75 

Palaeont. afr., 24 (1981) 

FOSSIL HYAENIDAE FROM THE MAKAPANSGAT LIMEWORKS DEPOSIT, 
SOUTH AFRICA 

by 

R.M. Randall 

Department ojZoology, University oj Port Elizabeth, P. O. Box 1600, Port Elizabeth 6000 

ABSTRACT 
The remains of three hyaena species have been recovered from the Makapansgat Lime­

works deposit. A common small form , Hyaena hyaena makapani, and a rare large form, Pachycro­
cuta brevirostris, were recovered from Member 3 (Lower Phase I grey breccia). The rare Crocuta 
crocuta was recovered from Member 4 (Upper Phase I breccia), and was the only hyaena from 
this horizon. 

Abundant cranial and dental material of H. h. makapani facilitated comparisons with extant 
and fossil forms to confirm its identification as a subspecies of the extant striped hyaena. De­
spite morphological differences in the skull and teeth, H. abronia from Langebaanweg is con­
firmed as its likely ancestor. Some deciduous teeth of H. h. makapani are described and the 
eruption sequence of permanent cheek teeth deduced. P. brevirostris appears to be the largest 
fossil hyaena from Africa, showing affinities to P. bellax from Kromdraai. C. crocuta is similar to 
the extant form and the fossil forms from East Africa. As in the East African deposits, C. cro­
cuta appears relatively late in the succession. 

The hyaena material has limited value in site faunal correlations for dating purposes, but 
does not contradict the palaeomagnetic age estimate of more than 2,9 My for Member 3 (grey 
breccia) (Partridge 1979). 

CONTENTS 
Page 

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 75 
MATERIALS AND METHODS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 
RESULTS AND DISCUSSION ............................................................... 76 

Hyaena hyaena makapani ................................................................ . . . .. 76 
Skull. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 76 
Perman en t den ti tion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 78 
Deciduous dentition .. ................................................ • . . . . . . . . . . . . . . .. 79 

Pachycrocuta brevirostris . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 81 
Crocuta crocuta . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 82 
Correlations and dating .................................................................... 83 

CONCLUSIONS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 
ACKNOWLEDGEMENTS .................................................................. 84 
REFERENCES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 84 

INTRODUCTION 
The family Hyaenidae is well represented ·in 

most of the Transvaal cave deposits (Toerien 1952, 
Ewer 1954, 1955a, Hendey 1974a, Collings et al. 
1975). The Makapansgat Limeworks deposit is no 
exception; from there Toerien (1952) described the 
remains of a small hyaena similar to the extant 
striped hyaena, Hyaena hyaena, which he called H. 
makapani, as well as a large hyaena which he called 
Crocuta cf. brevirostris. Since then additional 
material of both species has been recovered. This 
material was briefly reported in Collings et al. 
(1975), together with an account of a third species 
attributed to C. crocuta. 

small hyaena and the extant striped hyaena is gen­
erally acknowledged, and in a review of the African 
fossil hyaenids Ewer (1967) concluded that they 
were only subspecifically different. Ewer (1955b) 
suggested that the Pliocene hyaena H. namaquensis 
might be a possible ancestor of H. h. makapani, but 
this was later considered unlikely (Ewer 1967). 
The late Miocene/early Pliocene deposit at Lange­
baanweg has yielded much hyaenid material in­
cluding the species H. abronia, which has been pro­
posed as the likely ancestor of the Makapansgat 
form and the extant striped hyaena (Hendey 1978). 

This paper presents a detailed account of the 
hyaena material reported in Collings et al. (1975), 
and it covers all the cranial material on hand at the 
Bernard Price Institute for Palaeontological Re­
search in 1972. The relationships of the hyaenas 
and their significance is discussed primarily in re­
lation to finds elsewhere in Africa. 

The taxonomic and phyletic relationships of the 
Plio/Pleistocene Hyaenidae have been the subject 
of many papers (Ewer 1955b, 1967, Kurten 1956, 
Ficcarelli and Torre 1970, Hendey 1974b, 1978). 
The close relationship between the Makapansgat 

BP - F 



76 

MA TERIAL AND METHODS 
The fossil material was mechanically prepared, 

mostly by staff preparators at the Bernard Price 
Institute, using hammer and punch. A few speci­
mens were also prepared by the author using a 
"Vibro-tool" ar~d emery disc. 

With the exception of the carnassials, where ad­
ditional measurements were taken, two measure­
ments were taken on all teeth. These were the 
mesiodistal and buccolingual distances, corre­
sponding to the length (L) and breadth (B) respec­
tively. The measurements were taken at the level of 
the cingulum, and the mean of three measurements 
was recorded each time. Use was made of ratio 
diagrams in comparisons between species, and in 
particular those based upon percentages. 

In references to the stratigraphy of the Lime­
works the sequence and terminology of Partridge 
(1979) and Brain (1958) is followed. 

RESULTS AND DISCUSSION 
Hyaena hyaena makapani 

Except for specimen M244 from Member 2 
(basal red mud), all H. h. makapani specimens were 
recovered from Member 3 (Lower Phase 1 grey 
breccia) O.W. Kitching,_pers. comm.). 

Skull (table 1; figs. 1,2) 
The skull is morphologically similar to the two 

extant Hyaena species, but is smaller than either H. 
hyaena or the fossil H. abronia (table 2). 

TABLE 1 
Skull measurements of Hyaena hyaena makapani from the 
Makapansgat Limeworks deposit 

Mean Std. Dev. Range 
Dimension Sample (mm) (mm) (mm) 

Condylobasal 
length 5 200,4 6,5 194,0--210,0 

Zygomatic width I 138,0 
Palate width 4 63,3 3,4 60,3-68,1 
Interorbital width 6 45,3 2,9 40 ,4-47 ,8 

TABLE 2 
Comparison of mean skull measurements: Hyaena hyaena 
makapani, Hyaena abronia and extant Hyaena hyaena 

Condylobasal Zygomatic 1 n terorbi tal 
length width width 

Species (mm) (mm) (mm) 

H . h. makapani a 200,4 138,0 45 ,3 
H. abronia 

b 
217,0 c. 145,0 47,7 

H . hyaena 
b 

211 ,0 c. 146,0 44,0 

a = composite reconstruction (figs. I , 2) based on M 2530, M 
299, M 8348 
b = Hendey (l974b) 

A comparison of H. h. makapani with the illustra-
tion and description of H. abronia in Hendey 
(1974b) revealed a few cranial differences . In H. h. 
makapani the palate (fig. 2a) is relatively shorter 
and narrower, and there appears to be a slightly 
longer precanine diastema. In lateral view (fig. 1 a) 

the angular process of the mandible in H. h. maka­
pani is not bulbous as it is in H. abronia, and it has a 
slightly dorsally directed distal end. In all these 
features H. h. makapani is similar to the striped 
hyaena. 

Permanent dentition (table 3) 
The dental formula 3/3: 1/1 :4/3: 1/1 = 34 is 

identical to that in H. hyaena and H. brunnea (Ewer 
1973). 

With the exception of the robust caniniform P, 
the incisors are small teeth and are seldom recov­
ered. Of the 28 canines recovered one upper and 
two lower were still in their alveoli. Upper and 
lower canines could be distinguished because up­
per canines are straighter and more elliptical in 
cross-section. Premolars, both loose and in their 
alveoli, were the most common hyaenid teeth re­
covered. Upper and lower premolars could be dis­
tinguished because lower premolars have their 
cusps arranged in a straight line parallel to the jaw 
ramus, whereas upper premolars have the first 
cusp offset lingually. 

The cusp terminology used for the carnassials is 
taken from Gromova (1968), and is illustrated in 
Figure 3. The talonid of MJ has a mean value of 

TABLE 3 
Measurements of teeth of Hyaena hyaena makapani from 
the Makapansgat Limeworks deposit 

Mean Std. dey. Range 
Tooth Dimension Sample (mm) (mm) (mm) 

I I length 2 3, I 2,8-3,4 
breadth 2 4,4 

I' length 3 4,6 4,1-5 ,6 
breadth 3 6,6 6,2-7 ,2 

I' length 6 7,5 0,4 7,1-8,0 
breadth 6 8,9 0,6 8,2-9,7 

C length 12 13,0 0,5 12,3-14,0 
breadth 12 9,1 0,5 8,3-9,9 

p i length 5 6,4 0,4 6,0--7 ,0 
breadth 5 5,9 0,5 5,4-6,6 

P' length 21 14,6 I , I 12,6-16,1 
breadth 20 9,4 0,5 8,6-10,2 

P' length 29 19,6 I, I 16,5-21 , I 
breadth 31 12,6 0,7 11 ,0--13 ,8 

P' length 28 28,8 I, I 25,8-31 ,2 
breadth 26 17, I I, I 15,6-19,5 
length 
metastyle 26 10,5 0,5 9,3-11 ,5 

M I length 16 5,6 0,5 4,7-6,6 
breadth 16 12,8 0,7 12,0--14, I 

I, length 4 5,6 0,2 5,4-5,8 
breadth 4 6,2 0,1 6,1-6,3 

C length 15 12,9 0,6 11 ,6-13 ,9 
breadth 15 9,8 0,5 8,9-10,5 

P, length 13 13,6 0,7 11 ,9-14,8 
breadth 14 8, 1 0,6 7,2-9,1 

P, length 20 17,9 0,7 16,7-19, 1 
breadth 21 10,8 0,5 9,7-;-11 ,5 

P, length 23 19,2 0,7 17,4-20,4 
breadth 21 10,9 0,5 9,7-12,1 

M I length 17 20,8 0,9 19, 1-22,0 
breadth 19 10,1 0,6 9,3-11 ,0 
length talonid 15 4,3 0,3 3,7-4,7 



77 

a 

Figure 1. Skull of Hyaena hyaena makapani showing (a) lateral and (b) dorsal aspects (scale bar = 5 em). (Composite reconstruc­
tion based mainly on M 2530, M 8348 and M 299.) 



78 

20,6 per cent (n = 15) of the total length of the 
tooth. 

In keeping with the small skull size, the teeth of 
H. h. makapani are also smaller than the extant H. 
hyaena (fig. 4). The similarity between the two is 

b 

seen to be close in terms of various tooth ratios 
(table 4); this is especially true for the carnassials. 
The biggest differences in the cheek teeth are for the 
anterior upper premolars. In H. hyaena there has 
been a reduction of pi and an increase in size of p2 

Figure 2. Reconstructed skull and mandible of Hyaena hyaena makapani showing (a) ventral aspect of skull and (b) lateral aspect of 
mandible (scale bar = 5 em) . (Skull reconstruction as in Figure I; mandible based on M 262.) 



79 

LINGUAL BUCCAL OCCLUSAL 

pr pas amph mes 

pad prd med tal end hyd 

Figure 3. Lingual , buccal and occlusal aspects of the carnassia ls of Hyaena hyaena makapani. T erminology: pr= protocone, pas = 
parastyle, amph = amphicone, mes = metastyle , pad = pa raconid , prd = pro toconid , med = m etaconid , tal = tal­
onid , end = en toconid , hyd = hypoconid (after Gromova 1968). 

relative to H. h. makapani. The pattern for H. abro­
nia and H . namaquensis is similar to the other two 
except for the upper molar, and to a lesser extent 
the lower molar (fig. 4). This is not unexpected in 
view of the fact that there has been less reduction 
in the molars of H . abronia and H. namaquensis, 
where both M 2 and M 2 may be present (Hendey 
1978). Another difference between the cheekteeth 
of the older H. abronia and H. namaquensis and the 
other two is the presence of PI and the higher 
length:breadth ratio of most of the cheekteeth 
(table 4). 

These comparisons illustrate the close re­
lationship between H. h. makapani and the extant 
H. hyaena, and supports the contention of Ewer 
(1967) that they are not separate species, but 
rather that the Makapansgat form should be re­
garded as a subspecies. The relationship between 
H. h. makapani and the older species H. abronia and 
H. namaquensis is more obscure in terms of skull 
morphology and tooth measurements. Hendey 
(1978) also noted differences in postcranial el­
ements, where H. abronia had relatively longer hind 
limbs than the extant H. hyaena. Hendey (1978) 
concluded that H. abronia was the likely ancestor of 
H. hyaena, with H. h. makapani the intermediate 
form, and H. namaquensis the possible ancestor of H. 

brunnea. The comparisons in Figure 4 and Table 4 
indicate that H. abronia and H. namaquensis are more 
closely related to one another than to either of the 
others. The relationship appears even stronger 
when the entire dentition is considered, since they 
both possess PI ' M 2 and M 2 and have more similar 
postcranial elements than either of the extant 
Hyaena species or H. h. makapani. If H. abronia was 
the ancestor of H. h. makapani then there must have 
been a relatively rapid evolutionary developmental 
phase between the late Miocene/early Pliocene H. 
abronia at Langebaanweg, dated 4-5 My (Hendey 
1978), and the first appearance of H. hyaena, dated 
at 3 My in East Africa (Coppens and Howell 
1976). A form from East Rudolf exhibiting the den­
tal characters of H. hyaena and many of the cranial 
features of H. abronia may well be an intermediate 
form (Leakey 1976). 

Deciduous dentition (table 5, fig. 5) 
The deciduous dentition of hyaenas bears little 

resemblance to the permanent dentition, and with­
out good comparative material it is unlikely that 
deciduous teeth would be correctly identified. For­
tunately, two specimens from the Limeworks con­
tained deciduous teeth with permanent teeth 
erupting. By working the surrounding bone away 



80 

TABLE 4 
Comparison of cheektooth ratios in Hyaena hyaena makapani, two species of fossil hyaenas 
from southern Africa, and two extant hyaena species 

H. h. H. namaquensis 
makapani H. hyaena H . abronia (Namaqua- H. brunnea 

Teeth Dimensions (Makapansgat) ( extant) (Langebaan) land) ( extant) 

P ' LIB 1,09 1,04' 1,08' 1,18d 1,04' 
P' LIB 1,56 1,64' 1,82' 1,78d 1,49' 
p3 LIB 1,54 1,61 ' 1,64' 1,57d 1,43' 
p4 LIB 1,69 1,62b 1,73' 1,70d 1,70b 
M ' LIB 0,44 0,50' 0,62' 0,56d 0,44 
P, LIB 1,70 1,55' 1,77' 1,75d 1,41 ' 
P3 LIB 1,65 1,74' 1,82' 1,77d 1,47' 
P4 LIB 1,77 1,82' 1,90' 1,88d 1,7l' 
M , LIB 2,06 2,00b 2,02' 2,20d 2,05" 
M , Trigonid/ L 0,79 0,80' 0,75d 0,84' 
p4 Metastyle/ L 0,37 0,36' 0,38' 

p3, p4 L P3/ L p4 0,68 0,68b 0,71 ' O,71 d 0,65 b 

M ',P' WM'/LP' 0,44 0,46' 0,54' 0,54d 0,37' 
P3, P4 L P3/ L P4 0,92 0,93 b 0,92' 0,92d 0,87b 

P3, p. B P3/B p. 1,01 0,99' 0,96' 0,97d 1,04' 
P4,M, L P4/ L M , 0,92 0,98b 0,90' 0,89d 1,0 I b 

Key: 
a = Toerien (1952), b = Ficcarelli and Torre (1970) , c = Hendey (1974b), 
d = Hendey (1978), e = Ewer (l955a) , f = Kurten (1956). 
Unless otherwise stated = own measurements. 

a 

p1 

p2 

p3 

p4 

M1 

M1 

P4 

P3 

P2 

60 80 100 120 140 160 40 60 80 100 120 

Percentage scale Percentage scale 
Figure 4. Ratio diagram comparing (a) mean crown lengths and (b) mean crown widths in Hyaena hyaena makapani (0), Hyaena 

hyaena extant (.6), Hyaena abronia (A) and Hyaena namaquensis (T ). Standard of comparison (100 per cent) Hyaena 
brunnea extant (e). Data from Toerien (1952), Kurten (1956) , Hendey (1974b, 1978) and own measurements . 

with an emery disc and "Vibro-tool" the perma­
nent teeth were exposed and identification as H. h. 
makapani confirmed. 

Specimen M 603 is a portion of a right maxilla 
with DM\ DM4 and MI in situ, with p4 starting to 
erupt, and p3 about to do so. DM4 resembles MI in 
having a similar crown pattern and being located 

in a comparable position. Like MI, it also has a 
greater breadth (buccolingual axis) than length 
(mesiodistal axis). DM3 is a slender shearing tooth 
functioning as the upper carnassial, and it is com­
parable to P4. It has three cusps and a blade 
arranged roughly in a straight line, meaning that it 
has one more cusp than P\ its functional replace-



TABLE 5 
Measurements of Hyaena hyaena makapani deciduous 
teeth from the Makapansgat Limeworks deposit 

Specimen Length Breadth 
number Tooth (mm) (mm) 

M2340 DM' 21 ,3 
M603 DM' 20,2 12,0 

DM' 7,6 12,3 
M2284 DM, c. 14,0 c. 5,7 

DM, 15,0 6,3 
M2342 DM, 14,7 5,8 

Figure 5. Hyaena hyaena makapani mandibles from Maka­
pansgat showing specimen M 2284 with decidu­
ous teeth , and permanent teeth erupting, and adult 
paratype (M 252). The mandible of M 2284 was 
worked away to expose the permanent teeth. 

ment tooth. The extra cusp on DM3 is the paracone 
which is separate from the metacone, unlike in p4 
where these two cusps have fused to form the am­
phicone. The DM\ therefore, has a parastyle, 
paracone, metacone and metastyle in a line pro­
ceeding from the mesial to the distal end. It also 
has a large protocone situated lingually relative to 
the paracone and metacone. The amphicone of p4 
was erupting immediately distal to DM\ and the 
tip of the protocone of p4 had erupted anterior to 
the lingual extension of DM4. When the permanent 
premolars were exposed by working away the max­
illa, the parastyle of p4 could be seen erupting im­
mediately lingual to the metastyle of DM3, and p 3 
beneath the paracone and parastyle of DM3. It ap­
pears that p4 would have erupted before P3, and the 
order of eruption of the three hindmost permanent 
cheek teeth was MI_P4_P3. 

Specimen M 2284 consists of the major portion 
of a right mandible with DM2 and DM3 in situ, and 
the permanent teeth C, P2, P3 and P4 about to re­
place. MI had been lost prior to fossilization and 
DM4 had either been shed, or like MI had been 
lost. Unlike in the adult, the mandible is short, 
shallow and comparatively fragile. Both DM2 and 
DM3 are unlike the permanent teeth, being high 
and narrow. The mandible was worked away on 
both lingual and buccal sides to expose the perma-

81 

nent teeth. P2 is erupting beneath DM2 with its 
main cusp between the two roots of DM2. P3 is re­
placing DM3, and P4 is replacing DM4 in a similar 
manner. The order of eruption of the permanent 
cheekteeth appears to be M I-P2-P4-P3. 

The state of development of the roots of the 
erupting permanent teeth indicates that many of 
the isolated teeth found are from animals of a simi­
lar age. 

The deciduous teeth have little taxonomic value 
at this stage primarily because of the paucity of 
comparative material, and because Kurten (1956) 
did not consider the lower deciduous teeth. The 
mean length of 14,85 mm (n = 2) for DM3 is 
identical to that recorded for the striped hyaena 
(Kurten 1956), which is surprising considering the 
larger size of the striped hyaena. The damaged 
DM2 has an approximate length of 14 mm, but this 
seems to be an overestimate in view of the mean 
length of 11,76 mm for H. hyaena (Kurten 1956) . 

Ewer (1967) expressed doubts about Toerien's 
(1952) identification of the maxillary fragment (M 
603) , but by exposing the erupting permanent 
teeth it has been possible to confirm the identifica­
tion. Toerien (1952) drew attention to the large 
size of DM3 compared to specimens of H. brunnea, 
and also to morphological differences between 
them. It has not been possible to verify these obser­
vations but, in relation to the lower dentition, the 
deciduous teeth of H . h. makapani appear larger 
than might be expected. 

The eruption sequence for the upper teeth de­
duced from the maxilla (M 603) fits the pattern 
shown by examples from the families Felidae, Mus­
telidae and Viverridae (Ewer 1973). In these famil­
ies the eruption sequence of the lower teeth is 
M I-P2-P3-P4, and not M I-P2-P4-P3 as it appears to 
be in specimen M 2284. 

Pachycrocuta brevirostris 
All Pachycrocuta brevirostris material obtained 

came from Member 3 (Lower Phase 1 grey brec­
cia) O.W. Kitching, pers. comm.). 

Since Toerien's (1952) description of the re­
mains of a large hyaena from the Limeworks three 
additional specimens attributed to the same species 
have been found. The isolated left lower canine (M 
606) described by Toerien fits into the canine al­
veolus of specimen M 2565, and may be from the 
same individual. The specimens were mostly frag­
mentary and the teeth extensively worn, so that ac­
curate measurements could not always be obtained 
(table 6, fig. 6). 

The most noteworthy feature of this form relates 
to its size, and only P. brevirostris has teeth of a 
comparable size (Howell and Petter 1980). It is 
principally on this basis that the Makapansgat 
specimens are assigned to the genus Paclrycrocuta. 
The features considered most characteristic of the 
genus relate to details of M I, P4 and p4 (Ficcarelli 
and Torre 1970, Howell and Petter 1980). Unfortu­
nately, in the Makapansgat specimens neither M I 
nor P4 is present, and in neither of the p4 specimens 



82 

TABLE 6 
Measurements of Pachycrocuta brevirostris teeth from the 
Makapansgat Limeworks deposit 

Specimen Length 
number Tooth (mm) 

M 2565 P, 18,4 

M604 
M6010 
M2533 

P, 23,0 
P, 24,9 
P' 42 ,0 
P' c. 25,2 
p4 c.44,0 
M' c. 6,0 

a 

b 

Breadth 
(mm) 

13,0 
17, I 
19,4 

c. 16, I 
c.20,5 
c. 11 ,5 

Figure 6. Pachycrocuta brevirostris mandible fragments from 
Makapansgat. (a) Specimen M 604 and (b) speci­
men M 2565. 

available can details of the crown pattern be dis­
cerned. Nevertheless the Makapansgat form has 
the robust premolars found in Pachycrocuta. In par­
ticular, the breadth:length ratios of 0,71 and 

0,74-0,78 for Pz and P3 respectively are closer to P. 
brevirostris than to other species of Pacfrycrocuta or 
Hyaena (Howell and Petter 1980). 

The only hyaena of comparable size from South 
African Plio/Pleistocene deposits is Hyaena bellax 
from Kromdraai (Ewer 1954). Ewer (1967) con­
cluded that the closest relative of the Kromdraai 
form was Hyaena brevirostris from Europe, a view­
point supported by Ficcarelli and Torre (1970) 
when they grouped them as two species under the 
genus Pachycrocuta. Howell and Petter (1980) went 
further and synonymized the Makapansgat and 
Kromdraai forms as P. bellax. There are, however, 
differences, notably the larger teeth in the 
Makapansgat form, particularly P4. In addition the 
breadth:length ratios in the Kromdraai form are 
smaller - 0,67 and 0,72 for Pz and P3 respectively. 
The MI is less developed in the Makapansgat form, 
being considerably smaller in relation to P4. These 
differences seem sufficient to warrant separation at 
the specific level and consequently the Makapans­
gat form has been assigned to P. brevirostris. 

Large hyaenas of the Pachycrocuta type have 
therefore twice appeared in southern Africa. As far 
as can be established the only other site in Africa 
where these large hyaenas have been recovered is 
Ain Brimba in Tunisia (Howell and Petter 1980). 

Crocuta crocuta 
One specimen (M 2567) attributed to C. crocuta 

was obtained from Member 4 (Upper Phase 1 
breccia) U.W. Kitching, pers. comm.) (fig. 7). No 
other hyaena material has been recovered from this 
horizon. 

Figure 7. Crocuta crocuta mandible M 2567 from Makapans­
gat. 

The specimen is the major part of a left man­
dible with Pz, P3 , P4 and MI. Part of the talonid of 
MI was damaged, but otherwise the teeth are well 
preserved and could be measured (table 7). The 
trigonid comprises about 87 per cent of the total 
length of MI. The damage to the talonid is such 
that it is impossible to establish if a metaconid was 
present. Absence of a metaconid would have con­
firmed identification as Crocuta, although its pres­
ence would not have excluded the possibility that 
the specimen belongs to Crocuta (Kurten 1956). 



B3 

TABLE 7 
Measurements of Crocuta crocuta teeth (specimen M 
2567) from the Makapansgat Limeworks deposit M 1 

Tooth 
Length 
(mm) 

14,2 
19,3 
21 ,5 

c.26,3 

Breadth 
(mm) 

9, I 
13,7 
13, I 
11 ,4 

Trigonid 
(mm) 

23 ,0 

The teeth are slightly smaller than those of the 
extant C. crocuta, but their similarity is immediately 
apparent, particularly when compared to the genus 
Hyaena (fig. 8). Ficcarelli and Torre (1970) estab-
1ished several useful dental characters for separat­
ing the genera Crocuta and Hyaena, including two 
based upon the lower dentition - P4 length:M, 
length, and M, breadth:M, length. Both of these 
ratios for specimen M 2567 fall within the Crocuta 
range (table 8) and outside the Hyaena range (table 
4) . 

A comparison of tooth dimensions between the 
extant C. crocuta and the specimens from Maka­
pansgat, East Rudolf, Olduvai, and Swartkrans 
plus Kromdraai combined shows the similarity be­
tween the forms (table 8). The most marked 
change is seen in M " where there has been a trend 
towards a relative increase in the length up to the 
extant form. There appears to have been little 
change in the lower dentition since the earliest oc­
currence of C. crocuta. Leakey (1976) stated that 
there had been little change in cranial or postcra­
nial morphology of C. crocuta for 2 My. 

Figure 8. 

80 90 100 110 120 

Percentage scale 

Ratio diagram comparing mean crown lengths in 
Crocuta crocuta from Makapansgat (0), C. crocuta 
from Swartkrans plus Kromdraai (f"l) , Hyaena 
hyaena makapani (0) and Hyaena brunnea ( . ) . 
Standard of comparison ( 100 per cent) Crocuta cro­
cuta extant (-). Data from Ewer (1954, 1955a), 
Kurten (1956), Leakey (1976) and own measure­
ments. 

TABLE 8 
Comparison of cheektooth ratios in Crocuta crocuta from several Plio/Pleistocene deposits 
in Africa, and the extant form 

Swartkrans 
East and 

Teeth Dimension Makapansgat Rudolf a Olduvai b Kromdraai c Extant d 

M , LIB 2,31 2,26 2,27 2,32 2,39 
P, LIB 1,64 1,66 1,61 1,72 1,77 
P3 LIB 1,41 1,44 1,44 1,50 1,44 
P, LIB 1,56 1,41 1,43 1,43 

P" M , L P,/L M , 0,82 0,77 0,76 0,80 0,81 
P3, P, L P3/ L P, 0,90 0,94 0,95 0,94 0,94 

Key : 
a = Leakey (1976), b = Petter (1973), c = Ewer (1954, 1955a), d = Ficcarelli and Torre (1970). 

Correlations and dating 
Both H . hyaena and C. crocuta have been recov­

ered at several sites of Plio/ Pleistocene age in East 
Africa and South Africa (table 9). These include 
some sites in East Africa for which K-Ar dates 
have been obtained. The earliest datable deposits 
at which a small hyaena similar to the Makapans­
gat form have been recovered are the lower mem­
ber of the Koobi Fora Formation, East Rudolf 
(Leakey 1976), and the Usno Formation and 
Member B of the Shungura Formation, Omo 
Group (Howell and Petter 1976). The earliest 
dated deposits containing C. crocuta are the lower 
member of the Koobi Fora Formation, East Rudolf 

(Leakey 1976) and Member G of the Shungura 
Formation, Omo Group (Howell and Petter 1976). 
Dates for these formations are: lower member 
Koobi Fora 2,61-3,18 My (Fitch and Miller 1976); 
Usno Formation> 2,97 My, Member B Shungura 
Formation 2,95-2,97 My, and Member G Shung­
ura Formation 1,83-1,93 My (Coppens and How­
ell 1976). It must be emphasized that H. hyaena, 
and to a lesser extent C. crocuta, occurred in later 
horizons from both East Rudolf and the Omo 
Group (Leakey 1976, Howell and Petter 1976). 
Leakey (1976) drew attention to the fact that the 
Hyaenidae, especially H. hyaena, have little value in 
site comparisons and correlations because the den-



84 

TABLE 9 
Sites in South Africa and East Africa where hyaena 
species represented at Makapansgat Limeworks have 
been recovered 

Sites 

South Africa 
Makapansgat 
Swartkrans' 
Kromdraaib 

East Africa 
Olduvai' 
Omo Groupd 
East Rudolf< 
Laetolil' 

Key: 

H. hyaena 

x 

x 
x 
x 

x 

C. crocuta 

x 
x 
x 

x 
x 
x 

x 

P. brevirostris 
and P. bellax 

x 

x 

a = Ewer (1955a), b = Ewer (1954), c = Petter (1973), 
d=Coppens and Howell (1976), e=Leakey ( 1976) , 
f = Dietrich (1942). 

tition has changed little in 3 My, and in C. crocuta 
there has been little cranial or postcranial change 
in 2 My. Thus, based upon the hyaena material, 
Member 3 (Lower Phase I grey breccia) could be 
any age up to 3 My and Member 4 (Upper Phase 1 
breccia) up to 2,6 My but, because of their conser­
vatism, hyaenas appear to have limited use in cor­
relations for dating purposes. Consequently there 
is nothing in the hyaena material to contradict the 
palaeomagnetic age estimate of> 2,9 My for the 
Makapansgat Limeworks grey breccia (Member 3) 
(Partridge 1979). 

CONCLUSIONS 
Cranial and dental characters confirm the 

identification of the abundant remains of the small 
hyaena in Member 3 (Lower Phase 1 breccia) as a 
subspecies of the extant striped hyaena. Compari­
sons support Hendey's (1978) suggestion that its 
likely ancestor is H. abronia from Langebaanweg. 
However, the relationship between H. h. makapani 
and extant H. hyaena seems to be closer than the re­
lationship between H. h. makapani and H. abronia; 
H. abronia seems to be more closely related to H. 
namaquenS1S. 

Fragmentary remains of a large hyaena in Mem­
ber 3 (Lower Phase 1 grey breccia) are identified 

as P. brevirostris. This large hyaena shows affinities 
to P. bellax from Kromdraai, and these specimens 
seem to be the only records of these large hyaenas 
from Africa. 

Hyaenas are represented by one specimen from 
Member 4 (Upper Phase 1 breccia), here identified 
as C. crocuta. This form is similar to the extant form 
and to fossil Plio/Pleistocene forms from East 
Africa. 

C. crocuta seems to have replaced H. h. makapani 
and P . brevirostris in the upper horizons. Kurten 
(1956) documented replacement of P. brevirostris by 
C. crocuta at several deposits in Europe and China, 
which he believed was due to direct competition. A 
similar replacement of H. hyaena by C. crocuta oc­
curred in the Upper Pleistocene of Palestine, with 
H. hyaena only reappearing after the extinction of C. 
crocuta (Kurten 1965). By contrast, in several East 
African deposits both H. hyaena and C. crocuta have 
been recovered from the same horizon (Leakey 
1976, Howell and Petter 1976), and the ranges of 
the extant forms overlap (Kruuk 1976). Conse­
quently the apparent disappearance of H. h. maka­
pani in later horizons at Makapansgat is unlikely to 
be due only to competition with C. crocuta. 

H. h. makapani and C. crocuta are morphologically 
similar to forms from datable deposits in East 
Africa. Correlations indicate that the age of Mem­
ber 3 (Lower Phase 1 grey breccia) could be up to 
3,0 My, and Member 4 (Upper Phase 1 breccia) 
up to 2,6 My. The hyaena material therefore pro­
vides general support for the palaeomagnetic age 
estimate of > 2,9 My for the grey breccia (Part­
ridge 1979). However, because the Hyaenidae, es­
pecially H. hyaena and to a lesser extent C. crocuta, 
are so conservative they have limited value in site 
correlations for dating purposes. 

ACKNOWLEDGEMENTS 
Dr. J .W. Kitching is thanked for his advice and assistance 

with the preparation and identification of material. Drs. 
A.R.I. Cruickshank, C . Gow, Q.B. Hendey, J.W. Kitching, 
J.M. Maguire and Mrs. B.M. Randall commented on drafts of 
the manuscript. 

This work was done at the Bernard Price Institute for Palae­
ontological Research, and financial support was provided in 
the form ofa C.S.I.R. postgraduate bursary. 

REFERENCES 
BRAIN, C.K. (1958). The Transvaal ape-man-bearing cave de­

posits, Tvl. Mus. Mem., 11, 131 pp. 
COLLINGS, G.E., CRUICKSHANK, A.R.I., MAGUIRE, 

J.M. and RANDALL, R .M. (1975). Recent faunal studies 
at Makapansgat Limeworks, Transvaal , South Africa. Ann. 
S. Afr. Mus., 71,153-165. 

COPPENS, Y. and HOWELL, F.C . (1976). Mammalian fau­
nas of the Omo Group: distributional and biostratigraphi­
cal aspects. In: Coppens, Y., Howell, F.C., Isaac, G.LI. and 
Leakey, R .E.F., Earliest man and environments in the Lake 
Rudolf Basin, 177-192. Chicago, University of Chicago 
Press . 

DIETRICH, W.O. (1942). Altestquartare Saugetiere aus der 
Siidlichen Serengeti, Deutsch-Ostafrika. Palaeontographica 
Abt. A, 94, 43-133. 

EWER, R.F. (1954). The fossil carnivores of the Transvaal 

caves. The Hyaenidae of Kromdraai. Proc. zool. Soc. Lond., 
124(3) , 565-585. 

- - - - (1955a). The fossil carnivores of the Transvaal caves. 
The Hyaenidae, other than Lycyaena, of Swartkrans and 
Sterkfontein. Proc. zool. Soc. Lond. , 124(4),815-837 . . 

---- (1955b). The fossil carnivores of the Transvaal caves. 
The Iycyaenas of Sterkfontein and Swartkrans, together 
with some general considerations of the Transvaal fossil 
hyaenids. Proc. zool. Soc. Lond. , 124(4),839-857. 

- - - - (1967). The fossil hyaenids of Africa - a reappraisal. 
In : Bishop, W.W. and Clark, J.D., Eds ., Background to evol­
ution in Africa, 109-123. Chicago, Chicago University Press. 

- - - - (1973). The carnivores. Cornell, Cornell University 
Press. 

FICCARELLI, H. and TORRE, D. (1970). Remarks on the 
taxonomy ofhyaenids. Palaeontogr. ital., 66(n.s.36), 13-33. 



FITCH, FJ. and MILLER, ].A. (1976). Conventional Potas­
sium-Argon and Argon-401 Argon-39 dating of volcanic 
rocks from East Rudolf. In: Coppens, Y., Howell, F .C ., 
Isaac, G .LI. and Leakey, R.E.F., Eds. ; Earliest man and 
environments in the Lake Rudolf Basin, 123-147. Chicago, 
University of Chicago Press. 

GROMOVA, V .I. (1968). Mammals. Fundamentals of Palaeon­
tology, 13,5-37. Israel program for scientific translation. 

HENDEY, Q.B. (1974a) . New fossil carnivores from the 
Swartkrans australopithecine site (Mammalia: Carnivora). 
Ann. Transv . Mus., 29, 27-5l. 

---- (1974b). The late Cenozoic Carnivora of the south­
western Cape Province. Ann. S. Afr. Mus ., 63, 1-369. 

---- (1978). Late T ertiary Hyaenidae from Langebaanweg, 
South Africa, and their relevance to the phylogeny of the 
family. Ann. S. Afr. Mus ., 76 (7), 265-297. 

HOWELL, F .C . and PETTER, G . (1976) . Carnivora from 
Omo Group Formations, southern Ethiopia. In: Coppens, 
Y., Howell , F .C ., Isaac, G.LI. and Leakey, R.E.F ., Eds. , 
Earliest man and environments in the Lake Rudolf Basin, 314-331. 
Chicago, University of Chicago Press . 

- - - - and - - - - (1980) . The Pachycrocuta and Hyaena lin­
eages (Plio-Pleistocene and extant species of the Hyaeni-

85 

dae). Their relationships with Miocene ictitheres: Palhyaena 
and Hyaenictitherium. Geobios 13(4) , 579-623. 

KRUUK, H. (1976). Feeding and social behaviour of the 
Striped Hyaena (Hyaena vulgaris Desmarest). E . Afr. Wildl . 
}., 14, 91-11l. 

KURTEN, B. (1956) . The status and affinities of Hyaena sinen­
sis Owen and Hyaena ultima Matsumoto. Am. Mus . Novit. , 
no. 1764, 1-48. 

- - - - (1965). The Carnivora of the Palestine Caves. Acta 
zool.fenn ., 107, 1-74. 

LEAKEY, M.G. (1976). Carnivora of the East Rudolf suc­
cession. In: Coppens, Y., Howell, F.C ., Isaac, G .LI. and 
Leakey, R .E.F., Eds., Earliest man and environments in the Lake 
Rudolf Basin, 302-313. Chicago, University of Chicago 
Press. 

PARTRIDGE, T.C. (1979). Re-appraisal of lithostratigraphy 
of Makapansgat Limeworks hominid site. Nature, Lond., 
279, 484-488. 

PETTER, G. (1973). Carnivores Pleistocenes du Ravin d ' 
Olduvai (Tanzanie) . In: Leakey, L.S.B. , Savage, RJ. and 
Coryndon, S.C ., Eds ., Fossil vertebrates of Africa, 3 , 43-100. 

TOERIEN, MJ. (1952). The fossil hyenas of the Maka­
pans gat Valley. S. Afr.}. Sci. , 48(9),293-300.