ChapterPDF Available

Abstract and Figures

792 specimens attributed to the Giraffidae were recovered by the Eyasi Plateau Paleontological Expedition (EPPE) from the three Pliocene stratigraphic units at Laetoli, with Giraffa stillei the most common taxon in all three levels. Giraffids are notably well represented in the Upper Laetolil Beds, with further evidence gathered by EPPE for the three previously recognized species from this unit. In the Lower Laetolil Beds Giraffa stillei is provisionally identified, as is Sivatherium. A third, large giraffid species may also be present. Based on a specimen recovered by Kohl-Larsen’s team during the first extensive exploration of Laetoli, we now provisionally recognize Giraffa pygmaea from the Upper Ndolanya Beds, along with Giraffa stillei and Sivatherium maurusium. Evidence for Giraffa jumae in the Upper Ndolanya Beds is not as convincing, as it is based on a small number of postcranial bones. In the time between the formation of the Upper Laetolil Beds and the Upper Ndolanya Beds, it appears that Giraffa stillei increased in size, which has been documented at other contemporary East African localities. This may relate to competition from the smaller Giraffa pygmaea. KeywordsGiraffidae- Giraffa - Sivatherium -Artiodactyla-Pliocene-Africa-Tanzania
Content may be subject to copyright.
Abstract 792 specimens attributed to the Giraffidae were
recovered by the Eyasi Plateau Paleontological Expedition
(EPPE) from the three Pliocene stratigraphic units at Laetoli,
with Giraffa stillei the most common taxon in all three
levels. Giraffids are notably well represented in the Upper
Laetolil Beds, with further evidence gathered by EPPE
for the three previously recognized species from this unit.
In the Lower Laetolil Beds Giraffa stillei is provisionally
identified, as is Sivatherium. A third, large giraffid species
may also be present. Based on a specimen recovered by
Kohl-Larsen’s team during the first extensive exploration of
Laetoli, we now provisionally recognize Giraffa pygmaea
from the Upper Ndolanya Beds, along with Giraffa stillei
and Sivatherium maurusium. Evidence for Giraffa jumae in
the Upper Ndolanya Beds is not as convincing, as it is based
on a small number of postcranial bones. In the time between
the formation of the Upper Laetolil Beds and the Upper
Ndolanya Beds, it appears that Giraffa stillei increased in
size, which has been documented at other contemporary
East African localities. This may relate to competition from
the smaller Giraffa pygmaea.
Keywords Giraffidae Giraffa Sivatherium Artiodactyla
Pliocene Africa Tanzania
The diversity of giraffids in the Pliocene of Africa was con-
siderably greater than it is today, and Laetoli is similar to
most other African Pliocene sites in having multiple giraffid
taxa represented in all of the stratigraphic units. In the Upper
Ndolanya Beds, for example, four giraffid species are now
identified with the recognition of Giraffa pygmaea, Harris
1991 based on a mandibular specimen described below.
These same four taxa have been identified at the Pliocene
sites of Hadar, Koobi Fora and the Omo Valley (Boaz et al.
1982; Harris 1991; Reed 2008). While the diversity of spe-
cies is similar to other sites, previous studies of fossil giraf-
fids have noted the unusually high number of giraffe
specimens at Laetoli compared to other East African Plio-
Pleistocene sites (Harris 1976a, 1987, 1991), which could be
related to favorable ecological conditions and/or a reduced
number of competitors compared to other early Pliocene
sites. The Laetolil Beds are unusual with respect to the num-
ber of giraffids recovered, but this abundance diminishes in
the Upper Ndolanya Beds, possibly due to competition from
the increased number of bovids.
Three species of Giraffa have been at least provisionally
identified from the various stratigraphic units at Laetoli,
Giraffa jumae, Leakey 1965, Giraffa pygmaea and Giraffa
stillei, Dietrich 1942. The earliest appearance of Giraffa in
East Africa is currently in the late Miocene at Aramis and
Lothagam (WoldeGabriel et al. 1994; Harris 2003). At that
time it appears that the genus had already split into at least
two species, probably G. jumae and G. stillei. The G. stillei
specimens from the Lower Laetolil Beds described below
represent some of the earliest fossils in East Africa that can be
provisionally attributed to Giraffa stillei. Based on the current
evidence, G. pygmaea does not appear until the late Pliocene,
where competition with G. stillei seems to have resulted in
increased size in the latter taxon (see below).
Also present at Laetoli, as it is in many East African faunal
assemblages, is Sivatherium maurusium, Pomel 1892. The
S. maurusium specimens from the Upper Laetolil Beds rep-
resent the earliest known representatives of that species,
although there are a number of earlier sivathere specimens
that cannot be attributed to species. A single sivathere pre-
molar was recovered from the Lower Laetoli Beds, which are
similar in age to sites in North, East and South Africa from
which Sivatherium hendeyi, Harris 1976 has been identified
(Harris 1976a, 1999; Likius 2002; Vignaud et al. 2002). The
tooth could represent this taxon or its possible descendant,
S. maurusium.
C.A. Robinson (*)
Department of Biology, Bronx Community College,
2155 University Avenue., Bronx, NY 10453, USA
Chapter 14
Chris A. Robinson
T. Harrison (ed.), Paleontology and Geology of Laetoli: Human Evolution in Context. Volume 2: Fossil Hominins
and the Associated Fauna, Vertebrate Paleobiology and Paleoanthropology, DOI 10.1007/978-90-481-9962-4_14,
© Springer Science+Business Media B.V. 2011
340 C.A. Robinson
Cranial Specimens and Taxonomic
Attribution of Giraffids
Species of Giraffidae are generally differentiated from one
another on the basis of size and cranial morphology, especially
on the shape of their ossicones (Harris 1976a, b, 1987, 1991;
Churcher 1978; Hamilton 1978; Geraads 1986, 1994; Likius
2002; Solounias 2007). No giraffid cranial specimens were
recovered by EPPE. Harris (1987) described two Giraffa stillei
ossicones from the Upper Laetolil Beds, and another (MB Ma
42325 in Berlin) was published by Dietrich (1942) that is
likely to be derived from the Upper Laetolil Beds (T. Harrison,
personal communication), confirming the presence of the
smaller Giraffa in this unit. In addition, following publication
of Harris (1987), right and left frontal ossicones (LAET
76-4193) from this unit were described and assigned to
Sivatherium maurusium (Harris, 1991). This makes it possible
to confidently recognize the presence of Sivatherium mauru-
sium in the Upper Laetolil Beds. It should be noted that, while
most authors identify the cranial appendages of sivatheres as
ossicones (Harris 1987, 1991; Harris 2003; Harris et al. 1988;
Solounias 1988, 2007; Churcher 1990), Geraads (1986, 1991)
argues that this term should be reserved for the cranial append-
ages of Giraffa, Okapia and, possibly, Palaeotragus. Geraads
(1991) describes the ossicone as “a bone originally indepen-
dent from those of the cranial roof, ossifying from a cartilagi-
nous matrix”. He also notes that the “true” ossicones of extant
giraffids are “hyper-ossified” and shifted posteromedially, fur-
ther from the supraorbital region. The ramified cranial append-
ages of Sivatherium, which show evidence of having had blood
vessels on them, appear to grow in a manner more similar to
cervids or bovids than Giraffa and do not exhibit the extremely
dense bone of that genus (Geraads 1986).
Other than in its cranial morphology, especially the form
of its ossicones, Giraffa jumae closely resembles Giraffa
camelopardalis, Linnaeus 1758 (Harris 1987, 1991; Geraads
et al. 2004). There are no G. jumae cranial specimens known
from Laetoli. Consequently, it is not possible to confirm the
specific attribution of giraffid fossils that are similar in size
to extant giraffes. However, the Upper Laetolil Beds, where
a number of specimens similar in size and morphology to G.
jumae from other sites are well documented, are dated to
between approximately 3.5 and 3.8 Ma (Drake and Curtis
1987; Deino 2011), making it likely that these specimens
represent G. jumae rather than G. camelopardalis (Harris,
1987). There are no specimens of G. camelopardalis from
the Pliocene, whereas G. jumae is known from a number of
Pliocene sites in East and South Africa (Harris 1976a–c,
1991; Churcher 1978; Harris et al. 2003; Kullmer et al.
2008). The earliest diagnostic evidence for G. cameloparda-
lis is from the Nariokotome Member at West Turkana, dated
to between 1.2 and 1.3 Ma (Harris et al. 1988; Harris 1991;
McDougall and Brown 2006).
Because Dietrich (1942) did not designate a holotype for
Okapia stillei, later referred to the genus Giraffa (Harris,
1976b), Harris (1987) selected two specimens from Laetoli
to be lectotypes of Giraffa stillei. One of the lectotypes,
based on Dietrich (1942) Figure 170, was identified as a par-
tial mandible with P3-P4 and M2-M3, although the teeth in the
figure are actually P4-M3 (Dietrich 1942; T. Harrison, per-
sonal communication). Moreover, as Dietrich often did, the
lectotype is comprised of a composite of specimens from dif-
ferent individuals and localities (Geraads et al. 2004;
T. Harrison, personal communication). It is proposed here
that the mandibular fragment with M2 and M3 (MB Ma
39078) be retained as a lectotype, as it clearly belongs to the
smaller Giraffa species from the Upper Laetolil Beds, and
conforms to the concept of Okapia stillei as proposed by
Dietrich (1942). The other unassociated specimens should be
recognized as paralectotypes.
Lower Laetolil Beds
Eleven giraffid specimens were recovered from the
Lower Laetolil Beds during the 1998–2005 field seasons
(Table 14.1). All were identified to taxon on a provisional
basis because of the lack of diagnostic specimens. Nine of
these specimens are attributed to Giraffa aff. stillei.
In the following descriptions the terminology used for the
postcranial elements follows that of Harris (1987, 1991) and
colleagues (Harris et al. 2003) other than for the external
cuneiform. The term external cuneiform is used rather than
lateral cuneiform to avoid confusion, since this bone is
formed by the fusion of the middle and lateral cuneiforms in
fossil giraffids from Laetoli.
Giraffa aff. stillei is best represented by a left mandibular
fragment from Kakesio retaining M1-M3 (EP 854/04)
(Fig. 14.1). The preserved morphology and dimensions of its
well worn molars are similar to those of smaller G. stillei
specimens from the Upper Laetolil Beds (see Fig. 14.4).
Small ectostylids are present on M1 and M2, and an ento-
stylid on M3, as they are on a number of G. stillei lower
molars from the Upper Laetolil Beds. Three isolated dental
elements and five postcranial specimens from Emboremony
and Kakesio are also provisionally attributed to this species.
The P4 (EP 2097/03) is similar in size to smaller specimens
from the Upper Laetolil Beds and indistinguishable morpho-
logically. EP 291/99, a relatively small P4 from Emboremony
1, shares an unusual feature with another smaller P4 from the
Upper Laetolil Beds, EP 3528/00, in that on both specimens
the anterior arm of the hypoconid extends lingually into the
talonid basin. The postcranial specimens resemble G. stillei
fossils from the Upper Laetolil Beds in both size and mor-
phology, with only the calcaneum (EP 202/03) and astrag-
alus (EP 375/99) reasonably well preserved. Although
34114 Giraf fidae
adhering matrix adds slightly to this dimension, the calca-
neum is similar in its length to larger G. stillei specimens
from the Upper Laetolil Beds. However, the Lower Lateolil
Beds specimen more closely resembles smaller Upper
Laetolil Bed calcanei in its more gracile tuber calcis. The
astragalus is more robust and slightly larger than the largest
of the G. stillei specimens from the Upper Laetolil Beds,
being similar in size to smaller female G. camelopardalis
specimens. It is possible that this specimen should be
included in a separate taxon with the large magnum from the
Lower Laetolil Beds described below but, pending further
evidence that would help to identify that species, it is retained
in G. aff. stillei.
Sivatherium is represented only by a single poorly pre-
served P4 from Esere (EP 1680/98), which is the only giraffid
fossil recovered from that site. The general morphology,
highly rugose enamel and massive buccolingual breadth
closely resemble those of Sivatherium maurusium P4s recov-
ered from the Upper Laetolil Beds. However, S. maurusium
cannot be distinguished in its dental morphology from
Sivatherium hendeyi and both species have been identified at
contemporary or older African sites (Singer and Boné 1960;
Harris 1976a; Churcher 1978; Likius 2002; Vignaud et al.
2002; Harris et al. 2003). Consequently, this specimen is
provisionally assigned to Sivatherium sp.
A magnum (EP 108/98) does not fit comfortably in
either of the two other giraffid taxa identified in the Lower
Laetolil Beds. It is similar in size to EP 4298/00, a Giraffa
aff. jumae specimen from the Upper Laetolil Beds, but its
scaphoid facet is unusually expanded mediolaterally
compared to other specimens from Laetoli. It also differs
from EP 4298/00, but is similar to magnums from the Mary
Leakey collection (Harris 1987), in that its anterolateral
corner is not projecting as far laterally as on extant
giraffe specimens. This specimen could be a representa-
tive of a third giraffid taxon in the Lower Laetolil Beds,
possibly Giraffa jumae. However, without diagnostic
fossils, it seems most appropriate to attribute the magnum
to Giraffidae indet.
Upper Laetolil Beds
Over 700 giraffid specimens were recovered from the Upper
Laetolil Beds (Table 14.2). Representatives of all three taxa
previously recognized from this unit (Harris 1987) have been
identified among these fossils.
Upper Laetolil Dento-Gnathic Specimens
All dento-gnathic specimens recovered by EPPE from the
Upper Laetolil Beds were attributed, at least provisionally, to
one of the three previously described giraffid species from
these strata (Harris 1987). Many of the isolated dental specimens
cannot be identified to a specific tooth because of morphological
similarities between serial teeth (Harris 1991). Included in
the EPPE collection are the first recorded giraffid incisors
and canines from Laetoli.
Table 14.1 Lower Laetolil Beds specimens. Measurements in mm
Specimen ID Element Locality Taxonomic attribution Measurements
EP 290/99 Distal radius fragment Emboremony 1 Giraffa aff. stillei
EP 291/99 Lower P4 Emboremony 1 BL – 16.2; MD – 21.3
EP 375/99 Astragalus Emboremony 1 Lat. Length – 93.7; Med. Length – 86.1a
EP 2097/03 Upper P4 Emboremony 2 BL – 21.9; MD – 17.5
EP 027/99 Partial lower M3 Kakesio
EP 202/03 Calcaneum Kakesio 8 Length – 160.2; Artic. AP – 72.2
EP 854/04 Mandible with M1-M3 Kakesio 8 M1: BL – 17.6; MD – 23.1: M2: BL – 18.3;
MD – 23.7: M3: BL – 17.9; MD – 34.0
EP 520/03 Astragalus fragment Kakesio 10 Med. Length – 77.2a
EP 860/04 Astragalus fragment Kakesio 10 Lat. Length – 86.3a; Med. Length – 76.7
EP 108/98 Magnum Kakesio South Giraffidae indet. Length – 77.9; Ant. DV – 36.2
EP 1680/98 Upper P4 Esere Sivatherium sp. BL – 49.2
Note: The maximum distance was measured for all tooth dimensions
BL buccolingual breadth, MD mesiodistal length, AP anteroposterior, DV dorsoventral, Artic. AP anteroposterior breadth at the proximal end of
the calcaneum
a Estimated measurement
Fig. 14.1 Occlusal view of EP 854/04, Giraffa aff. stillei left mandibu-
lar fragment with M1-M3 from Kakesio, Lower Laetolil Beds
342 C.A. Robinson
Table 14.2 Upper Laetolil Beds giraffid specimen numbers at each locality
Giraffa aff. jumae Giraffa stillei Sivatherium maurusium
Locality Dental Postcranial Total Dental Postcranial Total Dental Postcranial Total
1 2 2 4 5 8 13 4 4 8
1NW 0 0 0 1 3 4 1 0 1
2 3 6 9 16 16 32 6 2 8
3 0 1 1 11 5 16 2 1 3
4 3 2 5 0 3 3 0 3 3
5 1 5 6 14 7 21 3 4 7
6 2 3 5 17 8 25 1 3 4
7 3 2 5 4 13 17 8 5 13
8 4 4 8 14 7 21 4 5 9
9 7 8 15 14 8 22 2 1 3
9S 6 3 9 9 9 18 3 0 3
10 3 1 4 18 15 33 3 3 6
10E 10 5 15 25 19 44 4 5 9
10W 5 3 8 14 24 38 1 2 3
11 4 3 7 9 14 23 1 1 2
12 0 2 2 2 0 2 0 0 0
12/12Ea0 0 0 2 1 3 1 1 2
12E 3 1 4 6 4 10 1 1 2
13 2 7 9 6 8 14 5 0 5
15 0 5 5 10 1 11 0 0 0
16 0 10 10 5 3 8 1 5 6
19 0 0 0 1 2 3 1 0 1
20 1 0 1 0 2 2 0 0 0
21 5 3 8 6 11 17 5 5 10
22 1 5 6 9 10 19 3 0 3
22E 2 3 5 4 11 15 3 1 4
24 0 0 0 0 1 1 0 1 1
Totals 67 84 151 222 213 435 63 53 116
a When these five giraffid fossils were recovered, specimens from Locs. 12 or 12E were combined
Giraffa aff. jumae
Sixty-six isolated dental specimens attributable to Giraffa
aff. jumae were recovered by EPPE from the Upper Laetolil
Beds. In general, the teeth resemble those of Giraffa
camelopardalis in both size and morphology, and they are
substantially larger than specimens assigned to Giraffa stil-
lei. Given the relatively low sexual dimorphism in the
dentition of extant giraffes, it seems unlikely that specimens
attributed to these two Giraffa species are males and females
of a single species (Harris 1987).
Two incisors and two canines are provisionally assigned
to Giraffa aff. jumae. One of the incisors, EP 4297/00, is
unusually low crowned, due in part to its having been exten-
sively worn, being more similar in crown height to Giraffa
stillei specimens (Fig. 14.2). However, it has a long and
broad crown and a thick root, which distinguishes it from
specimens of the smaller Giraffa species. The two canines,
EP 432/98 and EP 3172/00, have a bilobed crown typical of
giraffids (Churcher 1978; Hamilton 1978; Geraads 1986;
Harris 1991; Solounias 2007). However, the mesial and dis-
tal lobules are evenly rounded and more similar to one
another morphologically than the asymmetrical lobules of
extant giraffes (Fig. 14.3). They are also lower crowned than
the canines of Giraffa camelopardalis, but which are other-
wise similar in size. An isolated canine from Makapansgat,
M 1801, attributed to Giraffa (Singer and Boné 1960), does
not share this morphology with the Laetoli specimens,
although it is approximately the same size. Its shape most
closely resembles that of the Sivatherium maurusium canine
(EP 498/00) from Laetoli (Fig. 14.3).
Five mandibular postcanine teeth are attributed to Giraffa
aff. jumae, all with length and breadth measurements in the
upper end of the extant giraffe size range, but otherwise
indistinguishable (Table 14.3).
Many more maxillary than mandibular postcanine teeth
of giraffids were recovered from the Upper Laetolil Beds.
This makes it possible to compare the degree and pattern of
variation between fossil and extant giraffe species for these
teeth. Data for Giraffa camelopardalis are derived from 13
wild shot specimens housed at the American Museum of
Natural History (AMNH) in New York. The coefficients of
variation for the length and breadth of all of the Giraffa
aff. jumae maxillary postcanine teeth are similar to, and
more often less than, those of Giraffa camelopardalis
(Table 14.3).
14 Giraf fidae
Modeling the extent of variation to be expected in a fossil
giraffid species from one site is difficult given that there is
only a single living species and museums typically have indi-
viduals that have been collected from multiple populations.
Consequently, the extent of variation among extant speci-
mens may be greater due to the geographic variation of the
sample. However, the Laetoli specimens sample a number of
populations spread over a substantial amount of time, albeit
seemingly under similar ecological conditions, possibly
leading to increased variation in the fossil sample as well.
Given the potential for temporal variation in a fossil sample
even from one site, some have suggested that collecting data
from multiple extant populations or subspecies is the most
appropriate means of modeling the extent of variation to be
expected in a fossil species (Richmond and Jungers 1995;
Lockwood et al. 1996; Uchida 1996).
Giraffa aff. jumae upper molars are slightly larger than
those of extant giraffes, but otherwise closely resemble them
in morphology. In a bivariate plot of the buccolingual
breadths and mesiodistal lengths of giraffid upper molars,
most of the G. aff. jumae upper molars are positioned above
those of extant giraffes on the Y-axis, indicating proportion-
ately broader teeth (Fig. 14.4).
Giraffa stillei
Two hundred and thirteen isolated teeth and eight specimens
with multiple associated teeth (EP 504/98 – M2-M3; EP
1594/00 – P3, P4 and M2; EP 1122/00 – RdP2-M2 and LdP2;
EP 1600/00 – mandible with M1-M2; EP 292/01 – P4-M1; EP
483/01 – P3-M2; EP 1700/04 – M2-M3; EP 174/05 – P3-M1;
EP 1294/05 – P2-P4) attributed to G. stillei were recovered
from the Upper Laetolil Beds. Unless discussed below, these
teeth are indistinguishable from those of Giraffa camelop-
ardalis and Giraffa jumae other than by size (Harris 1976b, c,
1987, 1991).
Mandibular dentition: A tiny deciduous canine (EP 1098/00)
that is most likely attributable to Giraffa stillei was recovered
from Loc. 8. This specimen resembles permanent canines,
although its mesial lobule is mesiodistally longer and more spatu-
late in shape. In addition, the specimen is over 20% smaller in its
mesiodistal dimension than the smallest giraffid permanent
canine from Laetoli. This is consistent with the difference in size
between the deciduous and permanent canines of extant giraffes.
Fig. 14.2 Giraffid incisors from the Upper Laetolil Beds. Top row
from left to right EP 2425/03, 438/98, 4296/00, 618/01 and 619/01
(Giraffa stillei); Bottom row from left to right – EP 4297/00 and 2270/03
(Giraffa aff. jumae); EP 1625/98, 2126/03, 1124/00 and 165/99
(Sivatherium maurusium)
Fig. 14.3 Giraffid canines from the Upper Laetolil Beds. From left to
right – EP 553/03 (Giraffa stillei); EP 432/98 and 3172/00 (Giraffa aff.
jumae); EP 498/00 (Sivatherium maurusium)
344 C.A. Robinson
Table 14.3 Measurements (in mm) of giraffid dental elements from the Upper Laetolil Beds and those of extant giraffes
Element Taxon Sample size Mean (range) Standard Dev. Coeff. of Variation
Incisor ML BL ML BL
G. aff. jumae 2 14.7 (13.1–16.3) 12.3 2.26 N/A
G. stillei 11 11.4 (10.2–12.7) 8.3 (7.0–9.3) 0.60 0.70
S. maurusium 4 16.4 (13.3–18.9) 14.9 (13.6–15.8) 1.34 0.97
G. camelopardalis 14 14.4 (8.6–18.6) 9.8 (7.2–12.5) 2.86 1.77
Lower dC BL MD BL MD
G. aff. jumae 0
G. stillei 1 5.1 12.6
S. maurusium 0
G. camelopardalis 2 6.0 (5.4–6.6) 20.1 (19.6–20.6) 0.88 0.68
Canine BL MD BL MD
G. aff. jumae 2 7.5 (7.3–7.7) 22.6 (22.2–22.9) 0.28 0.50
G. stillei 2 6.2 (6.1–6.3) 16.8 (15.9–17.7) 0.14 1.27
S. maurusium 1 12.5 20.2
G. camelopardalis 5 8.4 (6.5–10.1) 23.4 (16.1–31.7) 1.63 6.28
Lower dP3 BL MD BL MD
G. aff. jumae 0
G. stillei 0
S. maurusium 2 17.0 (16.8–17.3) 30.7 0.35 N/A
G. camelopardalis 5 12.9 (12.5–13.9) 20.5 (18.4–23.2) 0.59 1.87
Lower dP4 BL MD BL MD
G. aff. jumae 0
G. stillei 2 13.6 (12.7–14.6) 24.7 1.34 N/A
S. maurusium 0
G. camelopardalis 5 15.6 (14.9–16.7) 32.1 (30.8–33.8) 0.68 1.08
Lower P2 BL MD BL MD
G. aff. jumae 0
G. stillei 2 10.6 (9.9–11.2) 18.1 0.92 N/A
S. maurusium 1 19.9
G. camelopardalis 11 15.7 (12.6–19.4) 19.6 (15.6–22.8) 2.12 1.98
G. aff. jumae 1 17.2 21.8 N/A N/A
G. stillei 8 15.3 (15.1–16.9) 21.0 (20.4–21.9) 0.97 0.71 0.063 0.034
S. maurusium 4 22.9 (20.7–23.8) 34.0 (3.0–35.0) 1.48 1.41 0.065 0.041
G. camelopardalis 12 20.4 (16.7–23.9) 23.1 (19.0–25.5) 1.73 1.47 0.085 0.064
G. aff. jumae 1 24.2 29.1 N/A N/A
G. stillei 16 16.6 (14.2–18.4) 21.4 (18.4–23.9) 1.31 1.43 0.079 0.067
S. maurusium 3 24.9 (21.0–28.0) 39.0 (37.8–40.2) 3.58 1.70 0.144 0.044
G. camelopardalis 12 21.8 (15.8–23.4) 24.9 (21.3–26.6) 1.04 1.29 0.048 0.052
Lower M1/M2 a BL MD BL MD BL MD
G. aff. jumae 2 23.8 (22.7–24.9) 32.9 1.56 N/A 0.066 N/A
G. stillei 44 18.2 (15.1–22.0) 24.8 (21.0–27.5) 1.45 1.75 0.080 0.071
S. maurusium 6 31.8 (29.5–33.9) 42.7 (41.5–43.6) 1.65 1.07 0.052 0.025
G. camelopardalis 24 22.1 30.2 (26.5–34.5) 1.42 1.78 0.064 0.059
G. aff. jumae 5 23.8 (21.8–25.5) 45.2 (43.9–46.5) 1.69 1.84 0.071 0.041
G. stillei 18 18.4 (15.9–21.7) 34.6 (31.6–37.7) 1.69 1.83 0.092 0.053
S. maurusium 1 65.5
G. camelopardalis 12 22.6 (18.8–25.1) 42.0 (37.5–44.7) 1.76 2.45 0.078 0.058
Upper dP2 BL MD BL MD
G. aff. jumae 0
G. stillei 1 12.8 16.0
S. maurusium 0
G. camelopardalis 5 17.3 (16.9–17.7) 20.4 (18.9–22.4) 0.29 1.39
14 Giraf fidae
Mesio-distal length (mm)
G. camelopardalis
G. cf. jumae
G. stillei
S. maurusium
Bucco-lingual breadth (mm)
20 25 30 35 40 45 50
Fig. 14.4 Bivariate plot of the mesiodistal length and buccolingual breadth measurements taken on upper molars of giraffids from the Upper
Laetolil Beds and extant Giraffa camelopardalis specimens
Table 14.3 (continued)
Element Taxon Sample size Mean (range) Standard Dev. Coeff. of Variation
Upper dP3 BL MD BL MD
G. aff. jumae 0
G. stillei 2 16.9 (15.8–17.9) 19.6 (18.6–20.5) 1.48 1.34
S. maurusium 2 26.8 (24.3–29.4) 34.6 (33.6–35.7) 3.61 1.48
G. camelopardalis 5 19.9 (18.7–20.3) 24.4 (23.1–25.9) 0.70 1.34
Upper dP4 BL MD BL MD
G. aff. jumae 0
G. stillei 1 18.0 20.3
S. maurusium 0
G. camelopardalis 5 22.8 (21.2–25.7) 26.4 (24.4–28.0) 1.86 1.37
G. aff. jumae 7 26.2 (23.9–27.8) 23.5 (21.9–25.0) 1.67 1.03 0.064 0.044
G. stillei 23 18.8 (16.5–21.2) 18.4 (15.9–21.1) 1.28 1.23 0.068 0.067
S. maurusium 1 41.2 32.6
G. camelopardalis 10 25.2 (21.9–28.5) 21.3 (18.9–24.9) 1.78 1.78 0.071 0.084
G. aff. jumae 6 27.3 (25.8–29.3) 23.7 (22.4–26.0) 1.19 1.62 0.044 0.068
G. stillei 25 21.2 (18.9–22.8) 18.5 (16.7–28.5) 1.27 1.12 0.060 0.061
S. maurusium 3 41.7 (39.7–44.2) 28.5 2.30 N/A 0.055 N/A
G. camelopardalis 10 29.0 (24.9–32.7) 22.6 (20.3–24.3) 1.93 1.06 0.067 0.047
G. aff. jumae 6 30.2 (29.2–32.0) 22.2 (21.4–23.3) 1.26 0.71 0.042 0.032
G. stillei 16 23.8 (21.8–26.6) 19.2 (17.3–22.4) 1.29 1.51 0.054 0.079
S. maurusium 2 49.8 (49.4–50.3) 36.9 0.64 N/A 0.013 N/A
G. camelopardalis 11 29.8 (27.7–32.5) 22.9 (20.2–24.5) 1.78 0.87 0.060 0.038
G. aff. jumae 15 35.6 (26.4–40.0) 32.1 (28.8–34.9) 1.99 2.14 0.056 0.067
G. stillei 61 25.9 (20.5–29.5) 24.1 (19.2–28.5) 1.90 2.08 0.073 0.086
S. maurusium 15 46.0 (41.0–52.0) 45.8 (36.4–49.1) 4.04 4.13 0.088 0.090
G. camelopardalis 37 30.9 (26.9–34.1) 31.4 (24.8–36.6) 2.40 1.67 0.078 0.053
ML mediolateral breadth, BL buccolingual breadth, MD mesiodistal length
a It is not possible to consistently differentiate giraffid M1s and M2s or the three upper molars and, consequently, the data for these teeth are
346 C.A. Robinson
No consistent differences could be identified in Giraffa
camelopardalis specimens to distinguish between the three
lower incisors. Consequently, the isolated incisors from
Laetoli are not attributed to serial position. The incisors
from Laetoli ascribed to Giraffa stillei are substantially
smaller than those of G. camelopardalis and it is possible
that some of the smaller specimens could represent Okapia
or Giraffa pygmaea. To explore this possibility further, the
products of the length and breadth measurements for the
incisors and lower molars (excluding M3) were calculated
and averaged for the three fossil giraffid species from
Laetoli and extant giraffes. The results show that, on aver-
age, the incisors attributed to G. stillei are 26% of the size
of G. stillei molars, while G. camelopardalis incisors are
21% of the size of their molars. This suggests that the inci-
sors attributed to G. stillei are not unusually small. G. stillei
incisors have considerably thinner roots than those of G.
camelopardalis. On worn incisors (e.g., EP 618/01 and
619/01) a notch can form midway along the occlusal mar-
gin that causes these specimens to superficially resemble
the bilobed canines of giraffids (Fig. 14.2). Some extant
giraffes exhibit a similar wear pattern, although the notch is
not as deep on any G. camelopardalis incisor in the AMNH
Giraffa stillei canines from the Upper Laetolil Beds have a
more asymmetrical crown than those attributed to G. aff.
jumae, with a distal lobule that is lower and substantially lon-
ger mesiodistally than the mesial lobule (Fig. 14.3). This shape
more closely recalls the typical morphology of extant giraffes.
The Giraffa stillei dP4s from Laetoli are trilobed like those
of extant giraffes, but lack the prominent lingual stylids
observed in G. camelopdardalis.
Giraffa stillei P
2s are proportionately thinner buccolin-
gually than those of extant giraffes, but resemble them in
having a deep talonid basin and an elongated, buccolingually
thin trigonid. P3s from Laetoli exhibit little morphological or
size variation, with the coefficients of variation for both the
length and breadth of these teeth lower than in G. camelop-
ardalis (Table 14.3). Giraffa stillei P
4s from the Upper
Laetolil Beds are more variable than those of extant giraffes,
with higher coefficients of variation for both the buccolin-
gual and mesiodistal dimensions. EP 3528/00, the specimen
that resembles the Lower Laetolil Beds P4, is also unusual in
having multiple stylids on its cingula, including a pronounced
metastylid on the buccal cingulum.
There do not appear to be any consistent differences to
separate the lower M1s and M2s of Giraffa species (Harris
1991). Therefore, unless a lower molar is part of an associ-
ated dental series, it is identified only as a lower molar or
lower M3.
Accessory cusps are not common on the lower molars of
Giraffa stillei other than a variably present ectostylid. Among
the 22 lower M1 and M2s, there are two especially small
specimens, EP 1427/98 and EP 2119/00, which are 10–15%
smaller than the next smallest G. stillei lower molar. These
specimens are nearly identical in size to a lower M2 (LAET
75-2215) in the Mary Leakey collection (Harris 1987).
Earlier studies of the giraffid fossils from the 1959 Laetoli
collections provisionally attributed some of the smaller spec-
imens to Giraffa pygmaea (Harris 1976b), although Harris
(1987) later revised this opinion. The three lower molars
from Laetoli mentioned above are similar in their buccolin-
gual breadth to the holotype of G. pygmaea (KNM-ER 778)
but they are approximately 20% longer mesiodistally, and
are probably best retained within G. stillei. Moreover, the
smallest of the three lower molars, EP 2119/00, lacks an
ectostylid, which is typically found on G. pygmaea lower
molars (Churcher 1978; Harris 1991).
EP 1700/04 preserves M2 and M3 in association, with the
M3 not fully erupted. The dimensions of the M3 clearly
identify it as Giraffa stillei. However, the M2 is mesiodis-
tally longer than any other G. stillei specimen from Laetoli
and overlaps slightly with the lower end of the G. camelop-
ardalis range. It is also larger than five isolated molar speci-
mens that were identified as G. aff. jumae (LAET 75-1491,
75-1578, 75-2207, 75-3047 and 75-3461) (Harris 1987).
These molars are considerably smaller than the other speci-
mens attributed to G. aff. jumae from Laetoli (EP 021/00,
EP 1057/03 and LAET 76-3986) (Fig. 14.5). Given these
data, the evidence that G. aff. jumae teeth are generally
larger than those of extant giraffes, and the relative preva-
lence of these two taxa in the Upper Laetolil Beds, it is sug-
gested that these five smaller molars be reassigned to Giraffa
To explore whether the inclusion of these smaller and
larger specimens in the G. stillei sample from the Upper
Laetolil Beds increases its intraspecific variation beyond
what would be expected for a single giraffe species, the coef-
ficients of variation (CVs) of the buccolingual and mesiodis-
tal dimensions were calculated for all permanent postcanine
teeth, other than P2, of the three fossil giraffid species and
Giraffa camelopardalis (Table 14.3). The CVs for the
buccolingual breadth and mesiodistal length of Giraffa stillei
lower molars are similar to, albeit slightly greater than, those
of extant giraffes. Given the evidence discussed above from
the type specimen of Giraffa pygmaea and from EP 1700/04
there does not appear to be sufficient cause to refute the sin-
gle species hypothesis for the lower molars currently attrib-
uted to Giraffa stillei.
Most of the eight Giraffa stillei lower M3s recovered by
EPPE from the Upper Laetolil Beds lack accessory cusps,
although EP 862/03 has a prominent ectostylid and hypostylid
on its buccal cingulum. EP 1078/01 is about 7% smaller than the
other isolated M3s, but it is similar in size to the M3 in the LAET
75-563 mandible. Like the other smaller lower molars discussed
above, these two specimens are mesiodistally elongated
14 Giraf fidae
compared to the M3 of the Giraffa pygmaea holotype. Moreover,
including these smaller molars within G. stillei does not result in
its coefficient of variation being substantially greater than that
of extant giraffes (Table 14.3).
Maxillary dentition: EP 1122/00 is an associated series of
maxillary teeth of a juvenile (RdP2-dP4 and LdP2), along with
M1 and M2, attributed to Giraffa stillei (Fig. 14.6). These
specimens were recovered from Loc. 8 and are the first
deciduous maxillary teeth described from Laetoli. All the
teeth are minimally worn and M2 appears to have still been
developing within its crypt. Deciduous P2s of Giraffa resem-
ble mesiodistally elongated permanent upper premolars and
frequently have small accessory cuspules on their pro-
nounced buccal cingula. Upper dP3s are also mesiodistally
elongated, due primarily to a lengthening of the mesial cusps,
but are otherwise similar morphologically to permanent
upper molars. The EP 1122/00 dP3 is notably longer mesi-
odistally than the other dP3 from the Upper Laetolil Beds (EP
3599/00). Maxillary dP4s are morphologically indistinguish-
able from G. stillei upper molars, but are substantially
It is possible to divide the G. stillei permanent upper pre-
molars from the Upper Laetolil Beds into two size groups.
However, the coefficients of variation for the buccolingual
and mesiodistal dimensions of these pooled teeth are similar
to, and usually less than, those of modern giraffes (Table 14.3).
This suggests that these two groups are more likely to be
males and females of a single species rather than two differ-
ent taxa.
Two P2s, EP 398/00 and 1532/03, are distinct from the
other Giraffa stillei specimens in having well defined meso-
styles on their buccal cingula. EP 345/98, EP 347/98 and, to
a lesser extent, EP 3163/00, three of the largest P3s, resemble
one another in having their lingual margins folded, such that
they come to a point at approximately the midline of the
crown. The other G. stillei upper P3s have a more evenly
rounded lingual margin. The upper P4s exhibit less morpho-
logical variation, with the occlusal outline typically being
more triangular in shape than that of P3, due to the P4s nar-
rowing more sharply towards the lingual margin of the
There does not appear to be any consistent morphological
criteria to identify the serial position of giraffid isolated
upper molars (Leakey 1970; Harris 1987, 1991). Consequently,
all isolated G. stillei specimens are simply identified, follow-
ing Harris (1987), as upper molars. A small entostyle is pres-
ent on two of the specimens (EP 2869/00 and EP 1667/04).
The largest G. stillei specimens overlap in size with the
smallest G. camelopardalis molars, but not with any G. aff.
jumae specimens from Laetoli (Fig. 14.4). The coefficients
of variation for the length and breadth dimensions of G. stil-
lei upper molars are similar to those of G. camelopardalis
(Table 14.3).
1020 25
Mesio-distal length (mm)
Bucco-lingual breadth (mm)
30 35 40
G. camelopardalis
G. cf. jumae
G. stillei
S. maurusium
Large ‘G. stillei’
EP 854/04
Fig. 14.5 Bivariate plot of the mesiodistal length and buccolingual
breadth measurements taken on lower molars (M1 and M2) of giraffids
from the Laetolil Beds and extant Giraffa camelopardalis specimens.
EP 854/04 is a mandibular specimen from the Lower Laetolil Beds
retaining M1-M3. The ‘large’ Giraffa stillei specimens are three of the
lower molars identified as Giraffa aff. jumae specimens by Harris
(1987) that may instead be large G. stillei specimens
Fig. 14.6 Occlusal view of EP 1122/00, an associated series of upper
teeth attributed to Giraffa stillei (RdP2-M2). Also associated with these
teeth, but not pictured, is a left dP2
348 C.A. Robinson
Sivatherium maurusium
Fifty four isolated teeth from the Upper Laetolil Beds are
attributed to Sivatherium maurusium based on their size and
greater hypsodonty compared with Giraffa specimens. In
addition, four specimens with associated dental elements were
assigned to this taxon (EP 588/00 – P4-M1; EP 1342/01 M1-M3;
EP 2324/03 – M1-M3; EP 1701/04 – P4-M1).
Sivatherium maurusium incisors are buccolingually
thickened compared to those of Giraffa, so that when worn
they have a more extensive occlusal surface area. The roots
of S. maurusium incisors are more robust relative to the size
of the crown than those of Giraffa. EP 1625/98 and EP
202/00 are more extensively worn on the buccal half of their
crowns. In EP 1625/98 the crescent-shaped wear pattern
formed is so pronounced that the apex of the crown comes to
a point and the tooth is nearly worn down to the root on its
buccal side (Fig. 14.2). This irregular wear pattern was not
observed on any other fossil or extant giraffid incisor and
could be due to a pathological condition, although the cause
is difficult to determine given the lack of extant examples.
The Sivatherium maurusium canine, EP 498/00, resem-
bles the incisors of this taxon in being relatively broader buc-
colingually than those of Giraffa (Fig. 14.3). In addition, the
robusticity of its root matches that of the incisors. It is
bilobed, with an elongated and lower distal lobule, compa-
rable to specimens of Giraffa stillei, but substantially more
robust. It is smaller, but otherwise resembles the canine asso-
ciated with the S. maurusium skeleton from West Turkana
(KNM-WT 16584). This is consistent with evidence sug-
gesting that there may be an increase in the size of S. mauru-
sium teeth through time (see discussion).
Sivatherium maurusium lower dP3s are longer mesiodis-
tally both absolutely and relative to buccolingual breadth
than those of Giraffa. Their talonids resemble those of
Giraffa dP3s, albeit with a mesiodistally expanded hypoco-
nid more similar to the hypoconid of a permanent molar.
Their trigonid is elongated, with a large separation on the
buccal margin of the tooth between the relatively small
metaconid and paraconid. Similar to the S. maurusium dP3s
from Ahl al Oughlam (Geraads 1996), the anterior lobe is
almost closed on both specimens.
Sivatherium maurusium P
2s are elongated mesiodistally
compared to those of Giraffa camelopardalis. They also nar-
row more sharply mesially, with the well developed proto-
conid in the midline of the tooth rather than being shifted
buccally as in extant giraffes. The crest extending from the
protoconid and curving lingually (Harris 1991) is weaker
than in G. camelopardalis. The distal end of the crown is flat-
tened compared to the more rounded and projecting distal
margin of Giraffa P2s (Harris 1991).
Sivatherium maurusium lower P3s typically differ from
those of Giraffa in having a reduced trigonid, particularly the
metaconid, and an expanded entoconid, although on the
S. maurusium specimen from Laetoli (EP 206/01) the ento-
conid is not as pronounced, with the hypoconid dominating
the talonid, leading to a buccolingually narrower tooth over-
all. This tooth has been described as in a more “primitive
stage of molarization” than those of Giraffa in terms of its
overall form (Geraads et al. 2004), although the lack of
molarization in spite of the large size of this tooth has also
been argued to be a derived feature of sivatheres (Geraads
The only identifiable Sivatherium maurusium P
4 recov-
ered by EPPE was EP 588/00, a fragmentary specimen found
in association with a partial M1. The EP 588/00 teeth are
relatively small, but they fall well above the range of Giraffa
camelopardalis. Only one other Sivatherium maurusium
lower molar was identified from the Upper Laetolil Beds (EP
695/05). This specimen is about 10% larger than the EP
588/00 M1, but closely resembles it morphologically. While
the CVs for both the buccolingual breadth and mesiodistal
length of most S. maurusium postcanine teeth are smaller
than those of extant giraffes, the coefficient of variation for
the buccolingual breadth of P4 is high (Table 14.3). This is
seemingly due to the inclusion of the P4 in the LAET 75-520
mandible, which is unusually narrow.
Sivatherium maurusium upper P3s from Laetoli exhibit
similar morphological variation to that observed among
Giraffa stillei specimens, with EP 401/04 resembling the
small Giraffa specimens EP 345/98 and 347/98 in having a
lingual margin that folds to create a point, while the other
Sivatherium specimens have a flatter lingual margin that
results in a more typically rectangular shape to their occlusal
outlines. Two Sivatherium upper P4s are identified, with the
better preserved, but well worn, EP 1101/05 appearing to be
more rectangular in shape than typical giraffid P4s, partly
due to extensive wear, but also because of a glue-filled crack
that distorts the lingual margin.
Two Sivatherium maurusium specimens, EP 1342/01 and
EP 2324/03, retain all three associated upper molars, although
none are well preserved. No entostyles were observed on any
of the S. maurusium upper molars. EP 139/98 is a relatively
small S. maurusium upper molar, but it is too broad buccolin-
gually to be classified as Giraffa.
Upper Laetolil Beds Postcranial Specimens
Given the small sample sizes and range of intraspecific varia-
tion (Harris 1987; Harris 1991), it is frequently difficult to
confidently attribute fragmentary and/or isolated giraffid post-
cranial specimens to particular species (Likius 2002). However,
following comparison of the material from the Upper Laetolil
Beds with extant and other fossil giraffe samples, most well
14 Giraf fidae
preserved postcranial specimens from Laetoli were provi-
sionally identified to species (Table 14.4). This was based pri-
marily on size, although there are some notable differences
between the postcrania of Giraffa and Sivatherium (Harris
1991; Likius 2002).
Giraffa aff. jumae
Two proximal radii and one distal radius are the only fore-
limb bones that can be provisionally attributed to Giraffa aff.
jumae from Laetoli. They are similar in size and morphologi-
cally indistinguishable from those of extant giraffes.
The carpals of Giraffa aff. jumae, especially the
cuneiforms, semilunars and unciforms, resemble those of
Giraffa camelopardalis, apart from some relatively minor
differences (Harris 1987). The anterolateral corner is more
extensively projecting on the EP 4298/00 magnum than on
specimens in the Mary Leakey collection (Harris 1987),
which more closely recalls the morphology of extant giraffes.
While G. aff. jumae scaphoids from Laetoli are similar in
their mediolateral breadth and proximodistal width to those
of extant giraffes, they are, on average, about 15% longer
anteroposteriorly, partly due to a posterior extension of the
bone. This gives the specimens the appearance of being
mediolaterally compressed (Harris 1987).
Table 14.4 Measurements (in mm) of giraffid postcranials from the Upper Laetolil Beds
Taxon Element Sample size Mean dimensions (range in parentheses)
G. aff. jumae Proximal radius 2 AP – 65.9 (61.0–71.7); ML – 116.6 (111.8–121.4)
Distal radius 1
Cuneiform 4 Length – 56.8 (56.1–58.3); Ant. DV – 52.9 (48.5–55.6); Post. DV – 65.5 (62.3–68.2)
Semilunar 3 Length – 75.4 (69.0–79.5); Ant. ML – 47.7 (41.5–52.1); Ant. DV – 52.0 (50.4–53.8)
Magnum 4 Length – 78.8 (75.3–82.2); Ant. DV – 34.4 (32.1–36.7); Post. ML – 54.2 (52.0–56.1)
Scaphoid 10 AP Length – 75.4 (69.6–84.4); Ant. DV – 38.7 (33.7–45.5); Ant. ML – 50.1 (46.6–52.2)
Unciform 2 Length – 70.8 (67.2–74.4); Breadth – 43.7 (42.3–45.1)
Pelvis fragment 2
Fibula 1
Astragalus 15 Lat. Length – 104.7 (102.6–112.1); Med. Length – 91.0 (88.0–99.1)
Calcaneum 6 Artic. ML – 89.4; t.c. AP – 60.3
External cuneiform 1 DV – 21.8
Naviculo-cuboid 3 ML Breadth – 97.6 (92.4–101.7); AP Length – 87.8 (84.4–91.3)
Proximal metacarpal 1 AP – 70.4; ML – 104.2
Proximal metatarsal 2 AP – 90.7; ML – 94.8
Distal metapodial 10 AP – 64.3 (57.1–76.3); ML – 99.1 (91.9–106.3)
Proximal phalanx 12 Length – 106.6 (106.1–107.1); Prox. AP – 49.5 (46.1–52.4); Prox. ML – 47.6 (43.9–52.4);
Dist. AP – 33.1 (32.5–33.8); Dist. ML – 42.8 (40.8–44.9)
Middle phalanx 4 Length – 60.0 (52.2–64.4); Prox. AP – 43.6 (43.1–47.1); Prox. ML – 42.8 (41.3–44.3);
Dist. AP – 45.8 (43.7–49.5); Dist. ML – 38.4 (36.0–42.2)
G. stillei Distal humerus 3 ML – 105.1
Proximal radius 4 AP – 56.9 (52.7–60.8); ML – 105.3 (103.2–107.3)
Distal radius 3 AP – 61.4 (59.1–63.7); ML – 87.9 (84.3–91.5)
Proximal ulna 3 Artic. ML – 56.5 (55.4–58.3)
Cuneiform 3 Length – 47.5 (46.2–48.7); Ant. DV – 40.1 (38.9–41.8); Post. DV – 54.6 (52.2–57.0)
Semilunar 10 Length – 60.0 (57.8–62.2); Ant. ML – 37.3 (31.4–44.0); Ant. DV – 42.4 (37.0–47.7)
Magnum 5 Length – 69.3; Ant. DV – 33.0 (31.3–33.9); Post. ML – 50.6 (50.0–51.2)
Scaphoid 6 AP Length – 57.0 (56.1–58.6); Ant. DV – 27.9 (27.7–28.1); Ant. ML – 41.6 (40.5–43.2)
Unciform 2 Length – 51.6 (51.3–52.0); Breadth – 40.2
Pelvis fragment 1 Acetabulum diameter – 70.2
Distal tibia 2 AP – 65.6; ML – 85.2 (84.6–85.8)
Fibula 17 Length – 48.2 (44.6–52.9)
Astragalus 57 Lat. Length – 87.3 (83.4–96.0); Med. Length – 76.5 (72.0–81.5)
Calcaneum 37 Length – 156.8 (147.3–166.2); Artic. ML – 71.7 (64.0–80.1); t.c. AP – 48.3 (37.6–59.1);
t.c. ML – 51.1 (40.0–59.1)
External cuneiform 11 Length – 71.3; DV – 17.1 (14.6–20.0)
Naviculo-cuboid 12 ML Breadth – 75.2 (66.7–85.2); AP Length – 71.9 (66.5–83.6)
Proximal metacarpal 1 AP – 53.0; ML – 79.7
Proximal metatarsal 2 AP – 69.6 (66.8–72.5); ML – 66.2 (62.3–70.2)
Distal metapodial 27 AP – 49.5 (47.7–52.9); ML – 75.4 (73.8–80.7)
350 C.A. Robinson
No hindlimb long bone specimens of Giraffa aff. jumae
were recovered by EPPE and the only fibula found was
incomplete, although it does retain the relatively narrow tib-
ial facet characterizing fossil Giraffa fibulae from Laetoli
(Harris 1987).
Most of the giraffid postcranial specimens recovered
from the Upper Laetolil Beds are carpals and tarsals (Harris
1987), with astragali making up approximately 25% of the
taxonomically identifiable postcranials. Because of the
large sample of astragali, measurements taken on this ele-
ment can be used to compare the extent of variation in post-
cranial specimens of the three Laetoli giraffid taxa and
Giraffa camelopardalis. Giraffa aff. jumae was found to
have higher coefficients of variation for both lateral and
medial dorsoventral lengths, while the other two fossil
giraffid species had lower CVs than in G. camelopardalis
(Table 14.5).
To explore the extent and pattern of intraspecific variation
among giraffid astragali further, the medial and lateral dimen-
sions of these specimens were graphed on a bivariate plot
(Fig. 14.7) and the least-squares regression line was calcu-
lated (y = 0.77x + 8.88). The results show that giraffes scale
consistently for these measurements (r2 = 0.95), with larger
giraffids, such as Sivatherium, having proportionately shorter
lateral condyles.
The three fossil giraffid taxa from the Upper Laetolil Beds
are reasonably well separated from one another, with some
overlap in the ranges of Giraffa aff. jumae and Sivatherium
maurusium. This overlap, due to four especially large Giraffa
astragali in the Mary Leakey collection that could potentially
be Sivatherium maurusium specimens, contributes to the
higher CVs for G. aff. jumae noted above. The range of G.
aff. jumae also overlaps with that of G. camelopardalis, but
the fossil taxon has larger astragali on average.
Most of the giraffid calcanei from the Upper Laetolil Beds
are distal or proximal fragments, with few preserving suffi-
cient morphology for accurate measurements to be taken.
The three fossil taxa are generally differentiated by the
Taxon Element Sample size Mean dimensions (range in parentheses)
Proximal phalanx 6 Length – 99.3 (94.1–102.8); Prox. AP – 46.1 (44.9–46.8); Prox. ML – 39.3 (36.5–43.2);
Dist. AP – 28.2 (25.9–30.2); Dist. ML – 36.2 (30.6–43.3)
Middle phalanx 3 Length – 48.4 (48.0–48.8); Prox. AP – 37.2 (37.2–37.3); Prox. ML – 39.0; Dist. AP –
38.0 (35.3–40.6)
S. maurusium Cuneiform 3 Ant. DV – 58.8 (56.4–61.6); Post. DV – 74.9 (73.4–76.3)
Semilunar 7 Length – 78.8 (74.9–81.1); Ant. ML – 53.3 (51.5–55.7); Ant. DV – 59.4 (57.3–62.9)
Magnum 4 Length – 87.9 (87.1–88.7); Ant. DV – 37.4 (36.0–38.8); Post. ML – 56.2
Scaphoid 6 AP Length – 93.9 (91.9–98.6); Ant. DV – 48.7 (42.7–54.1); Ant. ML – 56.2 (54.1–58.3)
Unciform 3 Length – 69.7 (65.9–72.2); Breadth – 57.9 (57.1–58.7)
Distal tibia 1 AP – 87.8
Fibula 2 Length – 66.4a
Astragalus 10 Lat. Length – 118.8 (114.2–120.6); Med. Length – 101.9 (98.6–104.3)
Calcaneum 1 Artic. ML – 93.2
External cuneiform 1 DV – 23.4
Naviculo-cuboid 2 ML Breadth – 117.4
Proximal metacarpal 1 AP – 81.9; ML – 114.8
Distal metapodial 4
Proximal phalanx 8 Length – 135.2 (124.2–146.2); Prox. AP – 60.0 (58.4–64.1); Prox. ML – 53.1 (50.5–56.5);
Dist. AP – 37.9 (35.9–39.0); Dist. ML – 49.3 (47.7–50.1)
Middle phalanx 4 Length – 66.2 (62.7–69.6); Prox. AP – 55.9 (53.3–58.5); Prox. ML – 56.0 (52.8–59.2);
Dist. AP – 54.2; Dist. ML – 49.0 (47.3–50.8)
AP anteroposterior, DV dorsoventral, ML mediolateral, Artic. AP anteroposterior breadth of the proximal end of the calcaneum, t.c. tuber calcis
a Estimated measurement
Table 14.4 (continued)
Table 14.5 Measurements (in mm) of all giraffid astragali recovered from the Upper Laetolil Beds and those of extant giraffes
Mean Standard Deviation Coefficient of Variation
Taxon Sample size Lateral Medial Lateral Medial Lateral Medial
G. aff. jumae 16 107.1 93.6 6.98 5.84 0.065 0.062
G. stillei 48 86.7 76.2 3.61 2.87 0.042 0.038
S. maurusium 6 120.7 103.3 5.84 5.00 0.048 0.048
G. camelopardalis 6 100.3 88.8 6.16 4.49 0.061 0.051
14 Giraf fidae
robusticity of the calcaneum, with the heel process being
considerably thicker in the larger taxa. The calcanei, navic-
ulo-cuboids and external cuneiforms of Giraffa aff. jumae
from Laetoli are indistinguishable from those of extant
Most of the distal metapodial specimens recovered from
the Upper Laetolil Beds preserve only one of the condyles,
which makes it difficult to attribute them to the hind- or fore-
limb. The identification of isolated condyles to species is
equally problematic (e.g., they could either be large G. stillei
metacarpals or small G. aff. jumae metatarsals). However,
the two fossil Giraffa species can generally be distinguished
from one another by their anteroposterior dimension and,
consequently, most specimens can be tentatively identified to
species. Only five giraffid metapodial specimens preserved
enough of the distal end to allow for a measurement of their
mediolateral breadths. The two attributed to G. aff. jumae are
broad mediolaterally relative to their anteroposterior dimen-
sion and are likely to be metacarpals. These specimens are
proportionately narrower anteroposteriorly than metacarpals
of G. camelopardalis, like most of the G. cf. jumae specimens
from Toros-Menalla and Kossom-Bougoudi (Likius 2002).
The three G. aff. jumae proximal metapodials, one meta-
carpal and two metatarsals, are larger, on average, than
those of G. camelopardalis, but can clearly be attributed to
Giraffa. The proximal metatarsals are differentiated from
those of Sivatherium in having four articular facets rather
than three, with Sivatherium lacking the posterior cuboid
facet (Harris 1991).
Attributing isolated proximal phalanges, particularly the
more fragmentary specimens from Laetoli, to taxon is
difficult given the high levels of intraspecific variation
among the various digits of the fore- and hindlimb of extant
giraffes. The more complete specimens attributed to G. aff.
jumae from the Upper Laetolil Beds have proportionately
large proximal and distal ends relative to length compared to
those of G. camelopardalis.
Giraffa stillei
Giraffid forelimb long bones are rare at Laetoli, and none
preserve more than the proximal or distal ends. The only
well preserved Giraffa stillei distal humerus recovered by
EPPE from the Upper Laetolil Beds (EP 593/05) closely
resembles in both size and morphology the two G. stillei dis-
tal humeri identified previously (Harris 1987). All three
specimens are similar in size to female G. camelopardalis
specimens, but with a relatively small medial condyle (Harris
1987). Two proximal radii and three distal radii are consider-
ably smaller than those of Giraffa camelopardalis, but indis-
tinguishable morphologically, and are assigned to G. stillei.
The three similarly-sized Giraffa stillei ulnae recovered by
EPPE are about 15% smaller in the mediolateral dimension
of their humeral articular facet than LAET 75-1629, a
proximal ulna in the Mary Leakey collection, which is
approximately the size of an extant giraffe. These specimens
likely represent female and male specimens of G. stillei as
they share a reduced lateral facet for the radius compared to
G. camelopardalis ulnae (Harris 1987).
Carpals of Giraffa stillei are relatively rare compared to
other skeletal elements attributed to this taxon. Cuneiforms,
unciforms and magnums of the two Giraffa species are simi-
lar morphologically to those of extant giraffes, differing from
one another primarily in size and in the G. stillei specimens
being somewhat less robust. Small Giraffa semilunars that
75 85 95
Lateral length (mm)
Medial length (mm)
105 115 125 135
G. camelopardalis
G. cf. jumae
G. stillei
S. maurusium
Ndolanya G. cf. stillei
Fig. 14.7 Bivariate plot of the medial and lateral dorsoventral lengths for giraffid astragali recovered from the Upper Laetolil and Upper Ndolanya
Beds and for extant Giraffa camelopardalis specimens
352 C.A. Robinson
articulate well with scaphoids attributed to G. stillei and have
‘waisted’ dorsal and ventral articular facets (Harris 1987) are
placed in the smaller Giraffa taxon. The anteroposterior
lengths of G. stillei scaphoids approach those of small female
G. camelopardalis specimens, while mediolateral breadth
measurements are on average about 20% smaller than those
of extant giraffes, which matches previous descriptions of G.
stillei scaphoids as being proportionately narrow (Harris
Giraffa camelopardalis tibiae exhibit considerable intraspe-
cific variation in the anteroposterior width of the distal
epiphysis compared to its mediolateral breadth. Giraffa stillei
distal tibiae are considerably mediolaterally narrower than
those of extant giraffes, but they overlap with smaller
G. camelopardalis specimens in their anteroposterior width.
The narrowness of G. stillei tibiae may be related to their
relatively small and modestly projecting lateral condyles
(Harris 1987) or to their smaller size as smaller ruminants are
thought to typically have proportionately smaller mediolateral
dimensions (D. Geraads, personal communication).
Of the giraffid fibula specimens recovered from the Upper
Laetolil Beds, 85% are attributed, based on size, to Giraffa
Tarsals of Giraffa stillei are much more common at Laetoli
compared to those of the other two fossil giraffid taxa. For
example, over twice as many G. stillei astragali were recov-
ered than those of the other two taxa combined. One G. stillei
astragalus has an unusually long lateral condyle, but other-
wise the specimens are quite similar to one another morpho-
logically and they cluster tightly together when comparing
their medial and lateral lengths (Fig. 14.7). Over five times
as many calcanei were identified as Giraffa stillei than those
attributed to the other two giraffids combined. Although they
exhibit a large variation in size, like extant giraffes, all
G. stillei calcanei are considerably smaller than those of
G. camelopardalis. EP 662/00 is notably smaller than the
other well preserved Giraffa stillei calcanei, particularly in
the robusticity of its heel process, but is similar in size to
LAET 76-3678 from the Mary Leakey collection (Harris
1987). The size and morphology of the tuber calcaneum is
quite variable in G. stillei, with the morphology of some
larger specimens resembling that of extant giraffes. As in the
other tarsal elements, there is a substantially larger sample of
Giraffa stillei external cuneiforms from the Upper Laetolil
Beds than those of the other two taxa. However, only one G.
stillei specimen (EP 144/98) is complete enough for a mea-
surement of anteroposterior length and, while this dimension
overlaps with the lower end of the extant giraffe size range,
it is about 10% smaller than the mean for G. camelopardalis,
and it articulates well with G. stillei naviculo-cuboids.
Moreover, like G. stillei specimens in the Mary Leakey
collection (Harris 1987), the metatarsal facets on the external
cuneiform are almost continuous. There are slight differ-
ences among the three fossil taxa in the dorso-ventral breadth
of their external cuneiforms, but the range of G. stillei for
this measurement was extensive, and substantially over-
lapped the range of extant giraffes. The naviculo-cuboids
follow the same pattern of relative abundance as the other
tarsals, with considerably more Giraffa stillei specimens
recovered than those of the other giraffids. The G. stillei
specimens are mediolaterally narrower relative to their
anteroposterior length than G. aff. jumae and Giraffa cam-
elopardalis naviculo-cuboids, which is likely related to a
mediolaterally reduced calcaneal facet. As Harris (1987)
noted, the facet for articulation with the medial edge of the
calcaneum does not extend posterior to the facet for the lat-
eral trochlea of the astragalus in most of the G. stillei speci-
mens from Laetoli.
Three Giraffa stillei distal metapodials are complete
enough to identify one as a metatarsal and the other two as
metacarpals. The metatarsal, EP 495/01, has both the proxi-
mal and distal ends preserved, but is missing about one-third
of the shaft of the bone. It is also associated with a naviculo-
cuboid specimen that is fused to the external cuneiform. The
metatarsal shaft is typical of giraffine metapodials in being
long and slender with a shallow posterior trough (Solounias
Giraffa stillei proximal phalanges are proximodistally
shorter than those of extant giraffes but, as in G. aff. jumae
specimens, they have proportionately large proximal and dis-
tal ends. Three middle phalanges are considerably smaller
than those of Giraffa aff. jumae and, although they overlap
with the lower end of G. camelopardalis size range for both
length and the size of their proximal and distal ends, they are
tentatively placed in G. stillei.
Sivatherium maurusium
Unlike at Koobi Fora (Harris 1991), massive postcranial
specimens have been recovered from the Upper Laetoli Beds
that seem likely to represent S. maurusium rather than
extremely large Giraffa aff. jumae specimens.
The carpals attributed to Sivatherium maurusium from
Laetoli are generally larger and more robust than those of
Giraffa. None of the S. maurusium cuneiforms recovered by
EPPE are complete, but they are substantially more robust
than any Giraffa specimen and they articulate well with unci-
forms attributed to this taxon. The anteroposterior lengths of
the Sivatherium maurusium semilunars overlap with those of
large Giraffa specimens, while the breadth of these elements
clearly separate specimens of the two genera from one
another. Dorsoventral breadth measurements on magnums
were not effective in differentiating giraffid taxa, and
14 Giraf fidae
S. maurusium and G. aff. jumae were also similar in
their mediolateral dimension. Since S. maurusium had a
substantially longer magnum than G. aff. jumae, this indicates
that either Giraffa has a relatively broad magnum (Harris
1991) or S. maurusium has a proportionately longer magnum.
No evidence of a longitudinal groove in the scaphoid facet
(Harris 1987) was noted on the S. maurusium magnum recov-
ered by EPPE (EP 3918/00). Sivatherium maurusium sca-
phoids are extremely robust compared to those of Giraffa and
considerably larger in all dimensions. The proximal (radial)
facet on the scaphoid is more excavated on Sivatherium speci-
mens than on those of Giraffa, although a relatively deep
facet has also been observed on some G. aff. jumae specimens
(Harris 1987) and one G. stillei specimen (EP 1128/04). S.
maurusium unciforms are similar to those of G. aff. jumae in
their anteroposterior length, but considerably broader medio-
laterally seemingly due to a mediolaterally expanded semilu-
nar facet.
One Sivatherium maurusium partial distal tibia was
recovered by EPPE (EP 402/00). It does not appear to be
especially anteroposteriorly compressed compared to Giraffa
tibiae like specimens from Koobi Fora (Harris 1991),
although its mediolateral dimension could not be accurately
measured due to incomplete preservation. The one sivathere
fibula recovered by EPPE (EP 668/98) resembles those from
the Mary Leakey collection in having a relatively broad tibial
facet, but its spine projects upwards like Giraffa fibulae and
not outwards like the other S. maurusium specimens from
Laetoli (Harris 1987).
Well preserved tarsals of Sivatherium maurusium are
relatively rare. Astragali of S. maurusium are notably more
robust and larger than those of the two Giraffa species.
Differences between this taxon and Giraffa in the morphol-
ogy of the two trochleae that were noted on Koobi Fora
specimens (Harris 1991) were not observed on these sub-
stantially larger Sivatherium astragali from Laetoli. Likius
(2002) has argued that these characters are extremely vari-
able in both genera and differentiating isolated astragali to
genus based exclusively on them would be difficult. There
was also a difference in the relative lengths of the medial
and lateral condyles of Sivatherium and Giraffa astragali
from Koobi Fora (Harris 1991). At Laetoli the disparity is
less pronounced. Sivatherium maurusium has a lateral
condyle that is about 17% longer on average than the medial
condyle, while it averages about 14–15% longer on Giraffa
astragali. The S. maurusium naviculo-cuboid from the
Upper Laetolil Beds (EP 1885/03) can be differentiated
from those of Giraffa by its massive size and reduced poste-
rior ridge (Harris 1991), which leads to an anteroposteriorly
narrower bone overall. Adhering matrix makes it difficult to
assess the morphology of the articular facets. The S. maurusium
calcaneum (EP 2543/00) recovered by EPPE only preserves
the proximal end and has a large and saddled-shaped astra-
galar facet, with its medial margin extending further distally
than on Giraffa specimens from Laetoli.
All Sivatherium maurusium distal metapodials recov-
ered by EPPE from the Upper Laetolil Beds are fragments,
but they appear to correspond to previous descriptions of
these elements in being anteroposteriorly compressed with
proportionately wide epiphyses (Harris 1976a, 1991). The
epiphysis of the Sivatherium maurusium proximal metacar-
pal recovered by EPPE (EP 3604/00) was anteroposteriorly
wider relative to its mediolateral breadth than Giraffa spec-
imens, leading to a squarer outline to its articular surface.
This is likely related to the medial and lateral facets remain-
ing distinct on the articular surface of the S. maurusium
specimen (Harris 1976a, 1991).
Sivatherium maurusium proximal and middle phalanges
from the Upper Laetolil Beds are notable for having substan-
tially enlarged proximal and distal ends compared to Giraffa
camelopardalis specimens of similar length. This is consis-
tent with descriptions of Sivatherium phalanges as being
relatively short compared to those of Giraffa (Likius 2002).
Upper Ndolanya Beds
Sixty-two giraffid specimens were recovered from the Upper
Ndolanya Beds by EPPE (Table 14.6). Since no diagnostic
specimens have been found from this unit, species designa-
tions should be considered provisional. However, all three of
Table 14.6 Upper Ndolanya Beds giraffid specimen numbers at each locality
Giraffa aff. jumae Giraffa aff. stillei Sivatherium aff. maurusium
Locality Dental Postcranial Total Dental Postcranial Total Dental Postcranial Total
7E 0 1 1 3 6 9 0 0 0
15 0 4 4 0 13 13 1 0 1
18 0 3 3 0 15 15 3 0 3
22S 0 2 2 1 2 3 0 0 0
Silal Artum 0 0 0 0 6 6 0 0 0
Totals 0 10 10 4 42 46 4 0 4
354 C.A. Robinson
the giraffid taxa from the Upper Laetolil Beds, along with
Giraffa pygmaea, appear to be represented in the Upper
Ndolanya Beds.
Upper Ndolanya Beds Dento-Dnathic
Seven isolated giraffid teeth were recovered from the Upper
Ndolanya Beds, with none of the specimens well preserved. Four
fragmentary teeth, including one dP4, were attributed to Giraffa
aff. stillei (Table 14.7). All three of the permanent teeth are mor-
phologically indistinguishable from, but larger than, the average
G. stillei specimen from the Upper Laetolil Beds, although none
overlapped with the size ranges of G. aff. jumae teeth.
One giraffid partial M3 and two upper molar fragments
were attributed to Sivatherium aff. maurusium because of
their massive size and degree of hypsodonty. In addition, a
rolled and fragmented partial mandible of S. aff. maurusium
that retains the broken crowns of P4-M2 (EP 1040/00) was
recovered from Loc. 18 (Fig. 14.8). The teeth in this speci-
men only preserve enough morphology to estimate their size.
Based on these estimates, the mesiodistal length and bucco-
lingual breadth of M1 and M2 are greater than those of the
largest S. maurusium specimen known from the Upper
Laetolil Beds. The mandible is more robust and approxi-
mately 40% deeper than the LAET 75-520 specimen from
Fig. 14.8 Buccal view of the Sivatherium maurusium mandibular
specimens LAET 75–520 from the Upper Laetolil Beds and EP
1040/00 from the Upper Ndolanya Beds
Table 14.7 Measurements (in mm) of giraffid specimens from the Upper Ndolanya Beds
Taxon Element Sample size Mean dimensions (ranges in parentheses)
G. aff. jumae Glenoid of scapula 1 Length – 94.3; Breadth – 79.6
Proximal radius 1 AP – 71.9; ML – 113.6
Distal radius 2 AP – 76.5; ML – 115.6
Magnum 1
Distal tibia 1 AP – 67.7; ML – 92.7
Calcaneum 1 Artic. ML – 75a
Naviculo-cuboid 1 ML Breadth – 91.3; AP Length – 84.5
Distal metapodial 2
G. aff. stillei Lower dP4 1
Lower molar 1 MD – 26.0
Upper P2 1 BL – 19.9
Upper molar 1 MD – 25.3
Glenoid of scapula 2 Length – 69.1; Breadth – 67.5
Distal humerus 1
Cuneiform 1 Length – 46.8; Post. DV – 55.2
Magnum 1 Length – 69.9; Post. ML – 47.8
Scaphoid 1 AP Length – 67.8; Ant. ML – 41.4
Distal tibia 3 AP – 64.6 (60.2–68.0); ML – 81.7
Astragalus 10 Lat. Length – 90.2 (86.3–92.7); Med. Length – 77.4 (73.5–80.6)
Calcaneum 4 t.c. AP – 43.0
External cuneiform 1 DV – 18.5
Naviculo-cuboid 2 ML Breadth – 76.6 (72.8–80.5); AP Length – 75.0 (66.5–83.6)
Distal metapodial 5 AP – 47.7; ML – 80.7
Proximal phalanx 10 Length – 95.4 (92.4–98.0); Prox. AP – 45.0 (43.2–46.3); Prox. ML – 43.3
(41.2–47.1); Dist. AP – 29.3 (28.5–30.3); Dist. ML – 41.8 (40.2–43.6)
Middle phalanx 1 Length – 47.5; Prox. AP – 44.9; Prox. ML – 41.5; Dist. AP – 40.9; Dist. ML – 42.1
S. aff. maurusium Mandible with P4-M2 1 Depth at P4/M1 – 73.7; Depth at M1/M2 – 82.8; M1: BL – 36.3a; MD – 48.5a;
M2: BL – 35.0a; MD - 50.1a
Lower M3 1
Upper Molar 2
BL buccolingual breadth, MD mesiodistal length, AP anteroposterior, DV dorsoventral, ML mediolateral, Artic. AP anteroposterior breadth of the
proximal end of the calcaneum, t.c. tuber calcis
a Estimated measurement
14 Giraf fidae
the Mary Leakey collection. While this could be attributed to
sexual dimorphism, the difference is quite striking, particu-
larly given that the depth of the LAET 75-520 mandible is
artificially increased by crushing and, in spite of its poor
preservation, this does not appear to be true of the EP 1040/00
No cranio-dental specimens of Giraffa aff. jumae were
identified among the giraffids from the Upper Ndolanya
Beds. Harris (1987) assigned one dental specimen from
Loc.14 to that taxon, but did not provide further details or
metric data, which would suggest that it is an incomplete
Upper Ndolanya Beds Postcranial Specimens
Giraffa aff. jumae
Ten postcranial specimens are provisionally assigned to
Giraffa aff. jumae pending recovery of cranio-dental elements
that can be attributed to that taxon or another giraffid of
similar size (Table 14.7). EP 3440/00 is a glenoid of a scapula
that is similar in size to those of large male extant giraffes,
although its articular surface is relatively narrow and, conse-
quently, more ovoid. A proximal radius, EP 3434/00, is
approximately the same size as those of extant giraffes and G.
aff. jumae specimens from the Upper Laetolil Beds, but it is
unusually wide anteroposteriorly compared to its mediolat-
eral breadth. Two distal radii are similar in both size and mor-
phology to large male G. camelopardalis specimens. A distal
tibia (EP 1224/00) approximates the average size and typical
morphology of female G. camelopardalis specimens, and is
considerably broader mediolaterally than G. stillei specimens
from Laetoli. A matrix encrusted and fragmentary proximal
calcaneum (EP 1203/03) is similar in its robusticity to G. aff.
jumae specimens from the Upper Laetolil Beds. A naviculo-
cuboid (EP 1685/03) from the Upper Ndolanya Beds is
approximately the same size as the smallest G. aff. jumae
specimen from the Upper Laetolil Beds. It is considerably
larger than any G. stillei specimen and does not share a
reduced calcaneal facet with those specimens.
Giraffa aff. stillei
Forty two postcranial specimens recovered from the Upper
Ndolanya Beds were assigned to Giraffa aff. stillei. These
fossils comprise approximately 75% of the giraffid specimens
collected by EPPE from this unit. Much of this material,
including a distal humerus, is relatively fragmentary and is
tentatively assigned to this taxon based on similarities to spec-
imens from the Upper Laetolil Beds. Two scapular glenoids
are smaller than, but indistinguishable morphologically from,
those of extant giraffes.
Three carpals are attributed to Giraffa aff. stillei, with the
cuneiform and magnum similar in size and morphology to
those from the Upper Laetolil Beds. A scaphoid (EP 4003/00)
is provisionally assigned to this taxon due to its being rela-
tively narrow mediolaterally compared to Giraffa camelop-
ardalis and Giraffa aff. jumae specimens, like the Upper
Laetolil Beds G. stillei scaphoids, although it is over 15%
longer anteroposteriorly than any G. stillei specimen from
that unit.
Three distal tibiae were recovered that are morphologi-
cally similar to Giraffa stillei specimens from the Upper
Laetolil Beds in being proportionately narrower mediolater-
ally than those of extant giraffes with a less protuberant
lateral condyle.
Just as in the Upper Laetolil Beds, a large proportion, over
40%, of the postcranial specimens attributed to Giraffa aff.
stillei from the Upper Ndolanya Beds are tarsals. Most of
these specimens are astragali, and in general they fall within
the upper end of the Upper Laetolil Beds G. stillei size range
(Fig. 14.7). A number of G. aff. stillei astragali from the Upper
Ndolanya Beds have a noticeably thicker medial condyle than
specimens from the Laetolil Beds. All of the G. aff. stillei
external cuneiforms and calcanei are fragmentary, although
one calcaneum, EP 3286/00, retained parts of its proximal
and distal ends and was associated with the smallest astrag-
alus from this unit. One of the naviculo-cuboids, EP 1327/05,
preserves a relatively broad articular facet for the medial con-
dyle of the astragalus, which corresponds with the thicker
medial condyles of the astragali mentioned above. Like many
of the G. stillei specimens from the Upper Laetolil Beds it
lacks a medial extension of the calcaneal facet.
Only one of the distal metapodial specimens attributed to
Giraffa aff. stillei preserved enough morphology for the
mediolateral breadth of the epiphysis to be measured and it is
likely to be a metatarsal.
The relatively large number of Giraffa aff. stillei proximal
phalanges from the Upper Ndolanya Beds generally have
larger proximal and distal articular surfaces, but are dors-
oventrally shorter, than G. stillei specimens from the Upper
Laetolil Beds. A dorsoventrally short middle phalanx recov-
ered from the Upper Ndolanya Beds also has relatively
enlarged proximal and distal ends compared to G. stillei
specimens from the Laetolil Beds. Based on the dimensions
of the proximal phalanges from these strata, it is tentatively
assigned to G. aff. stillei, although it may represent a dors-
oventrally shortened G. aff. jumae specimen.
No postcranial specimens were recovered by EPPE from
the Upper Ndolanya Beds that could be attributed to
Sivatherium aff. maurusium, although four specimens of this
taxon were listed as having been recovered from Loc. 7E by
Harris (1987).
356 C.A. Robinson
Additional Specimen from the Upper Ndolanya Beds
During an examination of the giraffid material recovered by
Kohl-Larsen’s team that is currently housed at the Museum
für Naturkunde in Berlin, Terry Harrison noted an unusu-
ally small mandibular specimen retaining P3-M3 (MB Ma.
39792) that is likely to be derived from the Upper Ndolanya
Beds (annotations on the specimen indicate that it was
recovered from the Gadjingero, which is equivalent to Mary
Leakey’s collecting Loc. 18) and was identified as “Okapia
2 sp. (Fig. 14.9). The dimensions of all teeth are smaller
than those of the holotype of Giraffa pygmaea (KNM-ER
778) with the M2 nearly identical in size to two M1s attrib-
uted to that species from Olduvai (Harris 1976b). As is
typical for specimens of that taxon (Churcher 1978; Harris
1991), ectostylids are present on all three molars and there
is a hypostylid on M3. MB Ma. 39792 lacks an entostylid,
present on the G. pygmaea M3s from Koobi Fora. However,
as discussed below, this appears to be a variable character
found on some G. stillei specimens as well. The mandible
is shallow, as in the type specimen of G. pygmaea, although
the Laetoli specimen deepens posteriorly. Since only a tiny
fragment of the mesiobuccal corner of the P3 crown remains
and P4 is missing much of the mesial portion of the tooth, it
is not possible to determine whether the specimen possesses
the diagnostic premolar morphology of Giraffa (Harris
1976b, 1987). However, there are no known Pliocene
Okapia fossils (Churcher 1978), while Giraffa pygmaea is
present at a number of East African Pliocene sites (Harris
1991; Kullmer et al. 2008; Reed 2008). Thus, it seems
likely that this specimen represents Giraffa. Since Giraffa
stillei is diagnosed as a species with teeth that are “always”
larger than those of Giraffa pygmaea (Harris 1991, 2003),
this specimen cannot be attributed to the former taxon.
Based on the size and morphology of its dental elements,
MB Ma. 39792 is provisionally assigned to Giraffa aff.
Lower Laetolil Beds
Giraffa aff. stillei
The Lower Laetolil Beds are dated to between 3.8 and older
than 4.3 Ma and they could date to as old as 4.6 Ma (Drake
and Curtis 1987; Su and Harrison 2007). If Giraffa stillei can
be confirmed from these strata, it would represent one of the
earliest known occurrences of this taxon in East Africa. This
species has also been identified from the Apak Member at
Lothagam, Kanapoi, and, provisionally, Aramis; three East
African sites that are of similar age to the Lower Laetolil
Beds (WoldeGabriel et al. 1994; Harris 2003; Harris et al.
2003). The specimens thus far recovered from the Lower
Laetolil Beds appear to be typical of the species and closely
resemble those found in the Upper Laetolil Beds.
Two Sivatherium species have been identified from sites dated
to the late Miocene and early Pliocene, Sivatherium mauru-
sium and Sivatherium hendeyi. The latter is known from ear-
lier sites and considered to be more primitive in its ossicone
morphology (Harris 1976a; Likius 2002), although some
researchers have argued that the differences between the two
species represent intraspecific variation (Churcher 1978).
Most of the earliest fossils attributed to Sivatherium from
sub-Saharan Africa cannot be identified to species because
of the lack of diagnostic cranial specimens or complete meta-
podials. The earliest well documented Sivatherium speci-
mens in East Africa come from the Middle Awash in Ethiopia,
dated to between 5.2 and 5.8 Ma (Haile-Selassie et al. 2004).
A fragmentary lower molar from the late Miocene/early
Pliocene Nkondo Formation in the Albertine Rift Valley of
Uganda also appears to represent Sivatherium (Geraads
1994). Sivatherium may also be present in the late Miocene
Lower Nawata Formation at Lothagam, although this is
based on only one large distal metapodial (Harris et al. 1988).
Finally, Gentry (1997) has tentatively identified Sivatherium
from the early Pliocene site of Kiloleli 2 in the Manonga
Valley. Sivatherium hendeyi is best known from the early
Pliocene site of Langebaanweg, in South Africa, where it is
Fig. 14.9 Occlusal and buccal views of the Giraffa aff. pygmaea man-
dibular specimen (MB Ma. 39792) from “Gadjingero”, which is thought
to correspond to Loc. 18 in the Upper Ndolanya Beds
14 Giraf fidae
the most common artiodactyl in the faunal assemblage
(Harris 1976a; Franz-Odendaal et al. 2003), but it has also
been provisionally identified on the basis of what was
described as an “ossicone fragment” at Kanapoi (Harris et al.
2003). The Sivatherium mandibles and postcranial material
from the late Miocene and early Pliocene sites of Toros-
Menalla and Kollé in Chad are provisionally assigned to that
species as well, based primarily on the metapodial specimens
being elongated compared to those of S. maurusium (Likius
2002; Vignaud et al. 2002), which is one of the diagnostic
characters of S. hendeyi. Thus, it appears that both Sivatherium
species are present in Africa by the early Pliocene, although
the first diagnostic evidence of S. maurusium in sub-Sarahan
Africa is from the Upper Laetolil Beds. Since the only evi-
dence for Sivatherium so far recorded from the Lower
Laetolil Beds is an isolated P4, and these species are not dis-
tinguishable by their dental morphology (Harris 1976a;
Harris 2003; Likius 2002), the specimen is provisionally
identified as Sivatherium sp.
Giraffidae indet.
The magnum recovered from the Lower Laetolil Beds is
substantially larger than those attributed to Giraffa stillei
from Laetoli, and is not proportionately elongated like the
Sivatherium maurusium specimens from the Upper Unit. It
is approximately the size of Giraffa aff. jumae specimens
from the Upper Laetolil Beds and could potentially be attrib-
uted to that taxon, although the shape of its scaphoid facet is
unusual. Giraffa jumae has been identified from the late
Miocene and early Pliocene sites of Kanapoi, Langebaanweg,
and, provisionally, Aramis and the various Mio-Pliocene
sites in the Djourab Desert of Chad (Harris 1976a;
WoldeGabriel et al. 1994; Likius 2002; Harris et al. 2003).
However, Palaeotragus is also known from late Miocene
and early Pliocene sites, including the Upper Nawata
Formation at Lothagam, the Middle Awash and, provision-
ally, the Pelletal Phosphorite Member at Langebaanweg
(Churcher 1978, 1979; Hendey 1981; Harris 1991, 2003;
Franz-Odendaal et al. 2003; Haile-Selassie et al. 2004).
There is also a giraffid maxilla recovered from Asbole
(below the “top conglomerate” fossils), that Geraads and
colleagues (2004) speculate could possibly be a Pliocene
palaeotragine based on differences in the premolar morphol-
ogy between this specimen and those of Giraffa species.
Finally, Geraads (1985) argued that an anterior ossicone
from the Pelletal Phosphorite Member at Langebaanweg,
attributed by Harris (1976a) to S. hendeyi, is probably
instead a palaeotragine. Palaeotragus germaini Arambourg,
1959 the largest known species within this genus (Harris
et al. 2003), is not substantially smaller than extant giraffes
(Churcher 1978, 1979; Harris 2003) and has “characteristi-
cally giraffid” limb bones (Churcher 1978). While Harris
and colleagues (2003) argue that sivatheres and giraffines
replaced palaeotragines at the end of the Miocene, it is con-
ceivable that some late surviving members were present in
the Lower Laetolil Beds and at Asbole and Langebaanweg.
Consequently, it seems prudent to not assign this specimen
to a genus at this time.
Upper Laetolil Beds
In the Mary Leakey collections from the Upper Laetolil Beds
Giraffa stillei specimens are approximately twice as com-
mon as G. aff. jumae specimens, while there are about 60%
more specimens of the large Giraffa taxon than those attrib-
uted to Sivatherium maurusium (Harris 1987). During the
1998–2005 field seasons EPPE recovered three times as
many G. stillei specimens as G. aff. jumae specimens
(Table 14.2). While more specimens were attributed to G.
aff. jumae by EPPE than to Sivatherium maurusium, the
difference in their numbers was not as substantial as in the
Mary Leakey collection.
Previously either Giraffa aff. jumae or Sivatherium was
the most common giraffid at 8 of the 25 Upper Laetolil local-
ities (Harris 1987). When the specimens recovered by EPPE
from the 28 currently recognized localities where giraffids
have been found in the Upper Laetolil Beds are added to
these totals, G. aff. jumae is the most common species only
at Localities 4 and 12 (Table 14.2). However, three Giraffa
stillei fossils were recovered when collections from Locs. 12
or 12E were combined, and if these are added to the totals
from Locality 12 then G. aff. jumae would no longer be the
most common species at that locality. There are no localities
at which Sivatherium is the most common taxon, although
the only giraffids recovered from Loc. 24 were one S. mau-
rusium and one G. aff. jumae specimen. Thus, there are only
three out of 28 localities in the Upper Laetolil Beds at which
G. stillei is not the most prevalent taxon. These three sites are
among the least productive sites for giraffids (and mammals
in general) in the Upper Laetolil Beds, suggesting that the
relative numbers of giraffids at these localities may not be
representative due to inadequate sampling.
Some unusual preservational patterns were observed
among the giraffid fossils recovered from the Upper Laetolil
Beds. First, postcranial remains were more common than
dental specimens only for Giraffa aff. jumae, although the
difference between the three giraffid taxa with respect to this
ratio was not as extreme as previously recorded (Harris
1987). It is not readily apparent why there is a persistent dif-
ference between G. aff. jumae and the other two giraffid taxa
in the proportion of postcranial specimens. Second, a com-
parison of the relative numbers of podials from the Upper
358 C.A. Robinson
Laetolil Beds shows that almost three times as many Giraffa
stillei tarsals were recovered than those of the other two
giraffid taxa combined, while for the carpals there have been
fewer G. stillei specimens found than those of the other two
taxa. This does not appear to be due to search image bias as
two different teams working a number of years apart pro-
duced similar results (Harris 1987). It seems unlikely, given
the similarities in the overall density of these bones that this
is due to a greater fragility of G. stillei carpals compared to
tarsals. Another possibility is that carnivores were preferen-
tially consuming the forelimbs of the smaller G. stillei indi-
viduals. However, 11 of the 16 forelimb long bones recovered
from the Upper Laetolil Beds were attributed to G. stillei,
which suggests that carnivore predation is unlikely to be the
causal factor.
Giraffa aff. jumae
Giraffa jumae and Giraffa camelopardalis are distinguished
from one another primarily based on cranial morphology
(Harris 1987, 1991; Geraads et al. 2004). However, a few
additional differences between these two taxa were provi-
sionally identified above. These include G. jumae having:
(1) bilobed lower canines that are lower crowned, with mesial
and distal lobules that are more evenly rounded and more
similar in shape to one another; (2) relatively buccolingually
broad upper molars (see discussion below) and; (3) metacar-
pals that are relatively narrow anteroposteriorly.
The teeth of Giraffa jumae have been described as similar
in size to the largest extant giraffe males (Leakey 1965,
1970; Harris 1976b; Churcher 1978), and Harris (1987) pre-
viously stated that the lower molars of Giraffa aff. jumae
from the Upper Laetolil Beds are buccolingually broader
than those of extant giraffes. However, it seems likely that he
was referring to the upper molars, rather than the lowers,
since, as mentioned above, five of the six M1s and M2s in the
Mary Leakey collection are smaller than those of G. cam-
elopardalis, while the upper molars from the Upper Laetolil
Beds are relatively broad compared to those of extant
giraffes. Measurements taken on G. jumae upper molars
from Koobi Fora and Bed II at Olduvai housed at the
National Museum of Kenya are similar to those from
Laetoli. Published data indicate that G. jumae specimens
from Langebaanweg, Kanjera and the type specimen from
Rawe also have proportionately broader upper molars than
extant giraffes (Harris 1976a, b). When all available data on
Giraffa jumae specimens are pooled, their upper molars are
found to be significantly broader (Student’s two-tailed t-test;
p < 0.0001), approximately 15% broader on average, than
those of G. camelopardalis, while there is no significant dif-
ference (p = 0.08) between the mesiodistal lengths of these
teeth. Data from Laetoli and Langebaanweg astragali sug-
gest that G. jumae is also larger than G. camelopardalis
postcranially, with the medial and lateral condyles signifi-
cantly longer in the fossil taxon (p < 0.01).
Unlike G. stillei (see below), there is no apparent trend for
Giraffa jumae to increase in dental size through time. It is
more difficult to assess whether a change occurred in the size
of the postcranial elements, given the problems with being
confident about the taxonomic attribution of isolated giraffid
postcranial bones at many sites. Harris (1991) suggests that
this possible stasis in size may be because potential food
sources, such as Acacia trees, place an upper size limit on
giraffes, which Giraffa jumae may have reached.
Giraffa stillei
While most of the Giraffa stillei from the Upper Laetolil
Beds are quite similar to specimens attributed to that taxon
from other East African sites, a few smaller lower molars
could potentially be derived from a fourth, smaller giraffid
species, possibly Giraffa pygmaea. It is not uncommon for
the two smaller Giraffa species to co-occur. Giraffa stillei is
known from many East African Pliocene sites including
Hadar, Kanapoi, Koobi Fora, Lothagam, Omo and South
Turkwel (Harris 1976b, c, 1991, 2003; Ward et al. 1999;
Harris et al. 2003; Reed 2008), while Giraffa pygmaea has
been identified at Hadar, Koobi Fora, Omo and, provision-
ally, Galili (Harris 1991; Kullmer et al. 2008; Reed 2008).
As discussed above, the smaller dental specimens from the
Upper Laetolil Beds are substantially longer mesiodistally
than those of the holotype of G. pygmaea, and the smallest of
them lacks an ectostylid, which has been observed on other
lower molars attributed to that species (Churcher 1978;
Harris 1991). Since these two Giraffa species are distin-
guished dentally primarily by size, the data suggest that the
lower molars from the Upper Laetolil Beds should all be
retained within G. stillei.
There is an M3 from Koobi Fora attributed to Giraffa
pygmaea (KNM-ER 3912) that is substantially larger than
the type specimen and is similar in its mesiodistal and buc-
colingual dimensions to typical Giraffa stillei M
3s from
Laetoli (Harris 1991). It seems that this specimen was
placed in G. pygmaea because the G. stillei teeth from
Koobi Fora are considerably larger than those from Laetoli,
and that this specimen is much smaller than any G. stillei
M3 from Koobi Fora. KNM-ER 3912 has an extra cusp join-
ing the posterior arm of the entoconid with the hypoconulid
that has been suggested to be potentially a characteristic of
G. pygmaea (Harris 1991). This appears to be an entostylid
and is present on one of the smaller M3s from Laetoli (LAET
75-563), but not the other (EP 1078/01). This cuspid is also
observed on the largest isolated G. stillei M3 from Laetoli,
EP 862/03, the M3 on the mandible from the Lower Laetolil
14 Giraf fidae
Beds, and the lectotype of G. stillei (MB Ma. 39078),
which suggests that it is a variable feature on smaller
Giraffa M3s.
Upper Ndolanya Beds
Giraffa aff. stillei is the most common taxon at all of the
Upper Ndolanya Beds localities other than Loc. 22S, from
which only five giraffid specimens were recovered. The
taxon makes up an even greater proportion of the total num-
ber of giraffids recovered from these strata than in the Upper
Laetolil Beds, as noted previously (Harris 1987), with there
being over three times as many G. aff. stillei specimens
recovered than the other two giraffids combined.
Giraffa aff. jumae
Given the additional large-bodied Giraffa postcranial fossils
recovered by EPPE, it seems probable that Giraffa jumae
was present in the Upper Ndolanya Beds, although without
cranial specimens to confirm its taxonomic identity, it is
possible that the material represents Giraffa camelopardalis.
It is also possible that some of the larger postcranial speci-
mens could be attributed to Sivatherium, particularly with
the documented reduction in the size of the Sivatherium
maurusium postcrania in the late Pliocene (Harris 1976b,
1991; Harris et al. 1988; Geraads 1996) and the lack of more
than a few fragmentary sivathere postcrania identified from
the Upper Ndolanya Beds. However, until further, more
complete specimens are recovered, it will be difficult to
resolve this issue. Therefore, because of their size and over-
all morphological similarities to Giraffa, the specimens are
provisionally retained in Giraffa aff. jumae.
Giraffa aff. stillei and Giraffa aff. pygmaea
If the same species is present in the Upper Laetolil Beds and
the Upper Ndolanya Beds, it appears that Giraffa stillei
increased in size through time at Laetoli. The dimensions of
all teeth that could be measured from the Upper Ndolanya
Beds are larger than the means of those from the Laetolil
Beds. In addition, the average sizes of the two postcranial
elements with relatively large samples from the Upper
Ndolanya Beds, astragali and proximal phalanges, are greater
than those of specimens from the Upper Laetolil Beds. This
seems to have been a general trend for G. stillei in East
Africa, with dental specimens from Koobi Fora and Omo
being larger than those from the Laetolil Beds, Kanapoi and
Lothagam (Harris 1976b, 1987, 1991, 2003; Harris et al.
2003). The difference was notable enough for the more
recent specimens from Koobi Fora and Omo to have been
initially placed into a different species, Giraffa gracilis
(Harris 1976b; Churcher 1978).
It is possible that the reason for this size increase is the
appearance of Giraffa pygmaea in the faunal assemblages of
East Africa during this time period, with interspecific com-
petition driving Giraffa stillei to increase in size. The first
documented appearance of G. pygmaea is at Hadar, but it
also co-occurs with G. stillei at Koobi Fora and in the Omo
Valley. The teeth of G. stillei from Koobi Fora, where the
same four giraffids from the Upper Ndolanya Beds co-
occurred throughout much of the sequence, are notably larger
than those of G. stillei from the Upper Laetolil Beds, where
G. pygmaea appears to be absent. Data on G. stillei speci-
mens from the Denen Dora and Kada Hadar Members at
Hadar and Member B in the Omo Valley, where the two
small Giraffa species also co-occurred (Boaz et al. 1982;
Reed 2008), would provide a test of this hypothesis.
Sivatherium aff. maurusium
Although Sivatherium hendeyi is indistinguishable from S.
maurusium in its dental morphology, the most recent diag-
nostic specimens of the former taxon are from the early
Pliocene at Langebaanweg (Harris 1976a). Thus, it seems
reasonable to provisionally attribute the Sivatherium dento-
gnathic specimens from the Upper Ndolanya Beds to
Sivatherium aff. maurusium.
A comparison of Sivatherium maurusium M1s and M2s
from Laetoli and specimens from Olduvai and East and
West Turkana found that those from the Upper Laetolil
Beds were smaller than those from the late Pliocene and
early Pleistocene sites (Fig. 14.10), which suggests that
Sivatherium aff. maurusium may also have undergone a size
increase through time. One might expect a change in dental
morphology as sivatheres shift from browsing to more graz-
ing near the end of the Pliocene (Harris 2003; Harris et. al.
2003), although at Laetoli there is no indication from stable
isotope analyses of any significant dietary shift in S. mauru-
sium between the two stratigraphic levels (Kingston and
Harrison 2007).
This hypothesis is supported by data from three lower
molars collected by Kohl-Larsen from “Gadjingero”, thought
to be equivalent to Loc. 18 in the Upper Ndolanya Beds,
which show them to be considerably larger than the Upper
Laetolil Beds specimens (T. Harrison, personal communica-
tion) and also by the partial mandible from the Upper
Ndolanya Beds, EP 1040/00, that is substantially deeper
than the LAET 75-520 specimen from the Upper Laetolil
Beds, and more similar to specimens from Koobi Fora
(Harris 1991; Likius 2002). Moreover, Sivatherium maurusium
360 C.A. Robinson
teeth from Hadar are said to be of similar size to, or slightly
smaller than, those from the Upper Laetolil Beds (Harris
1987). However, since M1s and M2s are combined in the
above analyses and only three of these teeth were recovered
from the Upper Laetolil Beds, it is possible that these three
teeth are all relatively small M1s. Moreover, a comparison of
two M3s from “Gadjingero” and two from the Upper Laetolil
Beds collected by Kohl-Larsen finds that the Laetolil Beds
specimens are considerably larger (T. Harrison, personal
communication). Finally, two isolated teeth from the Upper
Ndolanya Beds collected by Mary Leakey’s team are noted
to be approximately the same size as those from the Upper
Laetolil Beds (Harris 1987). These conflicting results are
consistent with Harris (1991) statement that there is no obvi-
ous trend in Sivatherium cheek tooth size in East Africa and
Likius (2002) contention that the extent of variation and
small sample sizes make it impossible to be confident of any
With no complete metapodials or limb bones attributed to
S. maurusium from the Upper Ndolanya Beds, it is not pos-
sible to determine whether the reduction in the length of these
elements at more recent sites, documented in specimens from
Ahl al Oughlam, Olduvai, West Turkana and the Upper Burgi
Member at Koobi Fora (Harris 1976b, 1991; Harris et al.
1988; Geraads 1996), also occurs at Laetoli. Limb length
reduction in S. maurusium is thought to be related to a shift
from browsing to grazing (Harris 2003; Harris et al. 2003), as
it would have brought sivatheres closer to ground level foods
(Harris 1991). Stable isotope analyses indicate that
Sivatherium maurusium retained its browsing lifestyle
between the Upper Laetolil and Upper Ndolanya Beds
(Kingston and Harrison 2007), which suggests that, if there is
a link between dietary behavior and limb bone length, its limb
elements may have been relatively elongated compared to
individuals from contemporary sites.
Diversity of Giraffids at Laetoli
The Upper Laetolil Beds are remarkable among Pliocene
sub-Saharan African localities in having such a high
proportion of giraffids among its mammalian faunal assem-
blage, with only Langebaanweg having a greater percentage,
due to an exceptional number of Sivatherium hendeyi speci-
mens (Harris 1976a, 1987, 1991). In the Upper Laetolil Beds,
giraffids make up approximately 15% of the ruminant artio-
dactyls recovered by EPPE. They are also relatively common
in the Lower Laetolil Beds, making up 12% of the ruminants.
In the Upper Ndolanya Beds the percentage of giraffids is
substantially reduced, to only 4% of ruminants. This change
partly reflects the extraordinary increase in the proportion of
bovids in the Upper Ndolanya Beds from 35% of the faunal
assemblage to 81% (Kingston and Harrison 2007).
Giraffids are relatively rare at late Miocene sites, such as
the Middle Awash and Upper and Lower Nawata Members at
Lothagam (Harris 2003; Haile-Selassie et al. 2004), making
up 6–8% of the ruminants in the Nawata Members (Harris
et al. 2003). By the early Pliocene at Kanapoi and the Apak
and Kaiyumung Members at Lothagam, they have become
more common, being over 10% of the ruminants at all of
these sites (Harris et al. 2003), although they are still rare at
other early Pliocene sites, such as Galili (Kullmer et al. 2008).
It has been suggested, though, that the fauna from the Apak
and Kaiyumung Members may not be representative due to
the limited numbers of mammalian fossils derived from these
30 31 32 33 34 35 36 37 38 39 40
mesio-distal length
bucco-lingual breadth
Koobi Fora
Laetolil Beds
Ndolanya Beds
West Turkana
Fig. 14.10 Bivariate plot of the mesiodistal length and buccolingual
breadth measurements (mm) of Sivatherium maurusium lower molars
(M1 and M2) from Koobi Fora, Laetoli, Olduvai and West Turkana. This
includes three specimens from “Gadjingero”, regarded as equivalent to
Loc. 18 in the Upper Ndolanya Beds
14 Giraf fidae
strata (Leakey and Harris 2003). After the early Pliocene,
giraffid numbers, but not taxonomic diversity, become reduced
at Hadar, Koobi Fora, West Turkana, and the Upper Ndolanya
Beds (Harris 1976b, 1991; Harris et al. 1988). In the Omo,
giraffids represent less than 10% of the ruminant artiodactyls
in all relatively fossiliferous members, other than Upper
Member B and Member C, dated to between about 2.5 and
3.1 Ma (Bobe et al. 2002), in the later part of the interval
between the Upper Laetolil Beds and Upper Ndolanya Beds.
It may be that the environmental conditions at Laetoli are
particularly well suited to a diverse and abundant giraffid
guild. Stable isotope analyses have found that the large her-
bivore fauna from Laetoli is unusual in having a large per-
centage of taxa adapted to a more generalized diet (Kingston
and Harrison 2007). This potentially could have enabled
giraffids to fill the specialized browser niche with little com-
petition, and could explain their diversity and relative
Although the most recent stable isotope analyses do not
provide evidence for a significant change in paleoenviron-
mental conditions between the Upper Laetolil Beds and the
Upper Ndolanya Beds (Kingston and Harrison 2007), previ-
ous analyses of the fossil mammal community structure
(Kovarovic et al. 2002) and the paleosol carbonates (Cerling
1992), suggest that there was a shift to a more savanna-like
environment. If so, it could be that the reduced number of
giraffids in the Upper Ndolanya Beds is associated with this
environmental change.
The giraffids from Laetoli are generally considered to be
obligate browsers (Harris and Cerling 1998; Harris et al.
2003; Leakey and Harris 2003; Kingston and Harrison 2007;
Solounias 2007). With three to four species surviving on
broadly similar foods at Laetoli, they must have occupied
slightly different niches. It may be that their diversity in size
allowed them to feed at different optimum heights (Harris
1976b, 1987; Kingston and Harrison 2007). For example, it
has been suggested that the prevalence of G. stillei is related
to there being a predominance of shorter trees at Laetoli
(Harris 1991). Alternatively, giraffids may have browsed on
different food items or in different microhabitats (Harris
1976b, 1991; Kingston and Harrison 2007). For example,
Harris (1991) suggests that Giraffa stillei may have fed in the
upper slopes or in riverine woodland, while G. jumae ranged
primarily in taller mid-slope woodland and G. pygmaea in
the valley-bottom korongo thickets. Evidence from carbon
isotopes from Laetoli indicate that Sivatherium maurusium
was feeding in more forested areas than Giraffa species
(Kingston and Harrison 2007). This is supported by the
robust tooth roots and massive crowns of Sivatherium which
may indicate that it was subsisting on more obdurate foods.
Acknowledgements I would like thank Dr. Terry Harrison for his
invitation to work on the giraffid fossils and contribute to this volume.
I would also like to thank him for data he gathered on the giraffid
specimens in Berlin and for many years of advice and guidance. I am
grateful for the discussions with Dr. Alan Gentry during the examina-
tion of this material and e-mail correspondence afterwards that pro-
vided important insights into artiodactyl morphology and evolution.
Comments from Denis Geraads and two anonymous reviewers substan-
tially improved this manuscript. Many thanks to the curatorial staffs at
the Kenya National Museum, the National Museum of Tanzania and the
American Museum of Natural History for providing access to the giraf-
fid specimens under their care. Fiona Bohane, Obed Gonzalez and
Tyiece Rose assisted in collecting data on extant giraffes. Travel funds
were provided by an NSF grant (BCS–0309513) to Terry Harrison.
Arambourg, C. (1959). Vertébrés continentaux du Miocène supérieur
de l’Afrique du Nord. Paléontologie Mémoires, 4, 1–161.
Boaz, N. T., Howell, F. C., & McCrossin, M. L. (1982). Faunal age of
the Usno, Shungura B and Hadar Formations, Ethiopia. Nature, 300,
Bobe, R., Behrensmeyer, A. K., & Chapman, R. E. (2002). Faunal
change, environmental variability and late Pliocene hominin evolu-
tion. Journal of Human Evolution, 42, 475–497.
Cerling, T. E. (1992). Development of grasslands and savannas in East
Africa during the Neogene. Palaeogeography, Palaeoclimatology,
Palaeoecology, 97, 241–247.
Churcher, C. S. (1978). Giraffidae. In V. J. Maglio & H. B. S. Cooke
(Eds.), Evolution of African Mammals (pp. 509–535). Princeton:
Princeton University Press.
Churcher, C. S. (1979). The large palaeotragine giraffid, Palaeotragus ger-
maini from Late Miocene deposits of Lothagam Hill, Kenya. Brevoria,
453, 1–8.
Churcher, C. S. (1990). Cranial appendages of Giraffoidea. In G. A.
Bubenik & A. B. Bubenik (Eds.), Horns, pronghorns, and antlers:
Evolution, morphology, physiology, and social significance (pp.
180–196). New York: Springer.
Deino, A. (2011). 40Ar/39Ar dating of Laetoli, Tanzania. In T. Harrison,
T. (Ed.), Paleontology and geology of laetoli: Human evolution in
context (Geology, geochronology, paleoecologys, and paleoenvi-
ronment, Vol. 1, pp. 77–97). Dordrecht: Springer.
Dietrich, W. O. (1942). Altestquartäre Säugetiere aus den südlichen
Serengeti, Deutsch Ostafrika. Palaeontographica, 94, 43–133.
Drake, R., & Curtis, G. H. (1987). K–Ar geochronology of the Laetoli
fossil localities. In M. D. Leakey & J. M. Harris (Eds.), Laetoli:
A Pliocene site in northern Tanzania (pp. 48–52). Oxford:
Franz-Odendaal, T. A., Lee-Thorp, J. A., & Chinsamy, A. (2003).
Insights from stable light isotopes on enamel defects and weaning in
Pliocene herbivores. Journal of Bioscience, 28, 765–773.
Gentry, A. W. (1997). Fossil ruminants (Mammalia) from the Manonga
Valley, Tanzania. In T. Harrison (Ed.), Neogene paleontology of the
Manonga Valley, Tanzania. A window into the evolutionary history
of East Africa (Topics in Geobiology, Vol. 14, pp. 107–137).
New York: Plenum.
Geraads, D. (1985). Sivatherium maurusium (Pomel) (Giraffidae,
Mammalia) du Pléistocène de la République de Djibouti.
Paläontologische Zeitschrift, 59, 311–321.
Geraads, D. (1986). Remarques sur la systématique et la phylogénie des
Giraffidae (Artiodactyla, Mammalia). Geobios, 19, 465–477.
Geraads, D. (1991). Derived features of giraffid ossicones. Journal of
Mammalogy, 72, 213–214.
Geraads, D. (1994). Girafes fossils d’Ouganda. In M. Pickford, &
B. Senut, (Eds.), Geology and palaeobiology of the Albertine Rift
Valley, Uganda-Zaire. Volume II: Palaeobiology (pp. 375–381).
Orléans: CIFEG Occasional Publication 29.
362 C.A. Robinson
Geraads, D. (1996). Le Sivatherium (Giraffidae, Mammalia) du Pliocène
final d’Ahl al Oughlam (Casablanca, Maroc), et l’évolution du genre
en Afrique. Paläontologische Zeitschrift, 70, 623–629.
Geraads, D., Alemseged, Z., Reed, D., Wynn, J., & Roman, D. C.
(2004). The Pleistocene fauna (other than Primates) from Asbole,
Lower Awash Valley, Ethiopia, and its environmental and biochrono-
logical implications. Geobios, 37, 697–718.
Haile-Selassie, Y., WoldeGabriel, G., White, T. D., Bernor, R. L.,
DeGusta, D., Renne, P. R., Hart, W. K., Vrba, E., Ambrose, S., &
Howell, F. C. (2004). Mio-Pliocene mammals from the Middle
Awash, Ethiopia. Geobios, 37, 536–552.
Hamilton, W. R. (1978). Fossil giraffes from the Miocene of Africa and a
revision of the phylogeny of the Giraffoidea. Philosophical
Transactions of the Royal Society of London. Series B, 283, 165–229.
Harris, J. M. (1976a). Pliocene Giraffoidea (Mammalia, Artiodactyla)
from the Cape Province. Annals of the South African Museum, 69,
Harris, J. M. (1976b). Pliocene Giraffoidea (Mammalia, Artiodactyla)
from East Rudolf. In R. J. G. Savage & S. C. Coryndon (Eds.),
Fossil vertebrates of Africa (Vol. 4, pp. 283–332). London:
Harris, J. M. (1976c). Giraffidae from the East Rudolf succession. In Y.
Coppens, F. C. Howell, G. L. Isaac, & R. E. F. Leakey (Eds.),
Earliest man and environments in the Lake Rudolf Basin (pp. 264–
267). Chicago: University of Chicago Press.
Harris, J. M. (1987). Fossil Giraffidae and Camelidae from Laetoli. In
M. D. Leakey & J. M. Harris (Eds.), Laetoli: A Pliocene site in
northern Tanzania (pp. 358–378). Oxford: Clarendon.
Harris, J. M. (1991). Family Giraffidae. In J. M. Harris (Ed.), Koobi Fora
Research Project (Vol. 3, pp. 93–138). Oxford: Clarendon.
Harris, J. M. (2003). Lothagam giraffids. In M. G. Leakey & J. M.
Harris (Eds.), Lothagam: The dawn of humanity in eastern Africa
(pp. 523–530). New York: Columbia University Press.
Harris, J. M., & Cerling, T. E. (1998). Isotopic changes in the diet
of giraffids. Journal of Vertebrate Paleontology Supplement, 18, 49A.
Harris, J. M., Brown, F. H., & Leakey, M. G. (1988). Stratigraphy and
paleontology of Plio-Pleistocene localities west of Lake Turkana,
Kenya. Natural History Museum of Los Angeles County,
Contributions in Science, 399, 1–128.
Harris, J. M., Leakey, M. G., Cerling, T. E., & Winkler, A. J. (2003).
Early Pliocene tetrapod remains from Kanapoi, Lake Turkana Basin,
Kenya. In J. M. Harris & M. G. Leakey (Eds.), Geology and
Vertebrate Paleolotology of the early Pliocene site of Kanapoi,
northern Kenya (pp. 39–115). Natural History Museum of Los
Angeles County, Contributions in Science, no. 498.
Hendey, Q. B. (1981). Palaeoecology of the late Tertiary fossil occur-
rences in ‘E’ Quarry, Langebaanweg, South Africa, and a reinter-
pretation of their geological context. Annals of the South African
Museum, 84, 1–104.
Kingston, J. D., & Harrison, T. (2007). Isotopic dietary reconstructions
of Pliocene herbivores at Laetoli: Implications for early hominin
paleoecology. Palaeogeography, Palaeoclimatology, Palaeoecology,
243, 272–306.
Kovarovic, K., Andrews, P., & Aiello, L. (2002). The palaeoecology of
the Upper Ndolanya Beds at Laetoli, Tanzania. Journal of Human
Evolution, 43, 395–418.
Kullmer, O., Sandrock, O., Viola, T. B., Hujer, W., Said, H., & Seidler,
H. (2008). Suids, elephantoids, paleochronology, and paleoecology
of the Pliocene hominid site Galili, Somali region, Ethiopia. Palaios,
23, 452–464.
Leakey, L. S. B. (1965). Olduvai 1951–1961: Fauna and background.
Cambridge: Cambridge University Press.
Leakey, L. S. B. (1970). Additional information on the status of Giraffa
jumae from East Africa. In L. S. B. Leakey & R. J. G. Savage (Eds.),
Fossil vertebrates of Africa (Vol. 2, pp. 325–330). London: Academic.
Leakey, M. G., & Harris, J. M. (2003). Lothagam: Its significance and
contributions. In M. G. Leakey & J. M. Harris (Eds.), Lothagam:
The dawn of humanity in eastern Africa (pp. 625–660). New York:
Columbia University Press.
Likius, A. (2002). Les grands Ongulés du Mio-Pliocène du Tchad
(Rhinocerotidae, Giraffidae, Camelidae): Systématique, implications
paléogéographiques et paléoenvironnementales. Doctoral thesis,
Université Poitiers, France.
Linnaeus, C. (1758). Systema Naturae per regna tria naturae, secun-
dum classes, ordines, genera, species, cum characteribus, differen-
tiis, synonymis, locis, Tomus I. Editio decima, reformata. Laurentii
Salvii: Holmiae.
Lockwood, C. A., Richmond, B. G., & Jungers, W. L. (1996). Randomization
procedures and sexual dimorphism in Australopithecus afarensis.
Journal of Human Evolution, 31, 537–548.
McDougall, I., & Brown, F. H. (2006). Precise 40Ar/39Ar geochronology
for the Upper Koobi Fora Formation, Turkana Basin, northern
Kenya. Journal of the Geological Society, 163, 205–220.
Pomel, A. (1892). Sur le Libytherium maurusium, grand ruminant du
terrain Pliocène de l’Algérie. Comptes Rendus Hebdomadaires des
Séances de l’Académie des Sciences, 115, 100–102.
Reed, K. E. (2008). Paleoecological patterns at the Hadar hominin site,
Afar Regional State, Ethiopia. Journal of Human Evolution, 54,
Richmond, B. G., & Jungers, W. L. (1995). Size variation and sexual
dimorphism in Australopithecus afarensis and living hominoids.
Journal of Human Evolution, 29, 229–245.
Singer, R., & Boné, E. L. (1960). Modern giraffes and the fossil giraf-
fids of Africa. Annals of the South African Museum, 45, 375–603.
Solounias, N. (1988). Prevalence of ossicones in Giraffidae
(Artiodactyla, Mammalia). Journal of Mammalogy, 69, 845–848.
Solounias, N. (2007). Family Giraffidae. In D. R. Prothero & S. E. Foss
(Eds.) The evolution of artiodactyls (pp. 257–277). Baltimore: The
John Hopkins University Press.
Su, D. F., & Harrison, T. (2007). The paleoecology of the Upper Laetolil
Beds at Laetoli. A reconsideration of the large mammal evidence. In R.
Bobe, Z. Alemseged, & A. K. Behrensmeyer (Eds.), Hominin
Environments in the East African Pliocene: An assessment of the
faunal evidence (pp. 279–313). Dordrecht: Springer.
Uchida, A. (1996). Craniodental variation among the great apes.
Peabody Museum Bulletin, 4, 1–176.
Vignaud, P., Duringer, P., Mackaye, H. T., Likius, A., Blondel, C.,
Boisserie, J.-R., de Bonis, L., Eisenmann, V., Etienne, M.-E.,
Geraads, D., Guy, F., Lehmann, T., Lihoreau, F., Lopez-Martinez,
N., Mourer-Chauviré, C., Otero, O., Rage, J.-C., Schuster, M.,
Viriot, L., Zazzo, A., & Brunet, M. (2002). Geology and palaeontol-
ogy of the Upper Miocene Toros-Menalla hominid locality, Chad.
Nature, 418, 152–155.
Ward, C. V., Leakey, M. G., Brown, B., Brown, F., Harris, J., & Walker,
A. (1999). South Turkwel: A new Pliocene hominid site in Kenya.
Journal of Human Evolution, 36, 69–95.
WoldeGabriel, G., White, T. D., Suwa, G., Renne, P., de Heinzelin, J.,
Hart, W. K., & Helken, G. (1994). Ecological and temporal place-
ment of early Pliocene hominids at Aramis, Ethiopia. Nature, 371,
... The measurements used to compare the material were collected during visits to the American Museum of Natural History (New York, U.S.A.), the Institut Català de Paleontologia-Miquel Crusafont (Barcelona, Spain), the Museo Nacional de Ciencias Naturales-CSIC (Madrid, Spain), the Musèum national d'Histoire naturelle (Paris, France), and the Natural History Museum (London, U.K.). The rest of the morphological and biometric information in this paper was collected from several publications (Bohlin, 1926;Churcher, 1970Churcher, , 1978Hamilton, 1973Hamilton, , 1978Harris, 1976aHarris, , 1976bHarris, , 1991Pilgrim, 1911;Matthew, 1929;Colbert, 1935;Singer and Boné, 1960;Robinson, 2011;Bhatti et al., 2012;Khan et al., 2014). Occurrence-Nagri (Nagri Formation, Pakistan); Dhok Pathan, Hasnot, Kauliar, Padhri (Dhok Pathan Formation, Pakistan); Perim Island (India). ...
Here, we report new dental remains of a large giraffid from the Late Miocene of Pakistan. The new material improves our knowledge of the dental morphology of this extinct sivatherine giraffid, as it sheds light on intra-clade morphological variability within Late Miocene sivatherine giraffids. Fossils were recovered from sediments within the Nagri and Dhok Pathan formations (Middle Siwaliks, Late Miocene) from the sites Nagri, Dhok Pathan, Kauliar and Padhri in Pakistan. Presence of these fossils in Late Miocene sediments in Pakistan demonstrates the stratigraphic range of Bramatherium grande, which was present in the Indo-Pakistani region and the Greco-Iranian region from 11.2 to ca. 3.4 Ma and 11.2 to 9.0 Ma, respectively. The new fossils provide additional data that help resolve the phylogenetic relationships of these large Late Miocene sivatheres. The new findings highlight the presence of Bramatherium grande in the Indo-Pakistani region during the Late Miocene and provide evidence for the synonymy of Bramatherium grande and Bramatherium magnum, with Bramatherium grande having priority.
... What is striking, and remarkable, about giraffe taxonomic relationships is that despite the differences of opinion concerning the number of species, the data concur in describing two distinctive clusters, and possibly three. During the Middle/Late Pleistocene Era, substantial geographic and atmospheric changes transformed the African continent and produced adaptive radiations among multiple mammalian lineages, including giraffes (Badlangana, Adams, & Manger, 2009;Lorenzen, Heller, & Siegismund, 2012;Robinson, 2011). Many African ungulate species underwent a north/ south division closely marked by the Zambezi River (Lorenzen et al., 2012). ...
... jumae Leakey 1965 Description EF-HR T2-L2-1410 is a lower p2 (AP Â T: 17.7 Â 12.1 mm) from Upper Bed II that falls within the size range of both extant G. camelopardalis and a specimen of G. jumae from Rawe (Harris, 1976a). It is larger than Upper Laetoli specimens referred by Robinson (2011) to G. stillei, but similar to specimens he referred to Giraffa aff. jumae. ...
Eight years of excavation work by the Olduvai Geochronology and Archaeology Project (OGAP) has produced a rich vertebrate fauna from several sites within Bed II, Olduvai Gorge, Tanzania. Study of these as well as recently re-organized collections from Mary Leakey's 1972 HWK EE excavations here provides a synthetic view of the faunal community of Olduvai during Middle Bed II at ∼1.7-1.4 Ma, an interval that captures the local transition from Oldowan to Acheulean technology. We expand the faunal list for this interval, name a new bovid species, clarify the evolution of several mammalian lineages, and record new local first and last appearances. Compositions of the fish and large mammal assemblages support previous indications for the dominance of open and seasonal grassland habitats at the margins of an alkaline lake. Fish diversity is low and dominated by cichlids, which indicates strongly saline conditions. The taphonomy of the fish assemblages supports reconstructions of fluctuating lake levels with mass die-offs in evaporating pools. The mammals are dominated by grazing bovids and equids. Habitats remained consistently dry and open throughout the entire Bed II sequence, with no major turnover or paleoecological changes taking place. Rather, wooded and wet habitats had already given way to drier and more open habitats by the top of Bed I, at 1.85-1.80 Ma. This ecological change is close to the age of the Oldowan-Acheulean transition in Kenya and Ethiopia, but precedes the local transition in Middle Bed II. The Middle Bed II large mammal community is much richer in species and includes a much larger number of large-bodied species (>300 kg) than the modern Serengeti. This reflects the severity of Pleistocene extinctions on African large mammals, with the loss of large species fitting a pattern typical of defaunation or 'downsizing' by human disturbance. However, trophic network (food web) analyses show that the Middle Bed II community was robust, and comparisons with the Serengeti community indicate that the fundamental structure of food webs remained intact despite Pleistocene extinctions. The presence of a generalized meat-eating hominin in the Middle Bed II community would have increased competition among carnivores and vulnerability among herbivores, but the high generality and interconnectedness of the Middle Bed II food web suggests this community was buffered against extinctions caused by trophic interactions.
This study deals with the issues around the synonymization of the large extinct giraffid genera Libytherium and Sivatherium. We performed a morphological and biometrical analysis of the cranial remains of these giraffid genera and resolved this question by formulating criteria to distinguish Libytherium from Sivatherium, and to justify their systematic position within the Giraffidae. The present study also reports a new species of the genus Libytherium, Libytherium proton nov. sp. from the Chinji Formation (late Middle Miocene) and the Dhok Pathan Formation (Late Miocene) of the Siwalik Group of the Indian subcontinent. This extends the biogeographic and biostratigraphic range of Libytherium because this taxon had never been reported from the Siwaliks, Asia and the Miocene before this study. This study also initiates a detailed reassessment of the African and Siwalik material currently attributed to genus Sivatherium (and some other genera) and their proper allocation within the Giraffidae.
The sivatherine clade includes some of the largest giraffids and emerged during the late Miocene. Sivatherium hendeyi, the earliest known species of the Sivatherium genus, was first described from the lower Pliocene of Langebaanweg (5.15 ± 0.1 Ma, Cape Province, South Africa). Here we describe the first possible occurrence of Sivatherium from western Europe from the lower Pliocene (MN14) of Puerto de la Cadena (4.9 Ma, Murcia, Spain). The new material consists of dental and postcranial remains. The Puerto de la Cadena Sivatherium, together with the presence of Macaca sp. and Debruijnimys sp., indicates a connection between African and European faunas during the early Pliocene and a possible relationship between Sivatherium and the stem Iberian sivatherines Decennatherium and Birgerbohlinia.
Full-text available
The Hadar giraffids belong to the genera Giraffa and Sivatherium. We recognize two species in the former genus; distinguishing between them on dental elements is straightforward, but it is more difficult on other remains. Giraffa jumae is about the size of the modern G. camelopardalis, but has a less pneumatized skull with more conical ossicones, more slender and often longer limb bones, and also differs in some dental features, making it an unlikely ancestor of the modern giraffe. Giraffa stillei has teeth that are always smaller than those of the modern form, less molarized premolars, ossicones smaller but otherwise similar to those of the larger Hadar species, and slender limb bones that are probably relatively long, thus not much shorter than those of the modern form. It might be the ancestor of the later G. gracilis from the Turkana Basin, and of G. camelopardalis. We assign no specimen to Giraffa pygmaea. Sivatherium maurusium, a well-known form of the African Pliocene and Pleistocene, is a rare form at Hadar. In contrast to most bovids, giraffids are more common in the Sidi Hakoma than in the overlying Denen Dora Member, perhaps as a result of grassland expansion at Hadar at that time.
Full-text available
The Gondolin palaeokarstic system, located in the UNESCO Fossil Hominids of South Africa World Heritage Site, has been sporadically excavated since the 1970s. Sampling of ex situ dumpsites in 1997 recovered the only two fossil hominin specimens recovered thus far from Gondolin. While one partial mandibular molar (GA 1) remains unattributed, the complete mandibular second molar (GA 2) represents the largest Paranthropus robustus Broom, 1938 tooth identified to date. While subsequent excavations and research at Gondolin has clarified the geological, temporal, taphonomic, and palaeoecologic context for the in situ deposits, this paper presents the first comprehensive description of the fossil assemblage ‘associated’ with the two ex situ hominins. Analysis of 42 calcified sediment blocks and naturally decalcified sediments excavated from three cubic metres of the Dump A deposits reinforce that the dump contains a heterogeneous aggregation of materials from across the Gondolin sedimentary deposits. A total of 15,250 individual fossil specimens were processed (via sifting or acetic-acid mediated processing of calcified sediment blocks), yielding a faunal record that largely mirrors that described from either (or both) the GD 1 and GD 2 in situ assemblages but includes representatives of four novel mammal groups (Families Cercopithecidae, Felidae, Herpestidae, Giraffidae) not recorded in either in situ sample. While basic assemblage characteristics including primary taphonomic data is presented, analysis and interpretation is limited by the ex situ origin of the sample. Ultimately, these results reinforce that the substantial mining-mediated obliteration of palaeokarstic deposits at Gondolin continue to obscure a clear association between the Gondolin Dump A hominins and any of the sampled and dated in situ deposits.
Full-text available
Giraffids include the only living giraffomorph ruminants and are diagnosed by the presence of bi-lobed canines and a special type of epiphyseal cranial appendages called ossicones. The family Giraffidae ranges from the latest early Miocene until today. However they are currently extant relics with only two living representatives, the African genera Okapia and Giraffa. Giraffids were much more diverse and widespread in the past, with more than 30 fossil species described. For the past decades a number of studies intended to resolve the phylogenetic relationships of the family, but due to the lack of really good cranial material no clear consensus was reached regarding the phylogenetic relationships amongst the different members of the group. The exceptionally complete remains of a new large giraffid from the late Miocene of Spain, Decennatherium rex sp. nov., allows us to improve and reassess giraffid systematics, offering a lot of new data, both anatomic and phylogenetic, on the large late Miocene giraffids of Eurasia. The results of our cladistic analysis show Decennatherium as a basal offshoot of a clade containing the gigantic samotheres and sivatheres, characterized by the presence of a Sivatherium-like ossicone-plan among other features. Decennatherium thus offers the most ancient evidence of this Sivatherium-plan and firmly establishes the early morphological patterns of evolution of a sivathere / samothere-clade that is defined as the less inclusive clade that contains Decennatherium and Sivatherium. Finally, this large group of four-ossiconed giraffids evolutionarily links Miocene Europe and Africa indicating vicariance / migration processes among the giraffid genetic pools separated by the Mediterranean Sea.
Evidence from the Pliocene hominin site of Laetoli in northern Tanzania demonstrates that there was a taxonomic turnover of the mammalian fauna between the Upper Laetolil Beds (3.6–3.85 Ma) and the Upper Ndolanya Beds (2.66 Ma). Paranthropus aethiopicus was one of the novel species that appeared locally as part of the restructured fauna. This turnover coincides with a major climatic shift at ~2.8–2.5 Ma, which had an important impact on the local environment and the composition of the faunal community. Investigation of the paleoecology of the Upper Ndolanya Beds provides critical evidence about how the vegetation and fauna at Laetoli, including the hominins, responded to these environmental changes. The preponderance of alcelaphin bovids and the reduced frequency of browsing ungulates, in conjunction with evidence from ecomorphology, mesowear and stable isotopes, indicate that the Upper Ndolanya Beds sample drier habitats with a greater proportion of grasslands compared with the earlier Upper Laetolil Beds. However, paleoecological inferences based on ostrich eggshells, rodents, and terrestrial gastropods present a more complicated picture, indicating instead that Upper Ndolanya habitats were more mesic and dominated by dense woodlands. Such confounding results can be reconciled as a consequence of the differential impact of climatic and environmental change on a global, regional and local scale.
Full-text available
Decennatherium pachecoi Crusafont, 1952, is one of two giraffid species described from the Miocene of the Iberian Peninsula. This species is recovered exclusively from Vallesian faunas (MN9–10, late Miocene, 10–11 Ma). Despite being relatively well represented in the fossil record, except for the skull and ossicones, the complete vertebral column, and part of the upper dentition, its systematics and phylogenetic position among giraffids are the subject of debate. We update our knowledge of D. pachecoi, revising all Spanish material assigned to this species, as well as previously undescribed fossils. We reassess the systematics of Decennatherium, including its potential relationship with the second Iberian giraffid, the early Turolian Birgerbohlinia Crusafont, 1952, by means of the first cladistic analysis of the Giraffidae that includes Decennatherium together with the most relevant African and Eurasian taxa, both fossil and extant. Our results link Decennatherium with a ‘samothere’ clade, whereas Birgerbohlinia is nested within a ‘sivathere’ clade, thus refuting a previously assumed direct relationship between the two Spanish forms. Finally, we discuss some other possible finds of the genus Decennatherium in Greece, Turkey, and Iran. SUPPLEMENTAL DATA—Supplemental materials are available for this article for free at Citation for this article: Rios, M., I. M. Sánchez, and J. Morales. 2016. Comparative anatomy, phylogeny, and systematics of the Miocene giraffid Decennatherium pachecoi Crusafont, 1952 (Mammalia, Ruminantia, Pecora): State of the art. Journal of Vertebrate Paleontology. DOI: 10.1080/02724634.2016.1187624. 2016
Olduvai is one of the most important prehistoric sites in the world; indeed, the only Middle Pleistocene site of comparable importance is Choukoutien and Olduvai can show deposits far older. The site has produced a mass of material of the highest archaeological and palaeontology importance and in this first of five volumes Dr Leakey and his collaborators make their preliminary reports. The story of the excavations initiated by Dr Leakey in 1951 is well known. Their purpose was to locate and uncover a series of living-floors of early Hand-axe man and, if possible, of the preceding Olduwan culture. The discoveries were of striking and far-reaching importance. They included, besides a mass of tools and artefacts, small animal and human remains and the famous skull of Zinjanthropus boisei, the earliest tool-making man. Against this background Leakey and his collaborators discuss the geological evidence, its relation to the fauna and other fossil evidence, the problems of climatic sequence and the use of potassium-argon dating. The purpose of this volume is to provide a context in which the fossil human remains and the Stone Age cultural sequence at Olduvai can be studied.
The development of savanna-type grasslands is a relatively recent phenomena in East Africa. The stable carbon isotopic composition of paleosol carbonates from fossil localities in East Africa show that C 4 vegetation was present by about 8-9 Ma but made up only a relatively small proportion of the total biomass. Although the proportion of C 4 vegetation increased in the Pliocene and Pleistocene there is no evidence for the development of virtually pure C 4 grasslands, as is characterized by tropical grasslands today, until Middle Pleistocene times. This has important implications concerning the evolution of mammals in Africa, including hominids.