2006; Yom-Tov et al., 2006) and rodents (Smith et al., 1995; Smith et al., 1998). Similarly, smaller artiodactyls, hypothetically, would have greater access to thermal refuges, potentially giving smaller artiodactyls another favourable adaptation to hot and dry environments, over their larger counterparts. In chapter four I described the investigation of the different thermoregulatory strategies of a small and large desert-adapted artiodactyl, namely the small Arabian sand gazelle (~ 15 kg) 204 and the larger Arabian oryx (~ 70 kg), inhabiting the same desert environment, at the same time. Despite the oryx having a body mass more than four-fold that of the sand gazelle, the two species responded remarkably similarly to changes in environmental conditions. Both species employed heterothermy and cathemerality, and selected the same cooler microclimates during times of heat stress. Yet, because of the difference in body mass, for the same amplitude of nychthemeral rhythm of body temperature, the oryx would store about four times as much heat as the gazelle as body temperature rises, and therefore save about four times as much water, that otherwise would be spent on evaporative cooling. If small artiodactyls are unable to find appropriate refuges, they may be disadvantaged in hot and dry conditions because of their high mass-specific metabolic rates, high water turnover and less capacity to store heat. Future studies would therefore need to establish whether changes in body mass result in changes in function, such as the functional change I showed for pelt colouration in chapter two, and whether smaller individuals within a species would be advantaged in hot and dry environments. Unfortunately all of the gazelle died within five months of my starting the project. I do not know whether the captive-bred gazelle were maladapted to the extremely hot and dry conditions of the desert environment or whether their fenced enclosure prevented them from seeking water and moving to potentially more appropriate environments. Since sand gazelle are known to travel large distances in search of suitable forage after rain (Harrison and Bates, 1991; Thouless et al., 1991), restricting the dispersal ability of this species may have increased the stress associated with the exceptionally hot and dry conditions prevalent during my study and, as such, highlights the adverse effect fences may have in restricting an animal?s ability to move to more appropriate environments. However, the high mortality of sand gazelle was not specific to the gazelle in my enclosure; nearly 700 sand gazelle (50% of the estimated population) died in the Mahazat as-Sayd over the same time period (Cunningham et al., 2008). Whether or not the death of my sand gazelle was the result of their fenced enclosure restricting their dispersal ability, fences are likely to 205 be detrimental to large mammalian species in South Africa, where the majority of national parks are fenced. 7.2 Autonomic adaptations In seeking to explore how artiodactyls will respond to climate change, I sought the most extreme of current environments and chose to investigate physiological mechanisms of an arid-adapted artiodactyl which survives there. The deserts of the Arabian Peninsula are considered to have one of the most inhospitable climates on earth. Of the seven artiodactyl species which inhabit this hot, hyper-arid environment, the species which arguably faces the greatest challenge to homeostasis is the Arabian oryx (Oryx leucoryx). With a body mass of 70-100 kg, this artiodactyl survives most of the year, including the hottest summer months, without access to drinking water. To investigate the thermoregulatory consequences of water limitation, in chapter three I investigated thermoregulatory strategies of the arid-adapted Arabian oryx over a one-year period during which conditions got progressively drier. I showed that heterothermy indeed was present but was dependent not only on high ambient temperatures but also on lack of water. The oryx, like other desert-adapted endotherms, would have to trade off thermoregulation, osmoregulation and energy acquisition (Cain et al., 2006). I believe the data obtained from the oryx supports the view that priority is given to osmoregulation. When the oryx had access to sufficient dietary water to maintain both osmoregulation and homeothermy (by evaporative cooling), they did so. However, when they did not have access to sufficient dietary water they abandoned homeothermy. When environmental temperature exceeds body temperature evaporative cooling provides the only mechanism to lose metabolic heat, but when water is scarce animals may attempt to conserve body water at the expense of homeothermy. Yet, adaptive heterothermy differs from dehydration-induced hyperthermia in that mean body temperature does not increase because adaptive heterothermy, by definition, also is 206 associated with a decreased minimum body temperature. The limited evidence for the drop in minimum body temperature associated with adaptive heterothermy in free- living animals led some researches to question whether large fluctuations in body temperature are rather the result of dehydration-induced hyperthermia (Mitchell et al., 2002) and, as such reflect a compromised thermoregulatory ability during water deprivation. The capacity to store heat, and thereby conserve body water, holds whether the heterothermy is a controlled thermoregulatory event, that constitutes adaptive heterothermy, or results from failure of homeothermy, but failure of homeothermy may result in potential lethal hyperthermia. A study by Dawson et al. (2007) provided further evidence that, at least in kangaroos, heterothermy may represent a compromised thermoregulatory ability, rather than an adaptation. At the same level of dehydration, the desert-adapted red kangaroo (Macropus rufus) displayed reduced heterothermy while still reducing its evaporative water loss, compared to its mesically-adapted counterpart, the eastern grey kangaroo (Macropus giganteus, Dawson et al., 2007). But whether heterothermy is adaptive in artiodactyls, or whether it reflects a compromised thermoregulatory ability, as in kangaroos, remains to be investigated. To do so, we would need to investigate whether desert- adapted artiodactyls are more likely to employ heterothermy than mesically-adapted species, under the same environmental conditions. In addition, we would need to investigate whether heterothermy confers any advantages in terms of water conservation and whether individuals that employ heterothermy are more or less likely to survive extremely hot and dry conditions. I believe that, unlike adaptive heterothermy, the heterothermy observed in my oryx was the result of a failure of homeothermy as a result of a dehydration-induced hyperthermia combined with a starvation-induced hypothermia. Adaptive heterothermy predicts that the a low minimum body temperature would allow an animal to anticipate hot ambient conditions (Maloney et al., 2004) and thus pre- emptively permit additional storage of heat. Yet, my oryx not only displayed a low minimum body temperature during the hot dry period, but their minimum body 207 temperatures remained low throughout the dry period, even though ambient temperatures were not excessively high during these periods. These low minimum body temperatures throughout the dry period may reflect a reduced metabolic heat production, as a result of the Q10 effect, as has been proposed for wild kangaroo (Dawson et al., 2007). A low metabolic rate during the dry season could result from the combined effect of dehydration and starvation, as has been shown in other desert ungulates (Brosh et al., 1986; Merkt and Taylor, 1994; Williams et al., 2001; Ahmed and El Kheir, 2004; Ostrowski et al., 2006a; Ostrowski et al., 2006b). Similarly, low minimum body temperatures, postulated to have resulted from low metabolic rates during times of energy restrictions, were observed in both my sand gazelle during, the hot and dry period (chapter four), and the white springbok colour morph during the dry Karoo winter (chapter two). Thus, long-term monitoring of body temperature may provide a sensitive index of physiological stress as a result of changing climatic conditions, as low minimum body temperatures are likely to reflect nutritional stress and high maximum body temperatures are likely to reflect water stress. A second autonomic mechanism that I believe my Arabian oryx used to conserve body water and facilitate homeostasis at high environmental heat loads was selective brain cooling. Arid-zone mammals possessing a carotid rete may employ selective brain cooling to attenuate thermal drive by reducing hypothalamic temperature, which reduces evaporative heat loss and ultimately conserves water by transferring cooling to non-evaporative means (Kuhnen, 1997; Jessen, 1998; Jessen, 2001; Mitchell et al., 2002). If the evolution of the carotid rete indeed promoted thermoregulatory flexibility and thus facilitated the invasion of arid zones during the highly seasonal post-Eocene climate (Mitchell and Lust, 2008), such an anatomical structure may well provide a key adaptation for artiodactyls to cope with aridity and heat stress predicted to occur with climate change (Fuller et al., 2008). However, not all artiodactyls exhibit the same capacity for selective brain cooling. Primitive artiodactyls, such as the mouse deer (Tragulus javanicus, Tragulus napu), do not possess a carotid rete (Fukuta et al., 2007), which hypothetically restricts this species 208 to the climatically-stable forest environment (Whittow et al., 1977). Whether all artiodactyls inhabiting arid environments have the same capacity for selective brain cooling is not known. Because the hypothalamus is exceptionally sensitive to changes in temperature, even a relatively small difference of a few tenths of a degree in selective brain cooling could result in a large difference in suppression of evaporative cooling and therefore water loss. To date, there has been no comparative investigation of selective brain cooling in sympatric species simultaneously occupying the same hot arid environment. Nevertheless, it is encouraging to note that the magnitude of selective brain cooling in the arid-adapted Arabian oryx was similar to that observed in other free-living artiodactyls (Jessen et al., 1994; Jessen and Kuhnen, 1996; Mitchell et al., 1997; Fuller et al., 1999; Maloney et al., 2002), implying that even relatively small magnitudes of selective brain cooling may be sufficient to substantially reduce evaporative water loss and thereby conserve body water. Since selective brain cooling does not disrupt other homeostatic systems, it may provide an economical form of autonomic thermoregulation. If artiodactyls are capable of phenotypic plasticity in the magnitude of selective brain cooling, the carotid rete may ultimately provide artiodactyls with greater acclimatization potential to the hot and dry conditions predicted to occur in the future. 7.3 Behavioural adaptations An additional mechanism proposed to conserve body water and reduce metabolic costs as conditions get progressively hotter and drier would be an adjustment in the pattern of activity, termed cathemerality. The term ?cathemerality? was originally coined for primates, which were observed to be active throughout the day and night, and was later defined as significant amounts of activity occurring during both light and dark phases of the 24-h daily cycle (Tattersall, 1987; Tattersall, 2006; Tattersall, 2008). It differs from both crepuscular and diel activity, which follow a 24-h periodicity (Aschoff, 1966), and, instead, describes variation in activity patterns in response to prevailing ecological conditions. Chronoecological factors, such as 209 temperature, light, competition for food resources and predation, appear to promote cathemerality by overriding the endogenous clock (Curtis and Rasmussen, 2006). Although seldom studied, cathemerality may be common among non-primate mammals and a study on Indonesian rain forest mammals showed all nine species of ungulates studied to be cathemeral (van Schaik and Griffiths, 1996). Identification of cathemerality requires measurement of both diurnal and nocturnal activity (Curtis and Rasmussen, 2006), which I achieved objectively by using activity loggers. Because the presence of human observers distorts the behaviour of artiodactyls (Recarte et al., 1998), activity loggers provided a long-term, remote and unbiased assessment of activity, over successive 24-h cycles, not otherwise feasible under field conditions. Unlike van Schaik and Griffiths (1996), who classified animals as cathemeral based on the proportion of time spent active during diurnal and nocturnal periods, cathemerality, by definition, requires a variation in activity in response to ecological conditions. Using activity loggers, I was able to show that both the oryx and the sand gazelle shifted from a continuous activity with crepuscular peaks during warm wet periods to nocturnal activity during the hot dry period. Such an increase in nocturnal activity is believed to be a compensatory response, as grazing activities during the heat of the day would be limited as a result of shade- seeking behaviour, in response to high ambient temperatures. Shade-seeking behaviour reduces the heat load on an animal and conserves body water by reducing the need for evaporative cooling. I was able to quantify such behaviour in my oryx and sand gazelle, without observers present, by using miniature globe thermometers attached to collars, a technique I developed (Appendix 1, Hetem et al., 2007). Both cathemerality and shade-seeking behaviour represent flexible behavioural processes, which are likely to become increasingly important and may act to buffer the adverse effects of the progressively hot and dry conditions predicted to occur with climate change. However, unlike the Arabian oryx and sand gazelle, which can freely shift between diurnal and nocturnal activity since they are not exposed to natural predators, 210 many species throughout Africa may be exposed to a greater predation pressure should they increase their nocturnal activity. If African species are unable to shift their activity to more appropriate times of day and are forced to be active throughout the heat of the day because of constraints by predatory pressure, they are likely to become more dependent on the already scarce and fast declining water sources, if they attempt to maintain homeothermy. When these water sources run dry, species may be forced to display heterothermy. Although, it is encouraging to note that six species of East African artiodactyls tolerated large fluctuations in rectal temperature, at least over the short term (Taylor, 1970b), it has been questioned whether these measurements were artefacts of measurement error (Mitchell et al., 2002). It therefore remains to be seen whether free-living African species will be able to cope with the large fluctuations in body temperature exhibited by Arabian artiodactyls. A seemingly simple solution to the water stress associated with desertification would be to provide artificial water sources to regions where water supplies are limited (Epaphras et al., 2008). However, such a management solution is not without risk and would potentially provide only a short-term solution to the problem. The recent decline in the roan antelope (Hippotragus equinus) population in the Kruger National Park, for example, has been attributed to an increase in the number of artificial water points (Harrington et al., 1999b). It is proposed that these water points made previously dry areas, where roan antelope thrived, accessible to water-dependent species, such as wildebeest and zebra, year round. Lions followed their prey and the increased predation on adult roan antelope is believed to have caused the roan antelope to decline from about 450 to just 45 individuals between 1986 and 1993. The provision of artificial water points not only affects resident herbivores through the indirect effects of competition and predation, but the increased grazing pressure associated with the artificially increased density of herbivores can also have detrimental effects on the vegetation structure, creating so- called piospheres (Thrash, 1998; Todd, 2006). In fact, the provision of artificial water points, particularly in arid regions, may exacerbate land degradation and lead to 211 massive die-offs when the degraded vegetation can no longer sustain the artificially high herbivore numbers, ultimately resulting in a loss of biodiversity (Nangula and Oba, 2004; Fensham and Fairfax, 2008). Thus, the adaptive strategy with the least disruptive consequences for ecosystem function would be for species to show phenotypic plasticity and to adapt to the increasingly desertified landscapes to which they are likely to be exposed. Even those arid-adapted species which can survive independent of drinking water would need to adapt phenotypically to the reduced water availability in their forage. 7.4 Adapting to desertification One of the consequences of climate change will be habitat transformation and there is debate about the relative detrimental effects of the climate itself and its consequences for habitat. Since I could now predict the kind of physiological mechanisms required to adapt to desert environments, in chapter five I investigated whether similar responses were exhibited by an arid-adapted artiodactyl, which is likely to have to adapt to desertified landscapes in the Eastern Cape Province, South Africa. Because Angora goats are long-lived artiodactyls, which for economic reasons cannot be translocated, they provide the ideal study species in which to investigate plasticity of herbivores in response to habitat transformation. Although, in general, goats that inhabited the transformed and intact sites responded similarly as a result of the relatively benign conditions and high rainfall prevalent during my study, I was able to demonstrate physiological changes in response to desertification when goats were subjected to a physiological stress imposed by shearing. Shearing changed the insulative property of the pelt and, as such, represents a morphological change which appeared to make the animals more susceptible to changes in environmental conditions (Appendix 2, Hetem et al., 2009). Post-shearing, goats that inhabited the transformed site had a faster rate of rise in body temperature in the morning, displayed an increased 24-h amplitude of body temperature rhythm and were generally less active compared to goats that inhabited the intact site. When goats were 212 denied access to free-standing water, the goats that inhabited the transformed site were unable to maintain a positive water balance as a result of a suppressed dietary water intake, combined with a higher water turnover, compared to the goats that inhabited the intact site. Such an increased amplitude of body temperature rhythm and reduced activity when goats were exposed to water stress on the transformed site may reflect a trend towards heterothermy and cathemerality, as was observed in the Arabian oryx and sand gazelle. Because forage availability varies by orders of magnitude between wet and dry periods (Stuart-Hill and Aucamp, 1993), such physiological strategies are likely to become obligatory with further desertification predicted with climate change. Future studies should therefore focus on the physiological plasticity of artiodactyls exposed to habitat transformation under conditions both hotter and drier than those prevalent during my study. Given that climate change is already exacerbating drought extremes (Chamaill?-Jammes et al., 2007), and that the semi-arid to arid subregions of Southern Africa are particularly vulnerable, understanding the effects of these extremes will become increasingly important in the future. The use of sophisticated physiological measurements applied here will allow not only an understanding of the extent of adaptive responses, but also improve the predictions of species responses to climate change in the future. If we are to successfully conserve biodiversity under changing climatic conditions, we need an understanding of which species respond well to hot and dry environments so that we can prioritise those species for conservation in the future. Indeed, studies are already suggesting which breed of Borana cattle should be prioritised for cultivation in East Africa (Zander et al., 2009), with the aim to prioritise species with the greatest genetic diversity to adapt to future challenges. Such recommendations are based on Weitzman?s cost-effectiveness methodology (Weitzman, 1998), which ranks species based on a combination of economic factors and changes in survival probabilities. However, the predicted survival probabilities of the culturally important livestock species were fairly arbitrary and were based on the opinions of local livestock- 213 keepers (Zander et al., 2009). Such an approach could be improved by incorporating measures of genetic diversity and a species physiological adaptive capacity. 7.5 Adapting to the spread of pathogens Although my transformed site appeared to be a more stressful environment than the intact site, both in terms of thermal microclimates and water availability, one potential advantage of more open habitats is a reduced parasite load (Mooring et al., 2004). The low albumin to globulin ratio of goats that inhabited the intact site is likely to have reflected a higher parasite load, since a low albumin to globulin ratio is associated with inflammation and bacterial invasion (Hurwitz and Whipple, 1917), decreased immune function in heifers (Piccinini et al., 2004) and low social status in pigs (Hicks et al., 1998). I propose that the goats that inhabited the transformed site were less immunologically stressed than the goats that inhabited the intact site. Although a reduced parasite load on the desertified site may presently be advantageous, climate change is predicted to bring emergent pathogens with the relocation of arthropod vectors (Rogers and Randolph, 2000; Olwoch et al., 2003; Cumming and van Vuuren, 2006). We therefore cannot simply focus on adaptation responses of healthy individuals, but also need to include the studies of innate immunity and sickness behaviour. In chapter six I investigated the physiological consequences of infection in free-living kudu. Although it was not my original intention to study the consequences of infection in free-living kudu, the opportunity provided the first evidence ever of the suite of responses of a sick animal in the combination of immune compromise and environmental stress likely to be the typical scenario with climate change. The fever was characterized by a significant increase in mean 24-h body temperature, but it did not alter the amplitude of nychthemeral rhythm of body temperature. Kudu reduced the metabolic cost of fever by increasing vasoconstriction and selecting warmer microclimates during the fever than when afebrile. In addition, some kudu displayed 214 an increased frequency of selective brain cooling during the fever, which would have inhibited evaporative heat loss and increased febrile body temperatures, without increasing the metabolic maintenance costs of high body temperatures. Not only did I record quantitative evidence for autonomic and behavioural fever, but I also recorded the first evidence of sickness behaviour, in the form of decreased activity, in a free- living ungulate species. As a regulated physiological phenomenon, one would expect both fever and sickness behaviour to have survival value (Hart, 1988; Kluger et al., 1996; Dantzer, 2004), but fever evolved during the so-called host-pathogen evolutionary arms race (Ewald, 1980; LeGrand and Brown, 2002) and could equally benefit the micro-organism causing the infection, as it could the host. With the spread of pathogens predicted to increase as the arthropod vectors increase their distribution range in the future, it seems likely that the pathogens will have an increasing advantage over their hosts. In addition to the artiodactyl hosts having to develop immunity to novel pathogens, they are likely to have to contend with the increased metabolic costs of immunity superimposed on the chronic physiological stress of having to adapt to the climatically unsuitable areas to which they are confined. Although my kudu reduced the metabolic costs of immunity, even a slight increase in metabolic requirements may be too costly for artiodactyls already at their physiological limit. Thus, the predicted spread of pathogens with climate change is likely to increase extinction risk. Future studies will need to monitor physiological variables of free-living species to assess whether infections are indeed becoming more prevalent. In addition responses of sick animals need to be assessed in hot environments where the increased risk of hyperthermia may make febrile response maladaptive. Ultimately we will need to include the responses of sick animals into both models and experimental studies to better anticipate the effects of climate change. 215 7.6 Perspectives and significance I have revealed the kinds of physiological plasticity which free-living artiodactyls are likely to implement as conditions in Africa get progressively hotter and drier with climate change and, in so doing, I have shown that physiological measures are an appropriate way to detect an animal?s adaptation to changing environments. Activity patterns and microclimate selection are flexible behavioural processes which are likely to represent an animal?s primary defence to changes in climatic conditions, particularly for large-bodied species with long generation times which cannot rely on microevolution because of the unprecedented rate of current climate change. If such behavioural processes are insufficient to maintain homeothermy, we are likely to observe changes in an animal?s body temperature. Body temperature provides a sensitive indicator of physiological stress in terms of infection, dehydration, loss of body condition and nutritive stress. We need to make such measurements in individuals of several species inhabiting a variety of environments. Such studies would fall into the recently defined field of macrophysiology, defined as ?the investigation of variation in physiological traits over large geographical and temporal scales and the ecological implications of this variation? (Chown et al., 2004a). Incorporating such macrophysiological data into bioclimatic envelope models will allow us to better predict how species will respond to climate change (Porter et al., 2002) and ultimately contribute towards effective management and conservation of Africa's biodiversity in the face of future climate change. Knowing which species demonstrate sufficient physiological plasticity to cope with the consequences of climate change will allow for more informed decisions as to which species need to be relocated (Hoegh-Guldberg et al., 2008) and potentially which species should be prioritised for conservation (Zander et al., 2009) in the future. Ultimately, we require an interdisciplinary approach, combining the expertise of ecologists, botanists, zoologists, veterinarians and physiologists to provide a more complete picture of the interplay between the numerous variables affected by changing climatic conditions. As technology, such as biotelemetry, develops the 216 range of physiological variables which can be monitored in free-living animals is likely to increase. Within this thesis I have extended the typical biotelemetric measures of body temperature and showed the value of including remote measurements of activity and thermoregulatory behaviour. Such new technology, combined with current methodology such as Geographical Information Systems (GIS), will allow for a multidisciplinary approach to aid our understanding of animal biogeographical distributions. Since long-term monitoring programmes are already in place for ecological and environmental variables, for example the South African Environmental Observation Network (SAEON), the next logical step would be to incorporate physiological responses of individual organisms into such long-term monitoring programmes. Such long-term physiological measurements will allow us to detect when physiological thresholds are exceeded and when behavioural processes become insufficient to cope with climate change. Data from such long-term monitoring programmes will ultimately help to elucidate the causal mechanisms underlying ecosystem changes and allow us to better detect, predict and react to future climate change. 217 8 Literature cited Addo-Bediako A., Chown S.L. and Gaston K.J. (2000). Thermal tolerance, climatic variability and latitude. Proceedings of the Royal Society of London. Series B, Biological Sciences 267: 739-745. Adolf E.F. (1956). General and specific characteristics of physiological adaptations. American Journal of Physiology 184: 18-28. Ahmed M.M.M. and El Kheir I.M. (2004). Thermoregulation and water balance as affected by water and food restrictions in Sudanese desert goats fed good-quality and poor-quality diets. Tropical Animal Health and Production 36: 191-204. Alamer M. (2006). Physiological responses of Saudi Arabia indigenous goats to water deprivation. Small Ruminant Research 63: 100-109. Alderman J.A., Krausman P.R. and Leopold B.D. (1989). Diel activity of female desert bighorn sheep in western Arizona. Journal of Wildlife Management 53: 264-271. Allison C.D. (1985). Factors affecting forage intake by range ruminants: a review. Journal of Range Management 38: 305-311. Altmann J. (1974). Observational study of behaviour: sampling methods. Behaviour 49: 227-267. Amstrup S.C., Marcot B.G. and Douglas D.C. (2007). Forecasting the range-wide status of polar bears at selected times in the 21st century. U.S. Geological Survey Administrative Report, Alaska Science Center, Anchorage, Alaska, USA. Anderson T.A., Hubbert F., Jr., Roubicek C.B. and Taylor R.E. (1962). Influence of protein depletion on vitamin A and carotene utilization by vitamin A-deficient sheep. Journal of Nutrition 78: 341-347. Andrews R.D. (1998). Instrumentation for the remote monitoring of physiological and behavioral variables. Journal of Applied Physiology 85: 1974?1981. Animut G., Goetsch A.L., Aiken G.E., Puchala R., Detweiler G., Krehbiel C.R., Merkel R.C., Sahlu T., Dawson L.J., Johnson Z.B. and Gipson T.A. (2005a). Performance and forage selectivity of sheep and goats co-grazing grass/forb pastures at three stocking rates. Small Ruminant Research 59: 203-215. Animut G., Goetsch A.L., Aiken G.E., Puchala R., Detweiler G., Krehbiel C.R., Merkel R.C., Sahlu T., Dawson L.J., Johnson Z.B. and Gipson T.A. (2005b). Grazing behavior and energy expenditure by sheep and goats co-grazing grass/forb pastures at three stocking rates. Small Ruminant Research 59: 191-201. Animut G., Goetsch A.L., Aiken G.E., Puchala R., Detweiler G., Krehbiel C.R., Merkel R.C., Sahlu T., Dawson L.J., Johnson Z.B. and Kiesler D.H. (2006). Performance by goats and sheep consuming a 218 concentrate-based diet subsequent to grazing grass/forb pastures at three stocking rates. Small Ruminant Research 66: 92-101. Appleman R.D. and Delouche J.C. (1958). Behavioral, physiological and biochemical responses of goats to temperature, 0 ?C to 40 ?C. Journal of Animal Science 17: 326-335. Ara?jo M.B. and Luoto M. (2007). The importance of biotic interactions for modelling species distributions under climate change. Global Ecology & Biogeography 16: 743-753. Arnold G.W. (1982). Some factors affecting the grazing behaviour of sheep in winter in New South Wales. Applied Animal Ethology 8: 119-125. Arnold G.W. and Dudzinski M.L. (1978). Ethology of Free-Ranging Domestic Animals. Elsevior Scientific Publishing Company, New York, USA. Arnold W., Ruf T. and Kuntz R. (2006). Seasonal adjustment of energy budget in a large wild mammal, the Przewalski horse (Equus ferus przewalskii) II. Energy expenditure. Journal of Experimental Biology 209: 4566-4573. Arnold W., Ruf T., Reimoser S., Tataruch F., Onderscheka K. and Schober F. (2004). Nocturnal hypometabolism as an overwintering strategy of red deer (Cervus elaphus). American Journal of Physiology ? Regulatory Integrative & Comparative Physiology 286: R174-R181. Aschoff J. (1966). Circadian activity pattern with two peaks. Ecology 47: 657-662. Aschoff J. (1982). The circadian rhythm of body temperature as a function of body size. In: A Companion to Animal Physiology. Taylor C.R., Johansen K. and Bolis L. (Eds.), Cambridge University Press, New York, pp. 173-189. Askins G.D. and Turner E.E. (1972). A behavioral study of Angora goats on west Texas range (Un estudio sobre las actividades de cabras de Angora en un pastizal natural en el oeste de Texas). Journal of Range Management 25: 82-87. Aubert A. (1999). Sickness and behaviour in animals: a motivational perspective. Neuroscience and Biobehavioral Reviews 23: 1029-1036. Aubert A., Goodall G., Dantzer R. and Gheusi G. (1997). Differential effects of lipopolysaccharide on pup retrieving and nest building in lactating mice. Brain, Behavior, and Immunity 11: 107-118. Aucamp A.J., Howe L.G., Smith D.W.W.Q. and Grunow J.O. (1980). The effect of defoliation on Portulacaria afra. Proceedings of the Grassland Society of Southern Africa 15: 179-184. Austin M. (2007). Species distribution models and ecological theory: a critical assessment and some possible new approaches. Ecological Modelling 200: 1-19. 219 Bae H.H., Larkin J.E. and Zucker I. (2003). Juvenile Siberian hamsters display torpor and modified locomotor activity and body temperature rhythms in response to reduced food availability. Physiological and Biochemical Zoology 76: 858-867. Baker M.A. (1989). Effects of dehydration and rehydration on thermoregulatory sweating in goats. Journal of Physiology 417: 421-435. Baker M.A. (1982). Brain cooling in endotherms in heat and exercise. Annual Review of Physiology 44: 85-96. Bakken G.S. (1992). Measurement and application of operative and standard operative temperatures in ecology. American Zoologist 32: 194-216. Bakken G.S., Santee W.R. and Erskine D.J. (1985). Operative and standard operative temperatures: tools for thermal energetic studies. American Zoologist 25: 933-943. Bale J.S., Masters G.J., Hodkinson I.D., Awmack C., Bezemer T.M., Brown V.K., Butterfield J., Buse A., Coulson J.C., Farrar J., Good J.E.G., Harrington R., Hartley S., Jones T.H., Lindroth R.L., Press M.C., Symrnioudis I., Watt A.D. and Whittaker J.B. (2002). Herbivory in global climate change research: direct effects of rising temperature on insect herbivores. Global Change Biology 8: 1-16. Banet M. (1979). Fever and survival in the rat. Pfl?gers Archiv - European Journal of Physiology 381: 35-38. Banet M. (1981). Fever and survival in the rat. Metabolic versus temperature response Cellular and Molecular Life Sciences (CMLS) 37: 1302-1304. Baracos V.E., Whitmore W.T. and Gale R. (1987). The metabolic cost of fever. Canadian Journal of Physiology and Pharmacology 65: 1248-1254. Barnosky A.D., Hadly E.A. and Bell C.J. (2003). Mammalian response to global warming on varied temporal scales. Journal of Mammalogy 84: 354-368. Bartholomew G. (1964). The roles of physiology and behaviour in the maintenance of homeostasis in the desert environment. In: Symposia of the Society of Experimental Biology vol 18: Homoestasis and Feedback Mechanisms. Hughes G. (Ed.), Cambridge University Press, Cambridge, UK, pp. 7-29. Bartholomew G. (1987). Interspecific comparison as a tool for ecological physiologists. In: New Directions in Ecological Physiology. Feder M., Bennett A., Burggren W. and Huey R. (Eds.), Cambridge University Press, Cambridge, UK, pp. 11-37. Bearhop S., Fiedler W., Furness R.W., Votier S.C., Waldron S., Newton J., Bowen G.J., Berthold P. and Farnsworth K. (2005). Assortative mating as a mechanism for rapid evolution of a migratory divide. Science 310: 502-504. 220 Beede D.K. and Collier R.J. (1986). Potential nutritional strategies for intensively managed cattle during thermal stress. Journal of Animal Science 62: 543-554. Belovsky G.E. and Jordan P.A. (1978). The time-energy budget of a moose. Theoretical Population Biology 14: 76-104. Belovsky G.E. and Slade J.B. (1986). Time budgets of grassland herbivores: body size similarities. Oecologia 70: 53-62. Bennett K.D. (1990). Milankovitch cycles and their effects on species in ecological and evolutionary time. Paleobiology 16: 11-21. Benton M.J. and Twitchett R.J. (2003). How to kill (almost) all life: the end-Permian extinction event. Trends in Ecology and Evolution 18: 358-365. Berger A., Scheibe K.-M., Eichhorn K., Scheibe A. and Streich J. (1999). Diurnal and ultradian rhythms of behaviour in a mare group of Przewalski horse (Equus ferus przewalskii), measured through one year under semi-reserve conditions. Applied Animal Behaviour Science 64: 1-17. Bergman C.M., Fryxell J.M., Gates C.C. and Fortin D. (2001). Ungulate foraging strategies: energy maximizing or time minimizing? Journal of Animal Ecology 70: 289-300. Bernardo J. (1996). Maternal effects in animal ecology. American Zoologist 36: 83-105. Berteaux D., Reale D., McAdam A.G. and Boutin S. (2004). Keeping pace with fast climate change: can Arctic life count on evolution? Integrative and Comparative Biology 44: 140-151. Berthold P. and Pulido F. (1994). Heritability of migratory activity in a natural bird population. Proceedings of the Royal Society of London. Series B, Biological Sciences 257: 311-315. Beynen A.C., Schonewillea J.T. and Terpstra A.H.M. (2000). Influence of amount and type of dietary fat on plasma cholesterol concentrations in goats. Small Ruminant Research 35: 141-147. Bianca W. and Kunz P. (1978). Physiological reactions of three breeds of goats to cold, heat and high altitude. Livestock Production Science 5: 57-69. Bigalke R.C. (1972). Observations on the behaviour and feeding habits of the springbok, Antidorcas marsupialis. Zoologica Africana 7: 333-359. Bligh E.G. and Dyer W.J. (1959). A rapid method of total lipid extraction and purification. Canadian Journal of Biochemistry and Physiology 37: 911-917. Bligh J. and Harthoorn A.M. (1965). Continuous radiotelemetric records of the deep body temperature of some unrestrained African mammals under near-natural conditions. Journal of Physiology 176: 145- 162. 221 Bligh J. and Johnson K.G. (1973). Glossary of terms for thermal physiology. Journal of Applied Physiology 35: 941-961. Block B.A. (2005). Physiological ecology in the 21st century: advancements in biologging science. Integrative and Comparative Biology 45: 305-320. Boko M., Niang I., Nyong A., Vogel C., Githeko A., Medany M., Osman-Elasha B., Tabo R. and Yanda P. (2007). Africa. Climate Change 2007: Impacts, Adaptation and Vulnerability. In: Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Parry M.L., Canziani O.F., Palutikof J.P., van der Linden P.J. and Hanson C.E. (Eds.), Cambridge University Press, Cambridge, UK, pp. 433-467. Both C., Bouwhuis S., Lessells C.M. and Visser M.E. (2006). Climate change and population declines in a long-distance migratory bird. Nature 441: 81-83. Bradshaw A.D. and McNeilly T. (1991). Evolutionary response to global climatic change. Annuals of Botany 67: 5-14. Bradshaw W.E. and Holzapfel C.M. (2006). Climate change: evolutionary response to rapid climate change. Science 312: 1477-1478. Bradshaw W.E. and Holzapfel C.M. (2008). Genetic response to rapid climate change: it?s seasonal timing that matters. Molecular Ecology 17: 157?166. Brommer J.E. (2004). The range margins of northern birds shift polewards. Annales Zoologici Fennici 41: 391-397. Brosh A., Shkolnik A. and Choshniak I. (1986). Metabolic effects of infrequent drinking and low- quality feed on Bedouin goats. Ecology 67: 1086-1090. Brown G.D. and Dawson T.J. (1977). Seasonal variation in the body temperatures of unrestrained kangaroos (Macropodidae: Marsupialia). Comparative Biochemistry and Physiology A: Molecular and Integrative Physiology 56: 59-67. Bunnell F.L. and Gillingham M.P. (1985). Foraging behavior: dynamics of dining out. In: Bioenergetics of Wild Herbivores. Hudson R.J. and White C.R. (Eds.), CRC Press, Inc., Boca Raton, Florida, USA, pp. 53-79. Burns C.E., Johnston K.M. and Schmitz O.J. (2003). Global climate change and mammalian species diversity in U.S. national parks. Proceedings of the National Academy of Sciences of the United States of America 100: 11474-11477. Burtt E.H. (1981). The adaptiveness of animal colors. BioScience 31: 723-729. 222 Bustamante D., Nespolo R., Rezende E. and Bozinovic F. (2002). Dynamic thermal balance in the leaf-eared mouse: the interplay among ambient temperature, body size and behavior. Physiological and Biochemical Zoology 75: 396-404. Cable N.T., Drust B. and Gregson W.A. (2007). The impact of altered climatic conditions and altitude on circadian physiology. Physiology & Behavior 90: 267-273. Cain J.W., Krausman P.R., Rosenstock S.S. and Turner J.C. (2006). Mechanisms of thermoregulation and water balance in desert ungulates. Wildlife Society Bulletin 34: 570-581. Cain J.W., Jansen B.D., Wilson R.R. and Krausman P.R. (2008). Potential thermoregulatory advantages of shade use by desert bighorn sheep. Journal of Arid Environments 72: 1518-1525. Calosi P., Bilton D.T. and Spicer J.I. (2008). Thermal tolerance, acclimatory capacity and vulnerability to global climate change. Biology Letters 4: 99-102. Canas R.C., Quiroz R.A., Leon-Velarde C., Posadas A. and Osorio J. (2003). Quantifying energy dissipation by grazing animals in harsh environments. Journal of Theoretical Biology 225: 351?359. Caputa M. (2004). Selective brain cooling: a multiple regulatory mechanism. Journal of Thermal Biology 29: 691-702. Cardillo M., Mace G.M., Jones K.E., Bielby J., Bininda-Emonds O.R.P., Sechrest W., Orme C.D.L. and Purvis A. (2005). Multiple causes of high extinction risk in large mammal species. Science 309: 1239-1241. Carey C. (2005). How physiological methods and concepts can be useful in conservation biology. Integrative and Comparative Biology 45: 4-11. Caro T. (2005). The adaptive significance of coloration in mammals. BioScience 55: 125-136. Caro T. (2009). Contrasting coloration in terrestrial mammals. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 364: 537-548. Caton J.S. and Dhuyvetter D.V. (1997). Influence of energy supplementation on grazing ruminants: requirements and responses. Journal of Animal Science 75: 533-542. Cena K. and Monteith J.L. (1975). Transfer processes in animal coats. I. Radiative transfer. Proceedings of the Royal Society of London. Series B, Biological Sciences 188: 377-393. Chamaill?-Jammes S., Fritz H. and Murindagomo F. (2007). Detecting climate changes of concern in highly variable environments: quantile regressions reveal that droughts worsen in Hwange National Park, Zimbabwe. Journal of Arid Environments 71: 321-326. 223 Charmantier A., McCleery R.H., Cole L.R., Perrins C., Kruuk L.E.B. and Sheldon B.C. (2008). Adaptive phenotypic plasticity in response to climate change in a wild bird population. Science 320: 800-803. Chesy G., Kadziela W., Lachowski A., Caputa M., Kozak W. and Wieczor H. (1980). Pyrogen fever and selective brain cooling in the ox. In: Contributions to Thermal Physiology. Szelenyi Z. and Szekely M. (Eds.), Pergamon Press, Oxford, UK, pp. 197-199. Choshniak I., Ben-Kohav N., Taylor C.R., Robertshaw D., Barnes R.J., Dobson A., Belkin V. and Shkolnik A. (1995). Metabolic adaptations for desert survival in the Bedouin goat. American Journal of Physiology ? Regulatory Integrative and Comparative Physiology 268: R1101-1110. Chown S.L., Gaston K.J. and Robinson D. (2004a). Macrophysiology: large-scale patterns in physiological traits and their ecological implications. Functional Ecology 18: 159-167. Chown S.L., Sinclair B.J., Leinaas H.P. and Gaston K.J. (2004b). Hemispheric asymmetries in biodiversity ? a serious matter for ecology. PLoS Biology 2: 1701?1707. Clapperton J.L., Joyce J.P. and Blaxter K.L. (1965). Estimates of the contribution of solar radiation to the thermal exchanges of sheep at a latitude of 55? north. Journal of Agricultural Science 64: 37-49. Clarke A. and Rothery P. (2008). Scaling of body temperature in mammals and birds. Functional Ecology 22: 58-67. Cloudsley-Thompson J.L. (1979). Adaptive functions of the colours of desert animals. Journal of Arid Environments 2: 95-104. Clutton-Brock T.H., Wilson K. and Stevenson I.R. (1997). Density-dependent selection on horn phenotype in Soay sheep. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 352: 839-850. Codron D., Brink J.S., Rossouw L. and Clauss M. (2008). The evolution of ecological specialization in southern African ungulates: competition- or physical environmental turnover? Oikos 117: 344-353. Cooke S.J., Hinch S.G., Wikelski M., Andrews R.D., Kuchel L.J., Wolcott T.G. and Butler P.J. (2004). Biotelemetry: a mechanistic approach to ecology. Trends in Ecology and Evolution 19: 334-343. Corp N., Spalton A. and Gorman M.L. (1998). The influence of rainfall on range in a female desert ungulate: the Arabian oryx (Oryx leucoryx) in the Sultanate of Oman. Journal of Zoology 246: 369- 377. Costa D.P. and Sinervo B. (2004). Field physiology: physiological insights from animals in nature. Annual Review of Physiology 66: 209-238. Cowles R.B. (1967). Black pigmentation: adaptation for concealment or heat conservation? Science 158: 1340-1341. 224 Cronj? P.B. (1992). Glucose metabolism and adrenal function in goats bred for fibre production (Angora goat) or meat production (Boer goat). South African Journal of Animal Science 22: 149-153. Cronj? P.B. (1995). Differences in glucose response to insulin stimulation and fasting between Angora goats of two different phenotypes. Small Ruminant Research 16: 121-127. Crozier L. and Dwyer G. (2006). Combining population-dynamic and ecophysiological models to predict climate-induced insect range shifts. American Naturalist 167: 853-866. Cumming G.S. and van Vuuren D.P. (2006). Will climate change affect ectoparasite species ranges? Global Ecology and Biogeography 15: 486-497. Cunningham P., Sher Shah M., ul-Islam Z., Robinson R. and al Boug A. (2008). Adaptive management a prerequisite for the reintroduction of sand gazelle in Saudi Arabia. In: Gnusletter vol 27. Newsletter of the IUCN/SSC Antelope Specialist Group. Shurter S. (Ed.), White Oak Conservation Center, Yulee, Florida, USA, pp. 19-22. Curtis D.J. and Rasmussen M.A. (2006). The evolution of cathemerality in primates and other mammals: a comparative and chronoecological approach. Folia Primatologica 77: 178-193. Dantzer R. (2004). Cytokine-induced sickness behaviour: a neuroimmune response to activation of innate immunity. European Journal of Pharmacology 500: 399-411. Davidson I. and Simkanin C. (2008). Skeptical of assisted colonization. Science 322: 1048-1049. Davies H.L. and Southey I.N. (2001). Effects of grazing management and stocking rate on pasture production, ewe liveweight, ewe fertility and lamb growth on subterranean clover-based pasture in Western Australia. Australian Journal of Experimental Agriculture 41: 161-168. Davies R.A.G. and Skinner J.D. (1986). Temporal activity patterns of springbok Antidorcas marsupialis and Merino sheep Ovis aries during a Karoo drought. Transactions of the Royal Society of Southern Africa 46: 133-147. Davis A.J., Jenkinson L.S., Lawton J.H., Shorrocks B. and Wood S. (1998). Making mistakes when predicting shifts in species range in response to global warming. Nature 391: 783-786. Davis M.B. and Shaw R.G. (2001). Range shifts and adaptive responses to Quaternary climate change. Science 292: 673-679. Davis M.B., Shaw R.G. and Etterson J.R. (2005). Evolutionary responses to changing climate. Ecology 86: 1704?1714. Dawson T.J. and Brown G.D. (1970). A comparison of the insulative and reflective properties of the fur of desert kangaroos. Comparative Biochemistry and Physiology 37: 23-38. 225 Dawson T.J. and Maloney S.K. (2004). Fur versus feathers: the different roles of red kangaroo fur and emu feathers in thermoregulation in the Australian arid zone. Australian Mammalogy 26: 145-151. Dawson T.J., Blaney C.E., McCarron H.C.K and Maloney S.K. (2007). Dehydration, with and without heat, in kangaroos from mesic and arid habitats: different thermal responses including varying patterns in heterothermy in the field and laboratory. Journal of Comparative Physiology B: Biochemical, Systemic, and Environmental Physiology 177: 797-807. Dawson W.R. (1992). Physiological responses of animals to higher temperatures. In: Global Warming and Biological Diversity. Peters R.L. and Lovejoy T.E. (Eds.), Yale University Press, London, UK, pp. 158-170. Dean W.R.J., Hoffman M.T., Meadows M.E. and Milton S.J. (1995). Desertification in the semi-arid Karoo, South Africa: review and reassessment. Journal of Arid Environments 30: 247-264. Demas G.E., Chefer V., Talan M.I. and Nelson R.J. (1997). Metabolic costs of mounting an antigen- stimulated immune response in adult and aged C57BL/6J mice. American Journal of Physiology ? Regulatory Integrative and Comparative Physiology 273: R1631-1637. Demment M.W. and Van Soest P.J. (1985). A nutritional explanation for body-size patterns of ruminant and non-ruminant herbivores. American Naturalist 125: 641-672. Derocher A.E., Lunn N.J. and Stirling I. (2004). Polar bears in a warming climate. Integrative and Comparative Biology 44: 163-176. Deutsch C.A., Tewksbury J.J., Huey R.B., Sheldon K.S., Ghalambor C.K., Haak D.C. and Martin P.R. (2008). Impacts of climate warming on terrestrial ectotherms across latitude. Proceedings of the National Academy of Sciences of the United States of America 105: 6668-6672. Dmi'el R. (1986). Selective sweat secretion and panting modulation in dehydrated goats. Journal of Thermal Biology 11: 157-159. Dmi'el R., Prevulotzky A. and Shkolnik A. (1980). Is a black coat in the desert a means of saving metabolic energy? Nature 283: 76 -762. Dole K.P., Loik M.E. and Sloan L.C. (2003). The relative importance of climate change and the physiological effects of CO2 on freezing tolerance for the future distribution of Yucca brevifolia. Global and Planetary Change 36: 137-146. Doris P.A. and Baker M.A. (1981). Effects of dehydration on thermoregulation in cats exposed to high ambient temperatures. Journal of Applied Physiology 51: 46-54. Downing B.H. (1978). Environmental consequences of agricultural expansion in South Africa since 1850. South African Journal of Science 74: 420-422. 226 du Toit J.T. (1993). The feeding ecology of a very small ruminant, the steenbok (Raphicerus campestris). African Journal of Ecology 31: 35-48. du Toit J.T. and Yetman C.A. (2005). Effects of body size on the diurnal activity budgets of African browsing ruminants. Oecologia 143: 317-325. Dudzinski M.L. and Arnold G.W. (1979). Factors influencing the grazing behaviour of sheep in a Mediterranean climate. Applied Animal Ethology 5: 125-144. Dussault C., Ouellet J.P., Courtois R., Huot J., Breton L. and Larochelle J. (2004). Behavioural responses of moose to thermal conditions in the boreal forest. Ecoscience 11: 321-328. Elliot S.L., Blanford S. and Thomas M.B. (2002). Host-pathogen interactions in a varying environment: temperature, behavioural fever and fitness. Proceedings of the Royal Society of London. Series B, Biological Sciences 269: 1599-1607. Epaphras A.M., Gereta E., Lejora I.A., Ole Meing?ataki G.E., Ng?umbi G., Kiwango Y., Mwangomo E., Semanini F., Vitalis L., Balozi J. and Mtahiko M.G.G. (2008). Wildlife water utilization and importance of artificial waterholes during dry season at Ruaha National Park, Tanzania. Wetlands Ecology and Management 6: 183-188. Erasmus B.F.N., van Jaarsveld A.S., Chown S.L., Kshatriya M. and Wessels K.J. (2002). Vulnerability of South African animal taxa to climate change. Global Change Biology 8: 679-693. Erasmus J.A. (2000). Adaptation to various environments and resistance to disease of the improved Boer goat. Small Ruminant Research 36: 179-187. Estes R.D. (1991). Introduction to the ruminants suborder Ruminatia. In: The Behavior Guide to African Mammals incl. Hooved Mammals, Carnivores, Primates. Russel Friedman Books, Halfway House, South Africa, pp. 3-25, 81-84. Etterson J.R. and Shaw R.G. (2001). Constraint to adaptive evolution in response to global warming. Science 294: 151-154. Ewald P.W. (1980). Evolutionary biology and the treatment of signs and symptoms of infectious disease. Journal of Theoretical Biology 86: 169-176. Fabricius C. and Mentis M.T. (1992). Modelling the habitat relations of kudu in arid savanna. South African Journal of Science 88: 280-284. Feder M.E. and Block B.A. (1991). On the future of animal physiological ecology. Functional Ecology 5: 136-144. Feder M.E., Bennett A.F. and Huey R.B. (2000). Evolutionary physiology. Annual Review of Ecology and Systematics 31: 315-341. 227 Fensham R.J. and Fairfax R.J. (2008). Water-remoteness for grazing relief in Australian arid-lands. Biological Conservation 141: 1447-1460. Fick L., Matthee A., Mitchell D. and Fuller A. (2006). The effect of boma-housing and long-acting tranquillizers on body temperature, physical activity and food intake of blue wildebeest (Connochaetes taurinus). Journal of Thermal Biology 31: 159-167. Finch V.A. (1972). Thermoregulation and heat balance of the east African eland and hartebeest. American Journal of Physiology 222: 1374-1379. Finch V.A. and Robertshaw D. (1979). Effect of dehydration on thermoregulation in eland and hartebeest. American Journal of Physiology ? Regulatory Integrative and Comparative Physiology 237: R192-196. Finch V.A., Bennett I.L. and Holmes C.R. (1984). Coat colour in cattle: effect on thermal balance, behaviour and growth, and relationship with coat type. Journal of Agricultural Science (Cambridge) 102: 141-147. Finch V.A., Dmi'el R., Boxman R., Shkolnik A. and Taylor C.R. (1980). Why black goats in hot deserts? Effects of coat colour on heat exchanges of wild and domestic goats. Physiological Zoology 53: 19-25. Forcada J., Trathan P.N. and Murphy E.J. (2008). Life history buffering in Antarctic mammals and birds against changing patterns of climate and environmental variation. Global Change Biology 14: 2473-2488. Forchhammer M.C., Clutton-Brock T.H., Lindstr?m J. and Albon S.D. (2001). Climate and population density induce long-term cohort variation in a northern ungulate. Journal of Animal Ecology 70: 721- 729. Fourie T.J. (1984). A comparative study of the effects of cold exposure on the heat production of Angora (Capra aegagrus) and Boer (Capra hircus) goats ('n Vergelykende studie van die effek van koue bloodstelling op die hitteproduksie van angora- (Capra aegagrus) en boerbokke (Capra hircus)). MSc, Universiteit van Port Elizabeth, Port Elizabeth, South Africa. Franklin J. (1995). Predictive vegetation mapping: geographic modelling of biospatial patterns in relation to environmental gradients. Progress in Physical Geography 19: 474-499. Fukuta K., Kudo H., Sasaki M., Kimura J., Ismail D.b. and Endo H. (2007). Absence of carotid rete mirabile in small tropical ruminants: implications for the evolution of the arterial system in artiodactyls. Journal of Anatomy 210: 112-116. Fuller A., Maloney S.K., Mitchell G. and Mitchell D. (2004). The eland and the oryx revisited: body and brain temperatures of free-living animals. International Congress Series 1275: 275-282. 228 Fuller A., Meyer L.C.R., Mitchell D. and Maloney S.K. (2007). Dehydration increases the magnitude of selective brain cooling independently of core temperature in sheep. American Journal of Physiology ? Regulatory Integrative and Comparative Physiology 293: 438-446. Fuller A., Maloney S., Kamerman P., Mitchell G. and Mitchell D. (2000). Absence of selective brain cooling in free-ranging zebras in their natural habitat. Experimental Physiology 85: 209-217. Fuller A., Hetem R.S., Meyer L.C.R., Mitchell D. and Maloney S.K. (2008). Selective brain cooling: a physiological mechanism for coping with aridity? In: 4th Comparative Physiologists & Biochemists meeting in Africa. "Molecules to Migration: The Pressures of Life". Morris S. and Vosloo A. (Eds.), Medimond Publishing S.r.l., Via Maserati 5, 40128 Bologna, Italy, Maasai Mara Game Reserve, Kenya, pp. 375-382. Fuller A., Moss D.G., Skinner J.D., Jessen P.T., Mitchell G. and Mitchell D. (1999). Brain, abdominal and arterial blood temperatures of free-ranging eland in their natural habitat. Pfl? gers Archiv - European Journal of Physiology 438: 671-680. Fuller A., Kamerman P.R., Maloney S.K., Matthee A., Mitchell G. and Mitchell D. (2005). A year in the thermal life of a free-ranging herd of springbok Antidorcas marsupialis. Journal of Experimental Biology 208: 2855-2864. Fusch C. (2000). Measurement of water turnover using deuterium dilution: impact of periodically varying turnover rates on precision and accuracy. Clinical Chemistry & Laboratory Medicine 38: 961- 964. Gannon W.L., Sikes R.S. and the Animal Care and Use Committee of the American Society of Mammalogists (2007). Guidelines of the American Society of Mammalogists for the use of wild mammals in research. Journal of Mammalogy 88: 809-823. Garant D., Forde S.E. and Hendry A.P. (2007). The multifarious effects of dispersal and gene flow on contemporary adaptation. Functional Ecology 21: 434-443. Garland T., Jr. and Kelly S.A. (2006). Phenotypic plasticity and experimental evolution. Journal of Experimental Biology 209: 2344-2361. Gates D.M. (1966). Spectral distribution of solar radiation at the earth's surface. Science 151: 523-529. G?nin F. and Perret M. (2003). Daily hypothermia in captive grey mouse lemurs (Microcebus murinus): effects of photoperiod and food restriction. Comparative Biochemistry and Physiology B: Biochemistry and Molecular Biology 136: 71-81. Gienapp P., Postma E. and Visser M.E. (2006). Why breeding time has not responded to selection for earlier breeding in a songbird population. Evolution 60: 2381-2388. Gienapp P., Teplitsky C., Alho J.S., Mills J.A. and Meril? J. (2008). Climate change and evolution: disentangling environmental and genetic responses. Molecular Ecology 17: 167-178. 229 Gilman S.E., Wethey D.S. and Helmuth B. (2006). Variation in the sensitivity of organismal body temperature to climate change over local and geographic scales. Proceedings of the National Academy of Sciences of the United States of America 103: 9560-9565. Gilmour K.M., Wilson R.W. and Sloman K.A. (2005). The integration of behaviour into comparative physiology. Physiological and Biochemical Zoology 78: 669-678. Gitay H., Noble I.R., Pilifosova O., Alijani B. and Safriel U.N. (1998). Middle East and arid Asia. In: The Regional Impacts of Climate Change: An Assessment of Vulnerability. Watson R.T., Zinyowera M.C., Moss R.H. and Intergovernmental Panel on Climate Change. Working Group II (Eds.), Cambridge University Press, Cambridge, UK, pp. 233-250. Goldstein D.L. and Pinshow B. (2006). Taking physiology to the field: using physiological approaches to answer questions about animals in their environments. Physiological and Biochemical Zoology 79: 237-241. Gomulkiewicz R. and Holt R.D. (1995). When does evolution by natural selection prevent extinction? Evolution 49: 201-207. Gordon C.J. (1994). 24-hour control of body temperature in the rat. II: Diisopropyl fluorophosphate- induced hypothermia and hyperthermia. Pharmacology Biochemistry and Behavior 49: 747-754. Gourine A.V., Rudolph K., Tesfaigzi J. and Kluger M.J. (1998). Role of hypothalamic interleukin-1 in fever induced by cecal ligation and puncture in rats. American Journal of Physiology ? Regulatory Integrative and Comparative Physiology 275: R754-761. Grabherr G., Gottfried M. and Pauli H. (1994). Climatic effects on mountain plants. Nature 369: 448. Graham N.M. (1962). Energy expenditure of grazing sheep. Nature 196: 289-289. Graham R.W. and Grimm E.C. (1990). Effects of global climate change on the patterns of terrestrial biological communities. Trends in Ecology and Evolution 5: 289-292. Graham R.W., Lundelius E.L., Jr, Graham M.A., Schroeder E.K., Toomey R.S.I., Anderson E., Barnosky A.D., Burns J.A., Churcher C.S., Grayson D.K., Guthrie R.D., Harington C.R., Jefferson G.T., Martin L.D., McDonald H.G., Morlan R.E., Holmes A.S., Jr, Webb S.D., Werdelin L. and Wilson M.C. (1996). Spatial response of mammals to late Quaternary environmental fluctuations. Science 272: 1601-1606. Gratten J., Wilson A.J., McRae A.F., Beraldi D., Visscher P.M., Pemberton J.M. and Slate J. (2008). A localized negative genetic correlation constrains microevolution of coat color in wild sheep. Science 319: 318-320. Gremillet D., Kuntz G., Woakes A.J., Gilbert C., Robin J.-P., Le Maho Y. and Butler P.J. (2005). Year-round recordings of behavioural and physiological parameters reveal the survival strategy of a poorly insulated diving endotherm during the Arctic winter. Journal of Experimental Biology 208: 4231-4241. 230 Grenot C.J. (1992). Ecophysiological characteristics of large herbivorous mammals in arid Africa and the Middle East. Journal of Arid Environments 23: 125-155. Groves C.P. (1988). A catalogue of the genus Gazella. In: Conservation and Biology of Desert Antelope, including the proceedings of the 25th anniversary celebration of ?operation oryx? symposium. Dixon A. and Jones D. (Eds.), Christopher Helm, London, UK, pp. 193-198 Guisan A. and Zimmermann N.E. (2000). Predictive habitat distribution models in ecology. Ecological Modelling 135: 147-186. Guisan A. and Thuiller W. (2005). Predicting species distribution: offering more than simple habitat models. Ecology Letters 8: 993-1009. Hales J.R., Bennett J.W., Baird J.A. and Fawcett A.A. (1973). Thermoregulatory effects of prostaglandins E1, E2, F1 and F2 in the sheep. Pfl?gers Archiv - European Journal of Physiology 339: 125-133. Hallett T.B., Coulson T., Pilkington J.G., Clutton-Brock T.H., Pemberton J.M. and Grenfell B.T. (2004). Why large-scale climate indices seem to predict ecological processes better than local weather. Nature 430: 71-75. Hamilton W.J. and Heppner F. (1967). Radiant solar energy and the function of black homeotherm pigmentation: an hypothesis. Science 155: 196-197. Hammel H.T. (1968). Regulation of internal body temperature. Annual Review of Physiology 30: 641- 695. Hannah L., Midgley G.F. and Millar D. (2002a). Climate change-integrated conservation strategies. Global Ecology and Biogeography 11: 485-495. Hannah L., Midgley G.F., Lovejoy T., Bond W.J., Bush M., Lovett J.C., Scott D. and Woodward F.I. (2002b). Conservation of biodiversity in a changing climate. Conservation Biology 16: 264-268. Hansen M.C. (1982). Desert bighorn sheep: another view. Wildlife Society Bulletin 10: 133-140. Hansen P.J. (1990). Effects of coat colour on physiological responses to solar radiation in Holsteins. Veterinary Record 127: 333-334. Hanson J.D., Baker B.B. and Bourdon R.M. (1993). Comparison of the effects of different climate change scenarios on rangeland livestock production. Agricultural Systems 41: 487-502. Harrington R., Woiwod I. and Sparks T. (1999a). Climate change and trophic interactions. Trends in Ecology and Evolution 14: 146-150. 231 Harrington R., Owen-Smith N., Viljoen P.C., Biggs H.C., Mason D.R. and Funston P. (1999b). Establishing the causes of the roan antelope decline in the Kruger National Park, South Africa. Biological Conservation 90: 69-79. Harrison D.L. and Bates P.J.J. (1991). The Mammals of Arabia. Harrison Zoological Museum, England, UK. Hart B.L. (1988). Biological basis of the behavior of sick animals. Neuroscience and Biobehavioral Reviews 12: 123-137. Hart S.P., Sahlu T. and Fernandez J.M. (1993). Efficiency of utilization of high and low quality forage by three goat breeds. Small Ruminant Research 10: 293-301. Harthoorn A.M., Finch V.A., Hopcraft D. and McGinnis S.M. (1970). Adaptation to solar radiation by African large herbivores. Journal of the South African Veterinary Medical Association 41: 17-24. Hattingh J., Ganhao M.F., Kruger F.J., De Vos V. and Kay G.W. (1988). Remote controlled sampling of cattle and buffalo blood. Comparative Biochemistry and Physiology A: Comparative Physiology 89: 231-235. Hawkins B.A., Field R., Cornell H.V., Currie D.J., Gu?gan J.-F., Kaufman D.M., Kerr J.T., Mittelbach G.G., Oberdorff T., O?Brien E.M., Porter E.E. and Turner J.R.G. (2003). Energy, water, and broad- scale geographic patterns of species richness. Ecology 84: 3105?3117. Hayes C.L. and Krausman P.R. (1993). Nocturnal activity of female desert mule deer. Journal of Wildlife Management 57: 897-904. Hebert J., Lust A., Fuller A., Maloney S.K., Mitchell D. and Mitchell G. (2008). Thermoregulation in pronghorn antelope (Antilocapra americana, Ord) in winter. Journal of Experimental Biology 211: 749-756. Heikkinen R.K., Luoto M., Araujo M.B., Virkkala R., Thuiller W. and Sykes M.T. (2006). Methods and uncertainties in bioclimatic envelope modelling under climate change. Progress in Physical Geography 30: 751-777. Helmuth B., Kingsolver J.G. and Carrington E. (2005). Biophysics, physiological ecology, and climate change: does mechanism matter? Annual Review of Physiology 67: 177-201. Hendry A.P. and Kinnison M.T. (1999). Perspective: the pace of modern life: measuring rates of contemporary microevolution. Evolution 53: 1637-1653. Hendry A.P. and Kinnison M.T. (2001). An introduction to microevolution: rate, pattern, process. Genetica 112-113: 1-8. Henley S.R. (2001). Habitat suitability and modeling for ungulates in the thicket biome, Eastern Cape. PhD, University of Port Elizabeth, Port Elizabeth, South Africa. 232 Henley S.R., Ward D. and Schmidt I. (2007). Habitat selection by two desert-adapted ungulates. Journal of Arid Environments 70: 39-48. Heppner F. (1970). The metabolic significance of differential absorption of radiant energy by black and white birds. Condor 72: 50-59. Herselman M.J., Hart S.P., Sahlu T., Coleman S.W. and Goetsch A.L. (1999). Heat energy for growing goats and sheep grazing different pastures in the summer. Journal of Animal Science 77: 1258-1265. Hersteinsson P. and MacDonald D.W. (1992). Interspecific competition and the geographical distribution of red and arctic foxes Vulpes vulpes and Alopex lagopus. Oikos 64: 505-515. Hetem R.S., Maloney S.K., Fuller A., Meyer L.C.R. and Mitchell D. (2007). Validation of a biotelemetric technique, using ambulatory miniature black globe thermometers, to quantify thermoregulatory behaviour in ungulates. Journal of Experimental Zoology A: Comparative and Experimental Biology 307: 342?356. Hetem R.S., de Witt B.A., Fick L.G., Fuller A., Kerley G.I.H., Maloney S.K., Meyer L.C.R. and Mitchell D. (2009). Summer shearing affects body temperature of Angora goats (Capra aegagrus) more than does winter shearing. Animal 3: 1025-1036. Hicks T.A., McGlone J.J., Whisnant C.S., Kattesh H.G. and Norman R.L. (1998). Behavioral, endocrine, immune, and performance measures for pigs exposed to acute stress. Journal of Animal Science 76: 474-483. Hill J.K., Thomas C.D. and Huntley B. (1999). Climate and habitat availability determine 20th century changes in a butterfly's range margin. Proceedings of the Royal Society of London. Series B, Biological Sciences 266: 1197-1206. Hill R.A. (2006). Why be diurnal? Or, why not be cathemeral? Folia Primatologica: International Journal of Primatology 77: 72-86. Hodkinson I.D. (1999). Species response to global environmental change or why ecophysiological models are important: a reply to Davis et al. Journal of Animal Ecology 68: 1259-1262. Hoegh-Guldberg O., Hughes L., McIntyre S., Lindenmayer D.B., Parmesan C., Possingham H.P. and Thomas C.D. (2008). Ecology: assisted colonization and rapid climate change. Science 321: 345-346. Hoffman M.T. and Cowling R.M. (1990). Desertification in the lower Sundays River Valley, South Africa. Journal of Arid Environments 19: 105-117. Hofmeyr H.S., Joubert D.M., Badenhorst F.J.G. and van D Steyn G.J. (1965). Adaptability of sheep and goats to a South African tropical environment. Proceedings of the South African Society of Animal Production 4: 191-195. 233 Hofmeyr M.D. (1981). Thermal physiology of selected African ungulates with emphasis on the physical properties of the pelage. PhD, University of Cape Town, Cape Town, South Africa. Hofmeyr M.D. and Louw G.N. (1987). Thermoregulation, pelage conductance and renal function in the desert-adapted springbok, Antidorcas marsupialis. Journal of Arid Environments 13: 137-151. Hohtola E., Hissa R., Py?rnil? A., Rintam?ki H. and Saarela S. (1991). Nocturnal hypothermia in fasting Japanese quail: the effect of ambient temperature. Physiology & Behavior 49: 563-567. Holt R.D. (1990). The microevolutionary consequences of climate change. Trends in Ecology and Evolution 5: 311-315. Hopkins P.S., Knights G.I. and Feuvre A.S.L. (1978). Studies of the environmental physiology of tropical Merinos. Australian Journal of Agricultural Research 29: 161-171. Hori T. and Harada Y. (1974). The effects of ambient and hypothalamic temperatures on the hyperthermic responses to prostaglandins E1 and E2. Pfl?gers Archiv - European Journal of Physiology 350: 123-134. Huey R. (1991). Physiological consequences of habitat selection. American Naturalist 137: S91-S112. Huey R., Hertz P. and Sinervo B. (2003). Behavioural drive versus behavioural inertia in evolution: a null model approach. American Naturalist 161: 357-366. Huey R.B. and Tewksbury J.J. (2009). Can behavior douse the fire of climate warming? Proceedings of the National Acadamy of Sciences of the United States of America 106: 3647-3648. Hughes L. (2000). Biological consequences of global warming: is the signal already apparent? Trends in Ecology and Evolution 15: 56-61. Humphries M.M., Thomas D.W. and Speakman J.R. (2002). Climate-mediated energetic constraints on the distribution of hibernating mammals. Nature 418: 313-316. Humphries M.M., Umbanhowar J. and McCann K.S. (2004). Bioenergetic prediction of climate change impacts on northern mammals. Integrative and Comparative Biology 44: 152-162. Hunter M. (2007). Climate change and moving species: furthering the debate on assisted colonization. Conservation Biology 21: 1356-1358. Huntley B. (2007). Evolutionary response to climatic change? Heredity 98: 247?248. Huntley B., Green R.E., Collingham Y.C., Hill J.K., Willis S.G., Bartlein P.J., Cramer W., Hagemeijer W.J.M. and Thomas C.J. (2004). The performance of models relating species geographical distributions to climate is independent of trophic level. Ecology Letters 7: 417-426. 234 Hurwitz S.H. and Whipple G.H. (1917). Studies on the blood proteins: ii. The albumin-globulin ratio in experimental intoxications and infections. Journal of Experimental Medicine 25: 231-253. Huston J.E. (1994). Effects of supplemental feeding on intake by kid, yearling, and adult Angora goats on rangeland. Journal of Animal Science 72: 768-773. Huston J.E., Thompson P.V. and Taylor C.A., Jr. (1993a). Combined effects of stocking rate and supplemental feeding level on adult beef cows grazing native rangeland in Texas. Journal of Animal Science 71: 3458-3465. Huston J.E., Taylor C.A., Lupton C.J. and Brooks T.D. (1993b). Effects of supplementation on intake, growth rate, and fleece production by female Angora kid goats grazing rangeland. Journal of Animal Science 71: 3124-3130. Hutchinson J.C. and Brown G.D. (1969). Penetrance of cattle coats by radiation. Journal of Applied Physiology 26: 454-464. IPCC (2007). Climate Change 2007: The Physical Science Basis. In: Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Solomon S., Qin D., Manning M., Chen Z., Marquis M., Averyt K.B., Tignor M. and Miller H.L. (Eds.), Cambridge University Press, Cambridge, UK and New York, USA. Isaac J.L., Vanderwal J., Williams S.E. and Johnson C.N. (2009). Resistance and resilience: quantifying relative extinction risk in a diverse assemblage of Australian tropical rainforest vertebrates. Diversity and Distributions 15: 280-288. IUPS Thermal Commission (2003). Glossary of terms for thermal physiology. Journal of Thermal Biology 28: 75-106. Jackson S.T. and Overpeck J.T. (2000). Responses of plant populations and communities to environmental changes of the late Quaternary. Paleobiology 26: 194-220. Janis C.M. (1989). A climatic explanation for patterns of evolutionary diversity in ungulate mammals. Palaeontology 32: 463-481. Jarman P.J. (1977). Behavior of topi in a shadeless environment. Zoologica Africana 12: 101-111. Jeschke J.M. and Tollrian R. (2005). Predicting herbivore feeding times. Ethology 111: 187-206. Jessen C. (1998). Brain cooling: an economy mode of temperature regulation in artiodactyls. News in Physiological Science 13: 281-286. Jessen C. (2001). Selective brain cooling in mammals and birds. Japanese Journal of Physiology 51: 291?301. 235 Jessen C. and Kuhnen G. (1996). Seasonal variations of body temperature in goats living in an outdoor environment. Journal of Thermal Biology 21: 197-204. Jessen C., Dmi'el R., Choshniak I., Ezra D. and Kuhnen G. (1997). Effects of dehydration and rehydration on selective brain cooling in the black Bedouin goat. In: Proceedings of the International Symposium on Thermal Physiology. Johannsen B.N. and Nielsen R. (Eds.), Copenhagen, Denmark, pp. 107-110. Jessen C., Dmi'el R., Choshniak I., Ezra D. and Kuhnen G. (1998). Effects of dehydration and rehydration on body temperatures in the black Bedouin goat. Pfl?gers Archiv - European Journal of Physiology 436: 659-666. Jessen C., Laburn H.P., Knight M.H., Kuhnen G., Goelst K. and Mitchell D. (1994). Blood and brain temperatures of free-ranging black wildebeest in their natural environment. American Journal of Physiology ? Regulatory Integrative and Comparative Physiology 267: R1528-1536. Jiguet F., Julliard R., Thomas C.D., Dehorter O., Newson S.E. and Couvet D. (2006). Thermal range predicts bird population resilience to extreme high temperatures. Ecology Letters 9: 1321-1330. Johnson R.W. (2002). The concept of sickness behavior: a brief chronological account of four key discoveries. Veterinary Immunology and Immunopathology 87: 443-450. Johnston K. and Schmitz O. (1997). Wildlife and climate change: assessing the sensitivity of selected species to simulated doubling of atmospheric CO2. Global Change Biology 3: 531-544. Kamau J.M. and Maloiy G.M. (1985). Thermoregulation and heat balance in the dik-dik antelope (Rhynchotragus kirki): a field and laboratory study. Comparative Biochemistry and Physiology A: Molecular and Integrative Physiology 81: 335-340. Kamerman P.R., Fuller A., Faurie A.S., Mitchell G. and Mitchell D. (2001). Body temperature patterns during natural fevers in a herd of free-ranging impala (Aepyceros melampus). Veterinary Record 149: 26-27. Katz R.W. and Brown B.G. (1992). Extreme events in a changing climate: variability is more important than averages. Climatic Change 21: 289-302. Kay R.N.B. (1997). Responses of African livestock and wild herbivores to drought. Journal of Arid Environments 37: 683-694. Kearney M. and Predavec M. (2000). Do nocturnal ectotherms thermoregulate? A study of the temperate gecko Christinus marmoratus. Ecology 81: 2984-2996. Kearney M. and Porter W.P. (2004). Mapping the fundamental niche: physiology, climate, and the distribution of a nocturnal lizard. Ecology 85: 3119-3131. 236 Kearney M., Shine R. and Porter W.P. (2009). The potential for behavioral thermoregulation to buffer "cold-blooded" animals against climate warming. Proceedings of the National Acadamy of Sciences of the United States of America 106: 3835-3840. Kerley G.I.H., Knight M.H. and Kock M. (1995). Desertification of subtropical thicket in the Eastern Cape, South Africa: are there alternatives? Environmental Monitoring and Assessment 37: 211-230. Kerr J. and Packer L. (1998). The impact of climate change on mammal diversity in Canada. Environmental Monitoring and Assessment 49: 263-270. Kingsolver J.G. and Huey R.B. (1998). Evolutionary analyses of morphological and physiological plasticity in thermally variable environments. American Zoologist 38: 545-560. Kingsolver J.G., Hoekstra H.E., Hoekstra J.M., Berrigan D., Vignieri S.N., Hill C.E., Hoang A., Gibert P. and Beerli P. (2001). The strength of phenotypic selection in natural populations. American Naturalist 157: 245-261. Kinnison M.T. and Hendry A.P. (2001). The pace of modern life II: from rates of contemporary microevolution to pattern and process. Genetica 112-113: 145-164. Klein D.R. and Fairall N. (1986). Associated energetics of impala and blesbok. Journal of Applied Ecology 23: 489-502. Kluger M.J. (1978). The evolution and adaptive value of fever. American Scientist 66: 38-43. Kluger M.J. (1979). Phylogeny of fever. Federation Proceedings 38: 30-34. Kluger M.J. (1986). Is fever beneficial? Yale Journal of Biology and Medicine 59: 89-95. Kluger M.J. (1991). Fever: role of pyrogens and cryogens. Physiological Reviews 71: 93-127. Kluger M.J., Kozak W., Conn C.A., Leon L.R. and Soszynski D. (1996). The adaptive value of fever. Infectious Disease Clinics of North America 10: 1-20. Kluger M.J., Kozak W., Conn C.A., Leon L.R. and Soszynski D. (1998). Role of fever in disease. Annuals of the New York Academy of Sciences 856: 224-233. Konsman J.P., Parnet P. and Dantzer R. (2002). Cytokine-induced sickness behaviour: mechanisms and implications. Trends in Neuroscience 25: 154-159. Kovarik M. (1964). Flow of heat in an irradiated protective cover. Nature 201: 1085-1087. Kozak W., Conn C.A. and Kluger M.J. (1994). Lipopolysaccharide induces fever and depresses locomotor activity in unrestrained mice. American Journal of Physiology ? Regulatory Integrative and Comparative Physiology 266: R125-135. 237 Kronfeld D.S. (1993). Starvation and malnutrition of horses: recognition and treatment. Journal of Equine Veterinary Science 13: 298-304. Kruger J.H. (1976). An investigation of certain dermatological, morphological and genetic aspects of the black springbok and the white springbok (Antidorcas marsupialis) [in Afrikaans]. BSc Honours Report, University of Pretoria, Pretoria, South Africa. Kruger J.H., Skinner J.D. and Robinson T.J. (1979). On the taxonomic status of the black and white springbok, Antidorcas marsupialis. South African Journal of Science 75: 411-412. Kuhnen G. (1994). Selective brain cooling during fever. In: Thermal Balance in Health and Disease: Advances in Pharmacological Sciences. Zeisberger E., Schonbaum E. and Lomax P. (Eds.), Birkhauser Verlag, Basel, Switzerland, pp. 353-358. Kuhnen G. (1997). Selective brain cooling reduces respiratory water loss during heat stress. Comparative Biochemistry and Physiology A: Molecular and Integrative Physiology 118: 891-895. Kuhnen G. and Jessen C. (1991). Threshold and slope of selective brain cooling. Pfl?gers Archiv - European Journal of Physiology 418: 176-183. Kuhnen G. and Jessen C. (1992). Effects of selective brain cooling on mechanisms of respiratory heat loss. Pfl?gers Archiv - European Journal of Physiology 421: 204-208. Kuhnen G. and Jessen C. (1994). Thermal signals in control of selective brain cooling. American Journal of Physiology ? Regulatory Integrative and Comparative Physiology 267: R355-359. La Sorte F.A. and Thompson F.R.I. (2007). Poleward shifts in winter ranges of North American birds. Ecology 88: 1803-1812. Laburn H.P., Mitchell D., Mitchell G. and Saffy K. (1988). Effects of tracheostomy breathing on brain and body temperatures in hyperthermic sheep. Journal of Physiology (London) 406: 331-344. Lachica M. and Aguilera J.F. (2003). Estimation of energy needs in the free-ranging goat with particular reference to the assessment of its energy expenditure by the 13C-bicarbonate method. Small Ruminant Research 49: 303-318. Lachica M. and Aguilera J.F. (2005a). Energy needs of the free-ranging goat. Small Ruminant Research 60: 111-125. Lachica M. and Aguilera J.F. (2005b). Energy expenditure of walk in grassland for small ruminants. Small Ruminant Research 59: 105-121. Le Houerou H.N. (1996). Climate change, drought and desertification. Journal of Arid Environments 34: 133-185. 238 Lechmere-Oertel R.G. (2003). The effects of goat browsing on ecosystem patterns and processes in succulent thicket, South Africa. PhD, University of Port Elizabeth, Port Elizabeth, South Africa. Lechmere-Oertel R.G., Cowling R.M. and Kerley G.I.H. (2005a). Landscape dysfunction and reduced spatial heterogeneity in soil resources and fertility in semi-arid succulent thicket, South Africa. Austral Ecology 30: 615-624. Lechmere-Oertel R.G., Kerley G.I.H. and Cowling R.M. (2005b). Patterns and implications of transformation in semi-arid succulent thicket, South Africa. Journal of Arid Environments 62: 459-474. LeGrand E.K. and Brown C.C. (2002). Darwinian medicine: applications of evolutionary biology for veterinarians. Canadian Veterinary Journal 43: 556-559. Lell B., Brandts C.H., Graninger W. and Kremsner P.G. (2000). The circadian rhythm of body temperature is preserved during malarial fever. Wiener Klinische Wochenschrift 112: 1014-1015. Leuthold B.M. and Leuthold W. (1978). Daytime activity patterns of gerenuk and giraffe in Tsavo National Park, Kenya. East African Wildlife Journal 16: 231-243. Levinsky I., Skov F., Svenning J.-C. and Rahbek C. (2007). Potential impacts of climate change on the distributions and diversity patterns of European mammals. Biodiversity and Conservation 16: 3803- 3816. Lochmiller R.L. and Deerenberg C. (2000). Trade-offs in evolutionary immunology: just what is the cost of immunity? Oikos 88: 87-98. Lonsdale E.M., Bradach B. and Thorne E.T. (1971). A telemetry system to determine body temperature in pronghorn antelope. Journal of Wildlife Management 35: 747-751. Louw G.N. and Seely M.K. (1982). Ecology of Desert Organisms. Longman, New York, USA. Lovegrove B.G. and Heldmaier G. (1994). The amplitude of circadian body-temperature rhythms in 3 rodents (Aethomys-namaquensis, Thallomys-paedulcus and Cryptomys-damarensis) along an arboreal- subterranean gradient. Australian Journal of Zoology 42: 65-78. Lovejoy T.E. and Hannah L. (2005). Climate Change and Biodiversity. Yale University Press, New Haven, Connecticut, USA. Lowe T.E., Cook C.J., Ingram J.R. and Harris P.J. (2001). Impact of climate on thermal rhythm in pastoral sheep. Physiology & Behavior 74: 659-664. Lu C.D. (1988). Grazing behavior and diet selection of goats. Small Ruminant Research 1: 205-216. Luker F.I., Mitchell D. and Laburn H.P. (2000). Fever and motor activity in rats following day and night injections of Staphylococcus aureus cell walls. American Journal of Physiology ? Regulatory Integrative and Comparative Physiology 279: R610-616. 239 Luoto M., P?yry J., Heikkinen R.K. and Saarinen K. (2005). Uncertainty of bioclimate envelope models based on the geographical distribution of species. Global Ecology and Biogeography 14: 575- 584. Lust A., Fuller A., Maloney S.K., Mitchell D. and Mitchell G. (2007). Thermoregulation in pronghorn antelope (Antilocapra americana Ord) in the summer. Journal of Experimental Biology 210: 2444- 2452. Lustick S. (1969). Bird energetics: effects of artificial radiation. Science 163: 387-389. Lyons S.K. (2003). A quantitative assessment of the range shifts of Pleistocene mammals. Journal of Mammalogy 84: 385-402. MacFarlane W.V., Howard B., Haines H., Kennedy P.J. and Sharpe C.M. (1971). Hierarchy of water and energy turnover of desert mammals. Nature 234: 483-484. Mack R.N. (1996). Predicting the identity and fate of plant invaders: emergent and emerging approaches. Biological Conservation 78: 107-121. Malcolm J.R. and Markham A. (2000). Global warming and terrestrial biodiversity decline. Report prepared for the WorldWildlife Fund for Nature, Gland, Switzerland. Malcolm J.R., Liu C., Neilson R.P., Hansen L. and Hannah L.E.E. (2006). Global warming and extinctions of endemic species from biodiversity hotspots. Conservation Biology 20: 538-548. Maloiy G.M. and Hopcraft D. (1971). Thermoregulation and water relations of two East African antelopes: the hartebeest and impala. Comparative Biochemistry and Physiology A: Molecular and Integrative Physiology 38: 525-534. Maloiy G.M.O. (1973a). Water metabolism of East African ruminants in arid and semi-arid regions. Journal of Animal Breeding and Genetics 90: 219-228. Maloiy G.M.O. (1973b). The water metabolism of a small east African antelope: the dik-dik. Proceedings of the Royal Society of London. Series B, Biological Sciences 184: 167-178. Maloney S.K. and Dawson T.J. (1995). The heat load from solar radiation on a large, diurnally active bird, the emu (Dromaius novaehollandiae). Journal of Thermal Biology 20: 381-387. Maloney S.K., Fuller A. and Mitchell D. (2009). Climate change: is the dark Soay sheep endangered? Biology Letters doi: 10.1098/rsbl.2009.0424. Maloney S.K., Moss G. and Mitchell D. (2005a). Orientation to solar radiation in black wildebeest (Connochaetes gnou). Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology 191: 1065-1077. 240 Maloney S.K., Fuller A., Mitchell G. and Mitchell D. (2002). Brain and arterial blood temperatures of free-ranging oryx. Pfl?gers Archiv - European Journal of Physiology 443: 437-445. Maloney S.K., Moss G., Cartmell T. and Mitchell D. (2005b). Alteration in diel activity patterns as a thermoregulatory strategy in black wildebeest (Connochaetes gnou). Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology 191: 1055-1064. Maloney S.K., Fuller A., Kamerman P.R., Mitchell G. and Mitchell D. (2004). Variation in body temperature in free-ranging western grey kangaroo Macropus fuliginosus. Australian Mammalogy 26: 135-144. Manteca X. and Smith A.J. (1994). Effects of poor forage conditions on the behaviour of grazing ruminants. Tropical Animal Health and Production 26: 129-138. Marshall T.C., Sunnucks P., Spalton J.A., Greth A. and Pemberton J.M. (1999). Use of genetic data for conservation management: the case of the Arabian oryx. Animal Conservation 2: 269-278. Martin K. and Nagy K.A. (2002). Animal physiology and global environmental change volume 2, the earth system: biological and ecological dimensions of global environmental change. In: Encyclopedia of Global Environmental Change. Mooney H.A. and Canadell J.G. (Eds.), John Wiley & Sons, Ltd., Chichester, UK, pp. 136-139. Martin R.A. and Barnosky A.D. (1993). Morphological change in Quaternary mammals of North America. Cambridge University Press, New York, USA. Martinez-Meyer E., Townsend Peterson A. and Hargrove W.W. (2004). Ecological niches as stable distributional constraints on mammal species, with implications for Pleistocene extinctions and climate change projections for biodiversity. Global Ecology and Biogeography 13: 305-314. Mason S.J. and Joubert A.M. (1997). Simulated changes in extreme rainfall over southern Africa. International Journal of Climatology 17: 291-301. Mattison E.H.A. and Norris K. (2005). Bridging the gaps between agricultural policy, land-use and biodiversity. Trends in Ecology and Evolution 20: 610-616. McCarty J.P. (2001). Ecological consequences of recent climate change. Conservation Biology 15: 320-331. McClean C.J., Lovett J.C., K?per W., Hannah L., Sommer J.H., Barthlott W., Termansen M., Smith G.F., Tokumine S. and Taplin J.R.D. (2005). African plant diversity and climate change. Annals of the Missouri Botanical Garden 92: 139-152. McCutchen H.E. (1981). Desert bighorn zoogeography and adaptation in relation to historic land uses. Wildlife Society Bulletin 9: 171-179. 241 McGinnis S.M., Finch V.A. and Harthoorn A.M. (1970). A radio-telemetry technique for monitoring temperatures from unrestrained ungulates. Journal of Wildlife Management 34: 921-925. McGraham N.C., Waimnan F.W., Blaxter K.L. and Armstrong D.G. (1958). Environmental temperature, energy metabolism and heat regulation in sheep. I. Energy metabolism in closely clipped sheep. Journal of Agricultural Science 52: 13-24. McGregor B.A. (1985). Heat-stress in Angora wether goats. Australian Veterinary Journal 62: 349- 350. McGregor B.A. (1998). Nutrition, management and other environmental influences on the quality and production of mohair and cashmere with particular reference to Mediterranean and annual temperate climatic zones: a review. Small Ruminant Research 28: 199-215. McLachlan J.S., Hellmann J.J. and Schwartz M.W. (2007). A framework for debate of assisted migration in an era of climate change. Conservation Biology 21: 297-302. McLaughlin J.F., Hellmann J.J., Boggs C.L. and Ehrlich P.R. (2002). Climate change hastens population extinctions. Proceedings of the National Academy of Sciences of the United States of America 99: 6070-6074. McManus C., Paludo G.R., Louvandini H., Gugel R., Sasaki L.C.B. and Paiva S.R. (2009). Heat tolerance in Brazilian sheep: physiological and blood parameters. Tropical Animal Health and Production 41: 95-101. McNab B.K. (1970). Body weight and the energetics of temperature regulation. Journal of Experimental Biology 53: 329-348. McNab B.K. (1983). Energetics, body size, and the limits to endothermy. Journal of Zoology (London) 199: 1-29. Meadows M.E. and Hoffman T.M. (2003). Land degradation and climate change in South Africa. Geographical Journal 169: 168-177. Merkt J.R. and Taylor C.R. (1994). "Metabolic switch" for desert survival. Proceedings of the National Academy of Sciences of the United States of America 91: 12313-12316. Mesteig K., Tyler N.J.C. and Blix A.S. (2000). Seasonal changes in heart rate and food intake in reindeer (Rangifer tarandus tarandus). Acta Physiologica Scandinavica 170: 145-151. Meyers L.A. and Bull J.J. (2002). Fighting change with change: adaptive variation in an uncertain world. Trends in Ecology and Evolution 17: 551-557. Meynecke J.-O. (2004). Effects of global climate change on geographic distributions of vertebrates in north Queensland. Ecological Modelling 174: 347-357. 242 Midgley G.F., Chown S.L. and Kgope B.S. (2007). Monitoring effects of anthropogenic climate change on ecosystems: a role for systematic ecological observation? South African Journal of Science 103: 282-286. Midgley G.F., Hannah L., Millar D., Rutherford M.C. and Powrie L.W. (2002). Assessing the vulnerability of species richness to anthropogenic climate change in a biodiversity hotspot. Global Ecology and Biogeography 11: 445-451. Millien V., Lyons S.K., Olson L., Smith F.A., Wilson A.B. and Yom-Tov Y. (2006). Ecotypic variation in the context of global climate change: revisiting the rules. Ecology Letters 9: 853-869. Mills A.J., Cowling R.M., Fey M.V., Kerley G.I.H., Donaldson J.S., Lechmere-Oertel R.G., Sigwela A.M., Skowno A.L. and Rundel P. (2005). Effects of goat pastoralism on ecosystem carbon storage in semiarid thicket, Eastern Cape, South Africa. Austral Ecology 30: 797?804. Milne T.A. (2008). The effects of thicket transformation on the diet and body condition of Angora goats. MSc, Nelson Mandela Metropolitan University, Port Elizabeth, South Africa. Milton S.J. and Dean W.R.J. (1995). South Africa's arid and semiarid rangelands: Why are they changing and can they be restored? Environmental Monitoring and Assessment 37: 245-264. Mitchell C.D. and Maher C.R. (2006). Horn growth in male pronghorns Antilocapra americana: selection for precocial maturation in stochastic environments. Acta Theriologica 51: 405-409. Mitchell D., Laburn H.P., Nijland M.J.M. and Zurovsky Y. (1987). Selective brain cooling and survival. South African Journal of Science 83: 598-604. Mitchell D., Maloney S.K., Laburn H.P., Knight M.H., Kuhnen G. and Jessen C. (1997). Activity, blood temperature and brain temperature of free-ranging springbok. Journal of Comparative Physiology B: Biochemical, Systemic, and Environmental Physiology 167: 335-343. Mitchell D., Maloney S.K., Jessen C., Laburn H.P., Kamerman P.R., Mitchell G. and Fuller A. (2002). Adaptive heterothermy and selective brain cooling in arid-zone mammals. Comparative Biochemistry and Physiology, Part B 131: 571-585. Mitchell G. and Lust A. (2008). The carotid rete and artiodactyl success. Biology Letters 4: 415-418. Mitchell N.J., Kearney M.R., Nelson N.J. and Porter W.P. (2008). Predicting the fate of a living fossil: how will global warming affect sex determination and hatching phenology in tuatara? Proceedings of the Royal Society of London. Series B, Biological Sciences 275: 2185-2193. Moolman J. and Cowling R.M. (1994). The impact of elephant and goat grazing on the endemic flora of South African succulent thicket. Biological Conservation 68: 53-61. Moore S.E. and Huntington H.P. (2008). Arctic marine mammals and climate change: impacts and resilience. Ecological Applications 18: s157-165. 243 Mooring M.S., Blumstein D.T. and Stoner C.J. (2004). The evolution of parasite-defence grooming in ungulates. Biological Journal of the Linnean Society 81: 17-37. Morris R.J.H., Howard B. and Macfarlane W.V. (1962). Interaction of nutrition and air temperature with water metabolism of Merino wethers shorn in winter. Australian Journal of Agricultural Research 13: 320-334. Morris W.F., Pfister C.A., Tuljapurkar S., Haridas C.V., Boggs C.L., Boyce M.S., Bruna E.M., Church D.R., Coulson T., Doak D.F., Forsyth S., Gaillard J.M., Horvitz C.C., Kalisz S., Kendall B.E., Knight T.M., Lee C.T. and Menges E.S. (2008). Longevity can buffer plant and animal populations against changing climatic variability. Ecology 89: 19-25. Morrison P.R. and Ryser F.A. (1952). Weight and body temperature in mammals. Science 116: 231- 232. Morrow J.D. and Opp M.R. (2005). Diurnal variation of lipopolysaccharide-induced alterations in sleep and body temperature of interleukin-6-deficient mice. Brain, Behaviour and Immunity 19: 40-51. Mphahlele N.R., Fuller A., Roth J. and Kamerman P.R. (2004). Body temperature, behavior, and plasma cortisol changes induced by chronic infusion of Staphylococcus aureus in goats. American Journal of Physiology ? Regulatory Integrative and Comparative Physiology 287: R863-869. Mucina L. and Rutherford M.C. (2006). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19: 1-807. Mu?oz-Garcia A. and Williams J.B. (2005). Cutaneous water loss and lipids of the stratum corneum in house sparrows Passer domesticus from arid and mesic environments. Journal of Experimental Biology 208: 3689-3700. Mu?oz-Garcia A. and Williams J.B. (2008). Developmental plasticity of cutaneous water loss and lipid composition in stratum corneum of desert and mesic nestling house sparrows. Proceedings of the National Academy of Sciences of the United States of America 105: 15611-15616. Munson L., Terio K.A., Kock R., Mlengeya T., Roelke M.E., Dubovi E., Summers B., Sinclair A.R.E. and Packer C. (2008). Climate extremes promote fatal co-infections during canine distemper epidemics in African lions. PLoS ONE 3: e2545. Murakami D.M., Horwitz B.A. and Fuller C.A. (1995). Circadian rhythms of temperature and activity in obese and lean Zucker rats. American Journal of Physiology ? Regulatory Integrative and Comparative Physiology 269: R1038-R1043. Murden S.B. and Risenhoover K.L. (1993). Effects of habitat enrichment on patterns of diet selection. Ecological Applications 3: 497-505. Murray D.L., Cox E.W., Ballard W.B., Whitlaw H.A., Lenarz M.S., Custer T.W., Barnett T. and Fuller T.K. (2006). Pathogens, nutritional deficiency, and climate influences on a declining moose population. Wildlife Monographs 166: 1-30. 244 Musher D.M., Fainstein V., Young E.J. and Pruett T.L. (1979). Fever patterns. Their lack of clinical significance. Archives of Internal Medicine 139: 1225-1228. Musolin D.L. (2007). Insects in a warmer world: ecological, physiological and life-history responses of true bugs (Heteroptera) to climate change. Global Change Biology 13: 1565-1585. Mustin K., Sutherland W.J. and Gill J.A. (2007). The complexity of predicting climate-induced ecological impacts. Climate Research 35: 165-175. Myers N. (1993). Questions of mass extinction. Biodiversity and Conservation 2: 2-17. Myers P., Lundrigan B.L., Hoffman S.M.G., Poor Haraminac A. and Seto S.H. (2009). Climate- induced changes in the small mammal communities of the Northern Great Lakes Region. Global Change Biology 15: 1434-1454. Mysterud A. (1998). The relative roles of body size and feeding type on activity time of temperate ruminants. Oecologia 113: 442-446. Mzilikazi N. and Lovegrove B.G. (2004). Daily torpor in free-ranging rock elephant shrews, Elephantulus myurus: a year-long study. Physiological and Biochemical Zoology 77: 285-296. Nagy K.A. and Knight M.H. (1994). Energy, water and food use by springbok antelope (Antidorcas marsupialis) in the Kalahari desert. Journal of Mammalogy 75: 860-872. Nangula S. and Oba G. (2004). Effects of artificial water points on the Oshana ecosystem in Namibia. Environmental Conservation 31: 47-54. Nardone A., Ronchi B., Lacetera N. and Bernabucci U. (2006). Climatic effects on productive traits in livestock. Veterinary Research Communications 30: 75-81. Nemani R.R., Keeling C.D., Hashimoto H., Jolly W.M., Piper S.C., Tucker C.J., Myneni R.B. and Running S.W. (2003). Climate-driven increases in global terrestrial net primary production from 1982 to 1999. Science 300: 1560-1563. Ngwa A.T., Pone D.K. and Mafeni J.M. (2000). Feed selection and dietary preferences of forage by small ruminants grazing natural pastures in the Sahelian zone of Cameroon. Animal Feed Science and Technology 88: 253-266. Nijland M.J. and Baker M.A. (1992). Effects of hydration state on exercise thermoregulation in goats. American Journal of Physiology ? Regulatory Integrative and Comparative Physiology 263: R201-205. Nilssen K.J., Sundsfjord J.A. and Blix A.S. (1984). Regulation of metabolic rate in Svalbard and Norwegian reindeer. American Journal of Physiology ? Regulatory Integrative and Comparative Physiology 247: R837-841. 245 Norton B.W. and Deery M.J. (1985). The productivity of Angora goats grazing improved and native pastures in south-eastern Queensland. Australian Journal of Experimental Agriculture 25: 35-40. Nussey D.H., Wilson A.J. and Brommer J.E. (2007). The evolutionary ecology of individual phenotypic plasticity in wild populations. Journal of Evolutionary Biology 20: 831?844. O'Reilly B., Vander A.J. and Kluger M.J. (1988). Effects of chronic infusion of lipopolysaccharide on food intake and body temperature of the rat. Physiology & Behavior 42: 287-291. Ogutu J.O. and Owen-Smith N. (2003). ENSO, rainfall and temperature influences on extreme population declines among African savanna ungulates. Ecology Letters 6: 412-419. Olwoch J.M., Rautenbach C.J.d.W., Erasmus B.F.N., Engelbrecht F.A. and van Jaarsveld A.S. (2003). Simulating tick distributions over sub-Saharan Africa: the use of observed and simulated climate surfaces. Journal of Biogeography 30: 1221-1232. Osovitz C.J. and Hofmann G.E. (2007). Marine macrophysiology: studying physiological variation across large spatial scales in marine systems. Comparative Biochemistry and Physiology A: Molecular and Integrative Physiology 147: 821-827. Ostrowski S. and Williams J.B. (2006). Heterothermy of free-living Arabian sand gazelles (Gazella subgutturosa marica) in a desert environment. Journal of Experimental Biology 209: 1421-1429. Ostrowski S., Williams J.B. and Ismael K. (2003). Heterothermy and the water economy of free-living Arabian oryx (Oryx leucoryx). Journal of Experimental Biology 206: 1471-1478. Ostrowski S., Mesochina P. and Williams J. (2006a). Physiological adjustments of sand gazelles (Gazella subgutturosa) to a boom or bust economy: standard fasting metabolic rate, total evaporative water loss, and changes in the sizes of organs during food and water restriction. Physiological and Biochemical Zoology 79: 810-819. Ostrowski S., Bedin E., Lenain D.M. and Abuzinada A.H. (1998). Ten years of Arabian oryx conservation breeding in Saudi Arabia?achievements and regional perspectives. Oryx 32: 209?222. Ostrowski S., Williams J.B., Bedin E. and Ismail K. (2002). Water influx and food consumption of free-living oryxes (Oryx leucoryx) in the Arabian desert in summer. Journal of Mammalogy 83: 665- 673. Ostrowski S., Williams J., M?sochina P. and Sauerwein H. (2006b). Physiological acclimation of a desert antelope, Arabian oryx (Oryx leucoryx), to long-term food and water restriction. Journal of Comparative Physiology B: Biochemical, Systemic, and Environmental Physiology 176: 191-201. Osuji P.O. (1974). The physiology of eating and the energy expenditure of the ruminant at pasture. Journal of Range Management 27: 437-443. 246 Owen-Ashley N.T. and Wingfield J.C. (2006). Seasonal modulation of sickness behavior in free-living northwestern song sparrows (Melospiza melodia morphna). Journal of Experimental Biology 209: 3062-3070. Owen-Ashley N.T., Turner M., Hahn T.P. and Wingfield J.C. (2006). Hormonal, behavioral, and thermoregulatory responses to bacterial lipopolysaccharide in captive and free-living white-crowned sparrows (Zonotrichia leucophrys gambelii). Hormones and Behavior 49: 15-29. Owen-Smith N. (1988). Megaherbivores: the Influence of Very Large Body Size on Ecology. Cambridge University Press, Cambridge, UK. Owen-Smith N. (1992). Grazers and browsers: ecological and social contrasts among African ruminants. In: Ongul?s/Ungulates 91. Spitz F., Janeau G., Gonzalez G. and Aulagnier S. (Eds.), SFEPM-IRGM, Toulouse, France, pp. 175-181. Owen-Smith N. (1998). How high ambient temperature affects the daily activity and foraging time of a subtropical ungulate, the greater kudu (Tragelaphus strepsiceros). Journal of Zoology 246: 183-192. Owen J.E., Norman G.A., Fisher I.L. and Frost R.A. (1977). Studies on the meat production characteristics of Botswana goats and sheep. 2. General body composition, carcase measurements and joint composition. Meat Science 1: 283-306. Palmer A., Kakembo V., Lloyd W. and Ainslie A. (2004). Degradation patterns and trends in the Succulent Thicket. In: Proceedings of the Thicket Forum. Wilson S, (Ed.), Zuurberg Inn, Addo Elephant National Park, South Africa, pp. 1-10. Parer J.T. (1963). The effects of wool length and nutrition on heat reactions of Merino sheep in the field. Australian Journal of Experimental Agriculture 3: 243-248. Parker K.L. and Robbins C.T. (1985). Thermoregulation in ungulates. In: Bioenergetics of Wild Herbivores. Hudson R.J. and White R.G. (Eds.), CRC Press Inc., Boca Raton, Florida, USA, pp. 161- 182. Parmesan C. (1996). Climate and species' range. Nature 382: 765-766. Parmesan C. and Yohe G. (2003). A globally coherent fingerprint of climate change impacts across natural systems. Nature 421: 37-42. Parmesan C., Root T.L. and Willig M.R. (2000). Impacts of extreme weather and climate on terrestrial biota. Bulletin of the American Meteorological Society 81: 443-450. Parmesan C., Ryrholm N., Stefanescu C., Hillk J.K., Thomas C.D., Descimon H., Huntley B., Kaila L., Kullberg J., Tammaru T., Tennent W.J., Thomas J.A. and Warren M. (1999). Poleward shifts in geographical ranges of butterfly species associated with regional warming. Nature 399: 579-583. 247 Parry M.L. and Carter T.R. (1989). An assessment of the effects of climatic change on agriculture. Climatic Change 15: 95-116. Pearson R.G. and Dawson T.P. (2003). Predicting the impacts of climate change on the distribution of species: are bioclimate envelope models useful? Global Ecology and Biogeography 12: 361-371. P?rez-Barber?a F.J. and Gordon I.J. (1999). The relative roles of phylogeny, body size and feeding style on the activity time of temperate ruminants: a reanalysis. Oecologia 120: 193-197. Peters R.H. (1983). The Ecological Implications of Body Size. Cambridge University Press, New York, USA. Peterson A.T. (2003). Predicting the geography of species' invasions via ecological niche modelling. Quarterly Review of Biology 78: 419-433. Phillips P.K. and Heath J.E. (1995). Dependency of surface temperature regulation on body size in terrestrial mammals. Journal of Thermal Biology 20: 281-289. Piccinini R., Binda E., Belotti M., Casirani G. and Zecconi A. (2004). The evaluation of non-specific immune status of heifers in field conditions during the periparturient period Veterinary Research 35: 539-550. Piccione G., Caola G. and Refinetti R. (2002). Circadian modulation of starvation-induced hypothermia in sheep and goats. Chronobiology International 19: 531 - 541. Piccione G., Caola G. and Refinetti R. (2007). Annual rhythmicity and maturation of physiological parameters in goats. Research in Veterinary Science 83: 239-243. Pigliucci M. (2005). Evolution of phenotypic plasticity: where are we going now? Trends in Ecology and Evolution 20: 481-486. Porter W.P., Budaraju S., Stewart W.E. and Ramankutty N. (2000). Calculating climate effects on birds and mammals: impacts on biodiversity, conservation, population parameters, and global community structure. American Zoologist 40. Porter W.P., Sabo J.L., Tracy C.R., Reichman O.J. and Ramankutty N. (2002). Physiology on a landscape scale: plant-animal interactions. Integrative and Comparative Biology 42: 431-453. P?rtner H.O. (2001). Climate change and temperature-dependent biogeography: oxygen limitation of thermal tolerance in animals. Naturwissenschaften 88: 137-146. P?rtner H.O. (2002). Climate variations and the physiological basis of temperature dependent biogeography: systemic to molecular hierarchy of thermal tolerance in animals. Comparative Biochemistry and Physiology A: Molecular and Integrative Physiology 132: 739-761. 248 P?rtner H.O. (2004). Climate variability and the energetic pathways of evolution: the origin of endothermy in mammals and birds. Physiological and Biochemical Zoology 77: 959-981. P?rtner H.O. and Farrell A.P. (2008). Physiology and climate change: studies of physiological mechanisms are needed to predict climate effects on ecosystems at species and community levels. Science 322: 690-692. P?rtner H.O., Langenbuch M. and Reipschlager A. (2004). Biological impact of elevated ocean CO2 concentrations: lessons from animal physiology and earth history. Journal of Oceanography 60: 705- 718. Post E. and Stenseth N.C. (1998). Large-scale climatic fluctuation and population dynamics of moose and white-tailed deer. Journal of Animal Ecology 67: 537-543. Post E. and Stenseth N.C. (1999). Climatic variability, plant phenology and northern ungulates. Ecology 80: 1322-1339. Post E., Stenseth N.C., Langvatn R. and Fromentin J.M. (1997). Global climate change and phenotypic variation among red deer cohorts. Proceedings of the Royal Society of London. Series B, Biological Sciences 264: 1317-1324. Postma E. (2006). Implications of the difference between true and predicted breeding values for the study of natural selection and micro-evolution. Journal of Evolutionary Biology 19: 309-320. Potts R. (1996). Evolution and climate variability. Science 273: 922-923. Potts R. and Deino A. (1995). Mid-Pleistocene change in large mammal faunas of east Africa. Quaternary Research 43: 106-113. Pounds J.A., Fogden M.P.L. and Campbell J.H. (1999). Biological response to climate change on a tropical mountain. Nature 398: 611-615. Prentice I.C., Cramer W., Harrison S.P., Leemans R., Monserud R.A. and Solomon A.M. (1992). A global biome model based on plant physiology and dominance, soil properties and climate. Journal of Biogeography 19: 117-134. Puchala R., Tovar-Luna I., Goetsch A.L., Sahlu T., Carstens G.E. and Freetly H.C. (2007). The relationship between heart rate and energy expenditure in Alpine, Angora, Boer and Spanish goat wethers consuming different quality diets at level of intake near maintenance or fasting. Small Ruminant Research 70: 183-193. Rasmussen K., Palacios D.M., Calambokidis J., Sabor?o M.T., Dalla Rosa L., Secchi E.R., Steiger G.H., Allen J.M. and Stone G.S. (2007). Southern Hemisphere humpback whales wintering off Central America: insights from water temperature into the longest mammalian migration. Biology Letters 3: 302-305. 249 R?ale D., Berteaux D., McAdam A.G. and Boutin S. (2003a). Lifetime selection on heritable life- history traits in a natural population of red squirrels. Evolution 57: 2416-2423. R?ale D., McAdam A.G., Boutin S. and Berteaux D. (2003b). Genetic and plastic responses of a northern mammal to climate change. Proceedings of the Royal Society of London. Series B, Biological Sciences 270: 591-596. Recarte J.M., Vincent J.P. and Hewison A.J.M. (1998). Flight responses of park fallow deer to the human observer. Behavioural Processes 44: 65-72. Refinetti R. (1999a). Amplitude of the daily rhythm of body temperature in eleven mammalian species. Journal of Thermal Biology 24: 477-481. Refinetti R. (1999b). Relationship between the daily rhythms of locomotor activity and body temperature in eight mammalian species. American Journal of Physiology ? Regulatory Integrative and Comparative Physiology 277: R1493-1500. Refinetti R. and Menaker M. (1992). The circadian rhythm of body temperature. Physiology & Behavior 51: 613-637. Reilly J. (1995). Climate change and global agriculture: recent findings and issues. American Journal of Agricultural Economics 77: 727-733. Reznick D.N. and Ghalambor C.K. (2001). The population ecology of contemporary adaptations: what empirical studies reveal about the conditions that promote adaptive evolution. Genetica 112-113: 183- 198. Rice K.J. and Emery N.C. (2003). Managing microevolution: restoration in the face of global change. Frontiers in Ecology and the Environment 1: 469?478. Richardson F.D., Hahn B.D. and Hoffman M.T. (2005). On the dynamics of grazing systems in the semi-arid succulent Karoo: the relevance of equilibrium and non-equilibrium concepts to the sustainability of semi-arid pastoral systems. Ecological Modelling 187: 491-512. Roberts S.C. and Dunbar R.I.M. (1991). Climatic influences on the behavioural ecology of Chanler's mountain reedbuck in Kenya. African Journal of Ecology 29: 316-329. Robertshaw D. (2006). Mechanisms for the control of respiratory evaporative heat loss in panting animals. Journal of Applied Physiology 101: 664-668. Robertson M.P. and Palmer A.R. (2002). Predicting the extent of succulent thicket under current and future climate scenarios. African Journal of Range and Forage Science 19: 21-28. Roche C. (2005). ?The springbok ?drink the rain?s blood?: indigenous knowledge and its use in environmental history ? the case of the /Xam and an understanding of springbok treks. South African Historical Journal 53: 1-22. 250 Rodbard S. (1950). Weight and body temperature. Science 111: 465-466. Rodr?guez-Trelles F. and Rodr?guez M.A. (1998). Rapid micro-evolution and loss of chromosomal diversity in Drosophila in response to climate warming. Evolutionary Ecology 12: 829-838. Rodriguez Iglesias R.M. and Kothmann M.M. (1998). Best linear unbiased prediction of herbivore preferences. Journal of Range Management 51: 19-28. Rogers D.J. and Randolph S.E. (2000). The global spread of malaria in a future, warmer world. Science 289: 1763-1766. Root T.L., Price J.T., Hall K.R., Schneider S.H., Rosenzweig C. and Pounds J.A. (2003). Fingerprints of global warming on wild animals and plants. Nature 421: 57-60. Rosenheim J.A. and Tabashnik B.E. (1991). Influence of generation time on the rate of response to selection. American Naturalist 137: 527-541. Rosenzweig C., Karoly D., Vicarelli M., Neofotis P., Wu Q., Casassa G., Menzel A., Root T.L., Estrella N., Seguin B., Tryjanowski P., Liu C., Rawlins S. and Imeson A. (2008). Attributing physical and biological impacts to anthropogenic climate change. Nature 453: 353-357. Rosenzweig M.L. (1968). The strategy of body size in mammalian carnivores. American Midland Naturalist 80: 299-315. Rouget M., Richardson D.M., Cowling R.M., Lloyd J.W. and Lombard A.T. (2003). Current patterns of habitat transformation and future threats to biodiversity in terrestrial ecosystems of the Cape Floristic Region, South Africa. Biological Conservation 112: 63-85. Sakurada S., Shido O., Sugimoto N., Hiratsuka Y., Yoda T. and Kanosue K. (2000). Autonomic and behavioural thermoregulation in starved rats. Journal of Physiology (London) 526: 417-424. Sala O.E., Chapin F.S., III, Armesto J.J., Berlow E., Bloomfield J., Dirzo R., Huber-Sanwald E., Huenneke L.F., Jackson R.B., Kinzig A., Leemans R., Lodge D.M., Mooney H.A., Oesterheld M., Poff N.L., Sykes M.T., Walker B.H., Walker M. and Wall D.H. (2000). Global biodiversity scenarios for the year 2100. Science 287: 1770-1774. Salem A.Z.M., Salem M.Z.M., El-Adawy M.M. and Robinson P.H. (2006). Nutritive evaluations of some browse tree foliages during the dry season: secondary compounds, feed intake and in vivo digestibility in sheep and goats. Animal Feed Science and Technology 127: 251-267. Sanon H.O., Kabore-Zoungrana C. and Ledin I. (2007). Behaviour of goats, sheep and cattle and their selection of browse species on natural pasture in a Sahelian area. Small Ruminant Research 67: 64-74. Sargeant G.A., Eberhardt L.E. and Peek J.M. (1994). Thermoregulation by mule deer (Odocoileus hemionus) in arid rangelands of southcentral Washington. Journal of Mammalogy 75: 536-544. 251 Schlesinger W.H. (2006). Global change ecology. Trends in Ecology and Evolution 21: 348-351. Schmidt-Nielsen K. (1984). Scaling: why is animal size so important? Cambridge University Press, New York, USA. Schmidt-Nielsen K., Schmidt-Nielsen B., Jarnum S.A. and Houpt T.R. (1957). Body temperature of the camel and its relation to water economy. American Journal of Physiology 188: 103-112. Schmidt-Nielsen K., Crawford E.C., Newsome A.E., Dawson K.S. and Hammel H.T. (1967). Metabolic rate of camels: effect of body temperature and dehydration. American Journal of Physiology 212: 341-346. Schneider S.H. and Root T.L. (1996). Ecological implications of climate change will include surprises. Biodiversity and Conservation 5: 1109-1119. Scholes R.J., Midgley G.F. and Wand S.J.E. (1999). The vulnerability and adaptation of rangelands. In: South African Climate Change Country Studies Contract Report. Division of Water, Environment and Forest Technology, CSIR Pretoria, South Africa, pp. 1-19. Schroter R.C., Robertshaw D. and Zine Filali R. (1989). Brain cooling and respiratory heat exchange in camels during rest and exercise. Respiration Physiology 78: 95-105. Schroter R.C., Robertshaw D., Baker M.A., Shoemaker V.H., Holmes R. and Schmidt-Nielsen K. (1987). Respiration in heat stressed camels. Respiration Physiology 70: 97-112. Seddon P.J. and Ismael K. (2002). Influence of ambient temperature on diurnal activity of Arabian oryx: implications for reintroduction site selection. Oryx 36: 50-55. Seif S.M., Johnson H.D. and Hahn L. (1973). Environmental heat and partial water restriction effects on body fluid spaces, water loss, body temperature, and metabolism of Holstein cows. Journal of Dairy Science 56: 581-586. Sheldon B.C., Kruuk L.E.B. and Merila J. (2003). Natural selection and inheritance of breeding time and clutch size in the collared flycatcher. Evolution 57: 406-420. Shkolnik A. and Choshniak I. (1987). Water depletion and rapid rehydration in the hot and dry terrestrial environment. In: Comparative Physiology: Life in Water and on Land. Dejous D., Bolls L., Taylor C.R. and Weibel E.R. (Eds.), Liviana Press, Padova, Italy, pp. 141-155. Shrader A.M., Kerley G.I.H., Kotler B.P. and Brown J.S. (2007). Social information, social feeding, and competition in group-living goats (Capra hircus). Behavioural Ecology 18: 103-107. Sigwela A.M., Lechmere-Oertel R.G., Kerley G.I.H. and Cowling R.M. (2003). Quantifying the costs of unsustainable domestic herbivory for biodiversity and ecosystem functioning in succulent thicket, Eastern Cape, South Africa. In: Proceedings of the VIIth International Rangelands Congress. Allsopp 252 N., Palmer A.R., Milton S.J., Kirkman K.P., Kerley G.I.H., Hurt C.R. and Brown C.J. (Eds.), Document Transformation Technologies, Durban, South Africa, pp. 1521-1526. Silanikove N. (1994). The struggle to maintain hydration and osmoregulation in animals experiencing severe dehydration and rapid rehydration: the story of ruminants. Experimantal Physiology 79: 281- 300. Silanikove N. (2000a). Interrelationship between water, food and digestible energy intake in desert and temperate goats. Appetite 12: 163-170. Silanikove N. (2000b). The physiological basis of adaptation in goats to harsh environments. Small Ruminant Research 35: 181-193. Sivakumar M.V.K. (2007). Interactions between climate and desertification. Agricultural and Forest Meteorology 142: 143-155. Skead C.J. (2007). Historical incidence of the larger land mammals in the broader Eastern Cape, 2nd edition. Boshoff A.F., Kerley G.I.H. and Lloyd P.H. (Eds.), Centre for African Conservation Ecology, Nelson Mandela Metropolitan University, Port Elizabeth, South Africa. Skinner J.D. and Louw G.N. (1996). The springbok Antidorcas marsupialis (Zimmermann, 1780). Transvaal Museum, Pretoria, South Africa. Skinner J.D. and Chimimba C.T. (2005). The Mammals of the Southern African Subregion. Cambridge University Press, Cape Town, South Africa. Smit B. and Skinner M.W. (2002). Adaptation options in agriculture to climate change: a typology. Mitigation and Adaptation Strategies for Global Change 7: 85-114. Smith B.N. and Epsten S. (1971). Two categories of 13C/12C ratios for higher plants. Plant Physiology 47: 380-384. Smith F.A. and Betancourt J.L. (2006). Predicting woodrat (Neotoma) responses to anthropogenic warming from studies of the palaeomidden record. Journal of Biogeography 33: 2061-2076. Smith F.A., Betancourt J.L. and Brown J.H. (1995). Evolution of body size in the woodrat over the past 25,000 years of climate change. Science 270: 2012-2014. Smith F.A., Browning H. and Shepherd U.L. (1998). The influence of climate change on the body mass of woodrats Neotoma in an arid region of New Mexico, USA. Ecography 21: 140-148. Smith N. and Wilson S.L. (2002). Changing land use trends in the thicket biome: pastoralism to game farming. Unpublished Report of the Subtropical Thicket Ecosystem Planning Project, Report No. 38. University of Port Elizabeth, Port Elizabeth, South Africa. 253 Spalton J.A. (1999). The food supply of Arabian oryx (Oryx leucoryx) in the desert of Oman. Journal of Zoology 248: 433-441. Stanley Price M.R. (1989). Animal Reintroductions: the Arabian Oryx in Oman. Cambridge University Press, Cambridge, UK. Stenseth N.C. and Mysterud A. (2002). Climate, changing phenology, and other life history traits: nonlinearity and match-mismatch to the environment. Proceedings of the National Academy of Sciences of the United States of America 99: 13379-13381. Stenseth N.C., Mysterud A., Ottersen G., Hurrell J.W., Chan K.-S. and Lima M. (2002). Ecological effects of climate fluctuations. Science 297: 1292-1296. Stephens D.W. and Krebs J.R. (1986). Foraging Theory. Princeton University Press, Princeton, USA. Stillman J.H. (2003). Acclimation capacity underlies susceptibility to climate change. Science 301: 65. Stirling I. and Parkinson C.l. (2006). Possible effects of climate warming on selected populations of polar bears (Ursus maritimus) in the Canadian Arctic. Arctic 59: 261-275. Stitt J.T. (1973). Prostaglandin E1 fever induced in rabbits. Journal of Physiology (London) 232: 163- 179. Stoner C.J., Caro T.M. and Graham C.M. (2003). Ecological and behavioural correlates of coloration in artiodactyls: systematic analyses of conventional hypotheses. Behavioral Ecology 14: 823-840. Stuart-Hill G.C. (1992). Effects of elephants and goats on the Kaffrarian Succulent Thicket of the Eastern Cape, South Africa. Journal of Applied Ecology 29: 699-710. Stuart-Hill G.C. and Aucamp A.J. (1993). Carrying capacity of the succulent valley bushveld of the Eastern Cape. African Journal of Range and Forage Science 10: 1-10 Sykes A.R. and Slee J. (1969). Cold exposure of Southdown and Welsh Mountain sheep: 2. Effects of breed, plane of nutrition and previous acclimatization to cold upon skin temperature, heart rate, shivering and respiratory rate. Animal Production 11: 77-89. Sykes M.T. and Prentice I.C. (1996). Climate change, tree species distributions and forest dynamics: a case study in the mixed conifer/northern hardwoods zone of northern Europe. Climatic Change 34: 161-177. Symonds M.E., Bryant M.J. and Lomax M.A. (1986). The effect of shearing on the energy metabolism of the pregnant ewe. British Journal of Nutrition 56: 635-643. Szelenyi Z., Hummel Z., Szekely M. and Petervari E. (2004). CCK-8 and PGE1: central effects on circadian body temperature and activity rhythms in rats. Physiology & Behavior 81: 615-621. 254 Tattersall I. (1987). Cathemeral activity in primates: a definition. Folia Primatologica: International Journal of Primatology 49: 200-202. Tattersall I. (2006). The concept of cathemerality: history and definition. Folia Primatologica: International Journal of Primatology 77: 7-14. Tattersall I. (2008). Avoiding commitment: cathemerality among primates. Biological Rhythm Research 39: 213-228. Taulman J.F. and Robbins L.W. (1996). Recent range expansion and distributional limits of the nine- banded armadillo (Dasypus novemcinctus) in the United States. Journal of Biogeography 23: 635-648. Taylor C.R. (1969a). The eland and the oryx. Scientific American 220: 88-95. Taylor C.R. (1969b). Metabolism, respiratory changes, and water balance of an antelope, the eland. American Journal of Physiology 217: 317-320. Taylor C.R. (1970a). Strategies of temperature regulation: effect on evaporation in east African ungulates. American Journal of Physiology 219: 1131-1135. Taylor C.R. (1970b). Dehydration and heat: effects on temperature regulation of east African ungulates. American Journal of Physiology 219: 1136-1139. Taylor C.R. and Lyman C.P. (1967). A comparative study of the environmental physiology of an east African antelope, the eland and the Hereford steer. Physiological Zoology 40: 280-295. Taylor C.R. and Lyman C.P. (1972). Heat storage in running antelopes: independence of brain and body temperatures. American Journal of Physiology 222: 114-117. Taylor W.A., Skinner J.D. and Krecek R.C. (2006). The activity budgets and activity patterns of sympatric grey rhebok and mountain reedbuck in a highveld grassland area of South Africa. African Journal of Ecology 44: 431-437. Tear T.H., Mosley J.C. and Ables E.D. (1997). Landscape-scale foraging decisions by reintroduced Arabian oryx. Journal of Wildlife Management 61: 1142-1154. Tegowska E. and Narebski J. (1985). Absence of selective brain cooling in pyrogen-induced fever in rabbits. Acta Physiologica Polonica 36: 256-266. Teplitsky C., Mills J.A., Alho J.S., Yarrall J.W. and Meril? J. (2008). Bergmann's rule and climate change revisited: disentangling environmental and genetic responses in a wild bird population. Proceedings of the National Academy of Sciences of the United States of America 105: 13492-13496. Tewksbury J.J., Huey R.B. and Deutsch C.A. (2008). Putting the heat on tropical animals. Science 320: 1296-1297. 255 Thomas C.D. and Lennon J.J. (1999). Birds extend their ranges northwards. Nature 399: 213. Thomas C.D., Bodsworth E.J., Wilson R.J., Simmons A.D., Davies Z.G., Musche M. and Conradt L. (2001a). Ecological and evolutionary processes at expanding range margins. Nature 411: 577-581. Thomas C.D., Cameron A., Green R.E., Bakkenes M., Beaumont L.J., Collingham Y.C., Erasmus B.F.N., de Siqueira M.F., Grainger A., Hannah L., Hughes L., Huntley B., van Jaarsveld A.S., Midgley G.F., Miles L., Ortega-Huerta M.A., Peterson A.T., L. P.O. and Williams S.E. (2004). Extinction risk from climate change. Nature 427: 145-148. Thomas D.W., Blondel J., Perret P., Lambrechts M.M. and Speakman J.R. (2001b). Energetic and fitness costs of mismatching resource supply and demand in seasonally breeding birds. Science 291: 2598-2600. Thompson I.D., Flannigan M.D., Wotton B.M. and Suffling R. (1998). The effects of climate change on landscape diversity: an example in Ontario forests. Environmental Monitoring and Assessment 49: 213-233. Thompson J.N. (1998). Rapid evolution as an ecological process. Trends in Ecology and Evolution 13: 329-332. Thorne E.T. (1975). Normal body temperature of pronghorn antelope and mule deer. Journal of Mammalogy 56: 697-698. Thouless C.R., Grainger J.G., Shobrak M. and Habibi K. (1991). Conservation status of gazelles in Saudi Arabia. Biological Conservation 58: 85-98. Thrash I. (1998). Impact of water provision on herbaceous vegetation in Kruger National Park, South Africa. Journal of Arid Environments 38: 437-450. Thuiller W. (2004). Patterns and uncertainties of species' range shifts under climate change. Global Change Biology 10: 2020-2027. Thuiller W. (2007). Biodiversity: climate change and the ecologist. Nature 448: 550-552. Thuiller W., Broennimann O., Hughes G., Alkemade J.R.M., Midgley G.F. and Corsi F. (2006). Vulnerability of African mammals to anthropogenic climate change under conservative land transformation assumptions. Global Change Biology 12: 424-440. Todd S.W. (2006). Gradients in vegetation cover, structure and species richness of Nama-Karoo shrublands in relation to distance from livestock watering points. Journal of Applied Ecology 43: 293- 304. Topp C.F.E. and Doyle C.J. (1996). Simulating the impact of global warming on milk and forage production in Scotland: 1. The effects on dry-matter yield of grass and grass-white clover swards. Agricultural Systems 52: 213-242. 256 Tovar-Luna I., Goetsch A.L., Puchala R., Sahlu T., Carstens G.E., Freetly H.C. and Johnson Z.B. (2007). Effects of diet quality on energy expenditure by 20-month-old Alpine, Angora, Boer, and Spanish wethers. Small Ruminant Research 72: 18-24. Tull J.C., Krausman P.R. and Steidl R.J. (2001). Bed-site selection by desert mule deer in southern Arizona. Southwestern Naturalist 46: 354-357. Van D.T.T., Mui N.T. and Ledin I. (2007). Effect of group size on feed intake, aggressive behaviour and growth rate in goat kids and lambs. Small Ruminant Research 72: 187-196. van Heezik Y., Ismail K. and Seddon P.J. (2003). Shifting spatial distribution of Arabian oryx in relation to sporadic water provision and artificial shade. Oryx 37: 295?303. van Schaik C.P. and Griffiths M. (1996). Activity periods of Indonesian rain forest mammals. Biotropica 28: 105-112. Via S., Gomulkiewicz R., De Jong G., Scheiner S.M., Schlichting C.D. and Van Tienderen P.H. (1995). Adaptive phenotypic plasticity: consensus and controversy. Trends in Ecology and Evolution 10: 212- 217. Visser M.E. (2008). Keeping up with a warming world; assessing the rate of adaptation to climate change. Proceedings of the Royal Society of London. Series B, Biological Sciences 275: 649-659. Visser M.E. and Both C. (2005). Review. Shifts in phenology due to global climate change: the need for a yardstick. Proceedings of the Royal Society of London. Series B, Biological Sciences 272: 2561- 2569. Visser M.E., Both C. and Lambrechts M.M. (2004). Global climate change leads to mistimed avian reproduction. Advances in Ecological Research 35: 89-110. Visser M.E., van Noordwijk A.J., Tinbergen J.M. and Lessells C.M. (1998). Warmer springs lead to mistimed reproduction in great tits (Parus major). Proceedings of the Royal Society of London. Series B, Biological Sciences 265: 1867-1870. Visser M.E., Adriaensen F., van Balen J.H., Blondel J., Dhondt A.A., van Dongen S., du Feu C., Ivankina E., V., Kerimov A., B., de Laet J., Matthysen E., McCleery R., Orell M. and Thomson D.L. (2003). Variable responses to large-scale climate change in European Parus populations. Proceedings of the Royal Society of London. Series B, Biological Sciences 270: 367-372. Vlok J., Euston-Brown D. and Cowling R. (2003). Acocks' Valley Bushveld 50 years on: new perspectives on the delimitation, characterisation and origin of Subtropical Thicket vegetation. South African Journal of Botany 69: 27-51. Voigt W., Perner J., Davis A.J., Eggers T., Schumacher J., B?hrmann R., Fabian B., Heinrich W., K?hler G., Lichter D., Marstaller R. and Sander F.W. (2003). Trophic levels are differentially sensitive to climate. Ecology 84: 2444-2453. 257 Walsberg G.E. (1982). Coat color, solar heat gain, and conspicuousness in the phainopepla. Auk 99: 495-502. Walsberg G.E. (1983). Coat color and solar heat gain in animals. Bioscience 33: 88-91. Walsberg G.E. (1990). Convergence of solar heat gain in two squirrel species with contrasting coat colors. Physiological Zoology 63: 1025-1042. Walsberg G.E. and Wolf B.O. (1995). Effects of solar radiation and wind speed on metabolic heat production by two mammals with contrasting coat colours. Journal of Experimental Biology 198: 1499-1507. Walsberg G.E., Campbell G.S. and King J.R. (1978). Animal coat color and radiative heat gain: a re- evaluation. Journal of Comparative Physiology B: Biochemical, Systemic, and Environmental Physiology 126: 211-222. Walther G.R., Beiner S. and Conradin B. (2005). Trends in the upward shift of alpine plants. Journal of Vegetation Science 16: 541-548. Walther G.R., Post E., Convey P., Menzel A., Parmesan C., Beebee T.J., Fromentin J.M., Hoegh- Guldberg O. and Bairlein F. (2002). Ecological responses to recent climate change. Nature 416: 389- 395. Weitzman M.L. (1998). The Noah's Ark problem. Econometrica 66: 1279-1298. Wentzel D., Viljoen K.S. and Botha J.J. (1979). Physiological and endocrinological reactions to cold stress in the Angora goat. Agroanimalia 11: 19-22. White C.R. and Seymour R.S. (2003). Mammalian basal metabolic rate is proportional to body mass2/3. Proceedings of the National Academy of Sciences of the United States of America 100: 4046-4049. Whittow G.C., Scammell C.A., Leong M. and Rand D. (1977). Temperature regulation in the smallest ungulate, the lesser mouse deer (Tragulus javanicus). Comparative Biochemistry and Physiology A: Physiology 56: 23-26. Wigginton J.D. and Dobson F.S. (1999). Environmental influences on geographic variation in body size of western bobcats. Canadian Journal of Zoology 77: 802-813. Wikelski M. and Cooke S.J. (2006). Conservation physiology. Trends in Ecology and Evolution 21: 38-46. Williams J.B., Ostrowski S., Bedin E. and Ismail K. (2001). Seasonal variation in energy expenditure, water flux and food consumption of Arabian oryx Oryx leucoryx. Journal of Experimental Biology 204: 2301-2311. 258 Wilson M.E. (1994). Disease in evolution introduction. Annals of the New York Academy of Sciences 740: 1-12. Wilson R.T. (1989). Ecophysiology of the Camelidae and Desert Ruminants. Springer-Verlag, New York, USA. Wingfield J.C. (2003). Control of behavioural strategies for capricious environments. Animal Behaviour 66: 807-816. Wolf M., Friggens M. and Salazar-Bravo J. (2009). Does weather shape rodents? Climate related changes in morphology of two heteromyid species. Naturwissenschaften 96: 93-101. Wronski T., Apio A. and Plath M. (2006). Activity patterns of bushbuck (Tragelaphus scriptus) in Queen Elizabeth National Park. Behavioural Processes 73: 333-341. Yirmiya R., Avitsur R., Donchin O. and Cohen E. (1995). Interleukin-1 inhibits sexual behavior in female but not in male rats. Brain, Behavior, and Immunity 9: 220-233. Yom-Tov Y. (2001). Global warming and body mass decline in Israeli passerine birds. Proceedings of the Royal Society of London. Series B, Biological Sciences 268: 947-952. Yom-Tov Y. and Yom-Tov S. (2006). Decrease in body size of Danish goshawks during the twentieth century. Journal of Ornithology 147: 644-647. Yom-Tov Y., Yom-Tov S., Wright J., Thorne C.J.R. and Du Feu R. (2006). Recent changes in body weight and wing length among some British passerine birds. Oikos 112: 91-101. Yoshida T., Jones L.E., Ellner S.P., Fussmann G.F. and Hairston N.G., Jr. (2003). Rapid evolution drives ecological dynamics in a predator-prey system. Nature 424: 303-306. Young B.A. and Corbett J.L. (1972). Maintenance energy requirement of grazing sheep in relation to herbage availability. I. Calorimetric estimates. Australian Journal of Agricultural Research 23: 57-76. Zander K.K., Drucker A.G., Holm-M?ller K. and Simianer H. (2009). Choosing the "cargo" for Noah's Ark - applying Weitzman's approach to Borana cattle in East Africa. Ecological Economics 68: 2051- 2057. Zervanos S.M. and Hadley N.F. (1973). Adaptational biology and energy relationships of the collared peccary (Tayassu tajacu). Ecology 54: 759-774. 259 9 Appendix 1 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 10 Appendix 2 275 276 277 278 279 280 281 282 283 284 285