Current Trends in Dental Morphology Research

PREMOLAR ROOT NUMBER VARIATION IN

HOMINOIDS: GENETIC POLYMORPHISM VS.
FUNCTIONAL SIGNIFICANCE

KORNELIUS KUPCZIK1,2*, FRED SPOOR2, ANDREAS POMMERT3,

M. CHRISTOPHER DEAN2

1
Department of Biology and Hull York Medical School, University of York, York YO10 5DD, UK;
2
Department of Anatomy & Developmental Biology, UCL, London, WC1E 6BT, UK; 3Institute of
Medical Informatics, University Hospital Eppendorf, 20246 Hamburg, Germany

ABSTRACT

Premolars are the tooth group with the largest variation in tooth root number and form among
extant and extinct primates. Yet it remains unclear whether the range in premolar root number
from one to three simply reflects genetic polymorphism or whether variation in root number is an
adaptation to dietary specialisation. It has been shown that differences in occlusal loading in
primates are strongly linked to differences in root attachment area. Teeth that are subjected to
relatively larger masticatory stress should hence bear larger roots. Here we address the question
whether there is empirical evidence to support the hypothesis that an increase in premolar root
number is a mechanism to increase root attachment area. Then, we describe the relationship
between occlusal crown area and root surface area and ask to what extent crown area and
concomitant root expansion and root length co-vary and/or combine to influence root surface area
in premolars with different root numbers.
High-resolution CT-scans were made of the mandibular and maxillary dentitions of four Pan
troglodytes, two Gorilla gorilla, two Pongo pygmaeus and two Homo sapiens skulls together with
40 extracted human premolars. Three-dimensional imaging software was used to visualise the
teeth and quantify root surface area, occlusal crown area and root length in order to establish any
relationship between these parameters in single and multi-rooted premolars.
Root attachment area is found to be not significantly different between maxillary and
mandibular premolar pairs. Irrespective of root number, root surface area increases as crown area
increases. When root surface area was scaled against root length, the slopes of the RMA

*
Correspondence to: Dr Kornelius Kupczik, Hull York Medical School, University of York, York
YO10 5DD, United Kingdom; e-mail: [email protected]
258 Kornelius Kupczik et al.

regressions indicate that a single-rooted premolar gains less root surface area than a multi-rooted
premolar of the same length. Furthermore, single-rooted premolars gain root attachment area either
from an enlarged root girth, a consequence of an increase in crown size, or from an independent
lengthening of the root. On the other hand, multi-rooted premolars achieve increased attachment
area by a combined increase in girth and root length where both these factors are associated with
enlarged crown area.
These findings show that an increase in premolar root number is not necessarily related to an
increase in root surface area and that root number alone does not then reflect a functional
adaptation related to area of root attachment. We suggest that premolar root number variation
among hominoids may primarily constitute a genetic polymorphism which may be a good
reflection of phylogeny. However, multiple roots do have a clear functional advantage when root
length is constrained: for a given increase in root length they gain substantially more in attachment
area than single-rooted premolars. Moreover, multiple roots can spread in different directions and
better sustain occlusal loads during different phases of the chewing cycle.

INTRODUCTION

Premolar teeth in both extant and extinct primates are more variable in their
root morphology than any other the tooth type (Bennejeant, 1936; Senyürek,
1953; Goh, 1957; Remane, 1960; Turner, 1981; Abbott, 1984; Wood et al.,
1988; Tobias, 1995). A number of studies of fossil hominins have demonstrated
that both mandibular and maxillary premolar root form and number emerge as
useful taxonomic indicators (Ward et al., 1982; Abbott, 1984; Wood and
Engleman, 1988; Wood et al., 1988; Tobias, 1995; Bermúdez de Castro et al.,
1999; White et al., 2000; Haile-Selassie, 2001; Brunet et al., 2002; Leakey et al.,
2001). Several premolar root configurations have been identified, either on the
basis of external morphology (Abbott, 1984; Wood and Engleman, 1988; Wood
et al., 1988) or number of root canals alone (Tobias, 1995). The plesiomorphic
condition, where mandibular premolars bear two roots and maxillary premolars
three, is found in great apes, while modern humans generally show an
apomorphic condition of single rooted premolars, but more often than not with a
double-rooted or bifurcated P3. Variation in premolar root number more than
likely represents a genetic polymorphism (Tobias, 1995), but might also be an
adaptation to food processing strategy and diet. For instance, it has been
demonstrated that differences in dietary specialisations among primates are
strongly linked to differences in root attachment area (Kupczik, 2003; Spencer,
2003). Teeth with greater root attachment areas seem likely to sustain higher
occlusal forces than teeth with smaller attachment areas. This increase in area
can potentially be achieved by either lengthening and widening the roots or by
increasing root number from one to two or from two to three. It has also been
suggested that there is a correlation between increased root number and
enlargement of the crown. Support for this comes from observations among
 Premolar Root Number Variation in Hominoids 259

robust australopiths (Sperber, 1974; Tobias, 1995) and other fossil hominins
such as Australopithecus afararensis and Ardipithecus ramidus where premolar
root form has even been related to specific features in cuspal morphology (White
et al., 2000; Haile-Selassie, 2001). Kovacs (1971) also linked crown with root
morphology and noted that modern human teeth with supernumerary roots often
have a correspondingly enlarged perimeter at the cervix, an enhanced number of
cusps and a larger crown overall.
In this study we present new data for premolar roots in great apes and
humans. First, we address the question whether there is empirical evidence to
support the hypothesis that an increase in premolar root number is a mechanism
to increase root attachment area. Then, we describe the relationship between
occlusal crown area and root surface area and ask to what extent crown area and
concomitant root expansion and root length co-vary and/or combine to influence
root surface area in premolars with different root numbers.

MATERIAL AND METHODS

Transverse computed tomography (CT) scans of skulls were made to include

the complete mandibular and maxillary dentitions of four Pan troglodytes (two
males, two females), two Gorilla gorilla (one male, one female), two Pongo
pygmaeus (two males) and two Homo sapiens (males). In addition, 40 extracted
human premolars (P3/P3 and P4/P4) of unknown sex were scanned. The skulls are
housed at The Royal College of Surgeons of England, London and the Grant
Museum of Zoology and the Department of Anatomy & Developmental Biology
of University College London. The human premolars were made available by
the Oral Biology Department at The University of Newcastle-upon-Tyne. The
skulls and isolated teeth were scanned on a Siemens Somatom Plus 4 using the
following parameters: 140 kV, 188 mAs, 1.0 mm slice thickness with a 0.5 mm
slice increment and a neutral SP90 convolution filter. Following the data
acquisition and a subsequent data processing of the CT datasets, the three
dimensional (3D) visualisation and metric analyses of the dental roots was
carried out with VOXEL-MAN, a high-resolution volume visualisation system
originally developed for medical applications (Höhne et al., 1995; Pommert et
al., 2001). The teeth were reconstructed using a threshold based segmentation
approach (see Kupczik, 2003 for details). This allowed for the analysis of the
gross morphology of the premolars and the quantification of the root surface
area, the root length and the occlusal crown area. Root length was measured as
the distance between the plane at the cemento-enamel junction and the root tip
along the long axis of the root (cf. Abbott, 1984). Occlusal crown area is the
260 Kornelius Kupczik et al.

product of mesio-distal and bucco-lingual diameters and not only broadly

reflects occlusal loads, but is also closely correlated with the girth of the root at
the cervix (Hillson et al., 2005). The distinction between two- and three-rooted
premolars was made following Abbott (1984). The non-parametric Wilcoxon
signed rank test was applied to test for significant differences between maxillary
and mandibular premolar tooth root area in the ten specimens with complete
dentitions (i.e. P3s versus P3s and P4s versus P4s, respectively). Moreover,
Spearman's rank correlation coefficients (rs) and reduced major axis (RMA)
regression were used to establish the size relationship between the parameters.
The significance level of rs coefficients were determined using t-tests, while the
regression slopes were compared using F-tests. Since only the geometric
relationships of premolar roots in general irrespective of species differences
were of interest here, irrespective of species differences, single and multi-rooted
premolars (mandibular and maxillary) of all specimens were pooled. A
significance level of p = 0.05 was used to reject the null hypothesis for each
case.

RESULTS

Figure 1 illustrates the maxillary and mandibular dentition of a male P.

troglodytes highlighting its premolar root morphology. All great apes except
those of three P. troglodytes and the female G. gorilla exhibit two roots (P3

Figure 1. Lingual view of the dentition of a male Pan troglodytes (left) with close-up of the
mandibular and maxillary premolars (left)
 Premolar Root Number Variation in Hominoids 261

= one mesio-buccal, one distal blade-like root; P4 = one mesial, one distal blade)
in the mandibular and three roots (P3/4 = two buccal, one palatal root) in the
maxillary premolars (Table 1). The two female P. troglodytes possess P3s with a
single root and fused double roots, respectively. One of these specimens and the
female G. gorilla bear only two roots on the maxillary premolars, whilst one
male P. troglodytes has a double-rooted P4. The human premolars have single
roots with the exception of two P3s with two roots (Table 1). Despite unequal
root number of P3 and P3 and of P4 and P4 differences in root surface area for
those tooth pairs in the skulls are not significant as was found by using the
Wilcoxon signed rank test (n = 10, T = 11 and 19, respectively, p > 0.05).

Table 1. Root number and form in great apes and human mandibular and maxillary premolars. B:
buccal; BD: bucco-distal; D: distal; MD: mesio-distal; ML: mesio-lingual; PAL: palatal

Species P3 P4 P3 P4
Pan, female 1 2 (1MB; 1D), fused 2 (1M; 1D blades) 2 (1B, 1PAL) 2 (1B, 1PAL)
Pan, female 2 1 (C-shaped) 2 (ML-BD) circular 3 (2B, 1PAL) 2 (1B, 1PAL)
Pan, male 1 2 (1MB, 1D blade) 2 (1M, 1D blades) 3 (2B, 1PAL) 2 (1B, 1PAL)
Pan, male 2 2 (1MB, 1D blade) 2 (1M, 1D blades) 3 (2B, 1PAL) 3 (2B, 1PAL)
Pongo, male 1 2 (1MB, 1D blade) 2 (1M, 1D blades) 3 (2B, 1PAL), 3 (2B, 1PAL)
Pongo, male 2 2 (1MB, 1D blade) 2 (1M, 1D blades) 3 (2B, 1PAL) 3 (2B, 1PAL)
Gorilla, female 2 (1MB, 1D blade) 2 (1M, 1D blades) 3 (2B, 1PAL) 2 (1B, 1PAL)
Gorilla, male 2 (1MB, 1D blade) 2 (1M, 1D blades) 3 (2B, 1PAL) 3 (2B, 1PAL)
Homo, male 1 1 (circular) 1 (circular) 1 (ellipsoid) 1 (ellipsoid)
Homo, male 2 1 (circular) 1 (circular) 2 (1B, 1PAL) 1 (ellipsoid)
Homo, isolated 1 circular to ellipsoid 1 circular to ellipsoid 1 ellipsoid, n = 8 1 ellipsoid root,
premolars root, n = 10 root, n = 10 2 (1B-PL) n = 2 n = 10

Figures 2 to 4 show bivariate plots between premolar root surface area, crown
area and root length. Spearman rank correlation and RMA statistics are
summarised in Table 2. An enlargement of the crown area correlates with an
increase in root surface area that is similar in one-, two- and three-rooted
premolars (Figure 2, Table 2). In other words, an increase from a single-rooted
to a multi-rooted premolar or from a two-rooted to a three-rooted premolar does
not result in a different scaling relationship between crown area and root surface
area. In contrast, whilst an increase in root length also correlates well with an
increase in root surface area, the scaling pattern of these parameters differs
significantly for single and multi-rooted premolars (Figure 3, Table 2). A tooth
with multiple roots gains relatively more surface area for a given increase in root
length than a single-rooted one. However, there is no difference in RMA slope
between premolars with two and three roots (Table 2). Only premolars with
multiple roots show a significant correlation between crown area and root length
and the RMA slopes are not significantly different (Figure 4, Table 2).
262 Kornelius Kupczik et al.

1000
3 roots
2 roots
900
1 roots
RMA
800

700

600
Root area [mm2]

500

400

300

200

100

0
0 50 100 150 200 250
Crown area [mm2]

Figure 2. Bivariate plot of the premolar root surface area against occlusal crown area showing
RMA regression for the pooled sample
 Premolar Root Number Variation in Hominoids 263

40
3 roots
35 2 roots
1 root
30
Sqrt root area [mm2]

RMA 1 root
25 RMA 2 roots
RMA 3 roots
20

15

10

0
0 5 10 15 20 25
Root length [mm]

Figure 3. Bivariate plot of premolar root surface area (square rooted) against root length showing
RMA regressions

25
3 roots
2 roots
20
1 root
Sqrt root area [mm2]

RMA 2 roots
15 RMA 3 roots

10

0
0 5 10 15 20 25
Root length [mm]

Figure 4. Bivariate plot of premolar crown area (square rooted) against root length showing RMA
regressions
264 Kornelius Kupczik et al.

Table 2. Spearman rank correlation (rs) and RMA statistics for premolar root surface area against
crown area (A), root surface area (square rooted) against root length (B) and crown area (square
rooted) against root length (C). F = F-test probability that slopes between one and two, one and
three and two and three rooted premolars are different. **p < 0.01, ***p < 0.001, n.s. = not
significantly different

A B C
1 root (n = 47), rs 0.46** 0.69*** –0.03, n.s.
RMA slope 5.12 0.82 –
95% conf. int. 3.79, 6.45 0.64, 1.01 –
Intercept –98.62 1.87 –
2 roots (n = 14), rs 0.84*** 0.85*** 0.79**
RMA slope 5.53 1.64 0.62
95% conf. int. 4.21, 6.84 1.23, 2.04 0.39, 0.85
Intercept –143.28 –5.32 0.32
3 roots (n = 19), rs 0.81*** 0.91*** 0.67**
RMA slope 5.27 1.54 0.63
95% conf. int. 3.93, 6.60 1.22 0.39, 0.86
Intercept –90.64 1.86 0.42
F (1 : 2) n.s. *** –
F (1 : 3) n.s. *** –
F (2 : 3) n.s. n.s. n.s.
1–3 roots (n = 80), rs 0.85*** 0.62*** 0.35***
RMA slope 5.47 2.07 0.78
95% conf. int. 5.04, 5.91 1.75, 2.38 0.64, 0.93
Intercept –126.20 –14.75 –3.31

DISCUSSION

The present study set out to investigate whether an increase in premolar root
number is a mechanism to enlarge root attachment area. In addition, we
addressed the question that root surface area might be associated with increasing
occlusal area and with increase in root length. We then asked if in one, two and
three-rooted premolars an increase in root attachment area simply follows on
from either an increase root length or from a widening of the roots (that occurs
as a consequence of a larger crown), or from both of these potential changes in
root morphology.
In this study we followed the definition of root number according to Abbott
(1984). This definition bears on our conclusion that no difference in root surface
area was found between two-rooted and three-rooted premolars. A tooth with a
mesio-buccal root and a blade-like distal root is considered here to be two-
 Premolar Root Number Variation in Hominoids 265

rooted. However, the distal portion may be fully “8” shaped in cross-section,
with two clearly distinct root canals, and in reality may be better described as
having two fused distal roots rather than one. Based on root canals (cf. Tobias,
1995; Ward et al., 1982) such premolars are perhaps better classified as three-
rooted. Hence, the strict division between two- and three-rooted specimens is to
some extent an artefact of the definition used to classify them, as the findings of
this study make clear.
The mandibular and maxillary premolar root morphologies observed in this
study are in agreement with previous findings for the three great ape species, and
also lend support for the suggestion that P. troglodytes females show more
variation in root form and number than the males (Abbott, 1984; Wood and
Engleman, 1988; Tobias, 1995). A common finding in the dentitions of great
apes and humans is that the root number in maxillary premolars is either equal to
the number of roots in mandibular premolars or exceeds it by one. Despite these
differences in root number between maxillary and mandibular premolar pairs
they show similar root attachment areas. Most strikingly however, in the full
sample an increase in root attachment area emerges as being directly associated
with an increase in crown area (Figure 2), where root number has no effect on
root surface area. This is clear evidence to suggest that increasing the number of
roots per se does not result in a larger root area which to support of a larger
crown with. From this perspective, root number seems less of a biomechanical
factor than a genetic polymorphism. However, the way in which a larger root
attachment area is achieved clearly differs in the single- and multi-rooted
premolars. In single-rooted premolars the increase either comes from the
enlarged root girth, a consequence of an increase in crown size, or from an
independent lengthening of the root. On the other hand, multi-rooted premolars
achieve increased attachment area by a combined increase in girth and root
length where both these factors are associated with enlarged crown area (Figures
2 and 4). This difference means that multiple roots have a clear advantage when
root length is constrained: for a given increase in root length they gain
substantially more in attachment area than single-rooted premolars (Figure 3).
Thus, multiple roots can compensate for shorter roots, which are less well
adapted to sustain occlusal loads (Levy and Wright, 1978), and thus provide
better tooth stability. In fact, this relationship has also been proposed for the P3s
of South African australopiths (Sperber, 1974). A further functional advantage
of multiple roots, but not related to attachment area as such, is that they can
spread in different directions and better sustain occlusal loads during different
phases of the chewing cycle. In this way transverse and oblique forces can be
resisted in several directions by the same tooth (Smith, 1986; Macho and Spears,
1999).
266 Kornelius Kupczik et al.

It is worth noting that the different scaling relationships observed for single
and multi-rooted premolars could follow from factors other than those
intrinsically distinguishing these two morphotypes. Nearly all of the single-
rooted premolars are modern human and this might reflect an interspecific, non-
functional difference between humans and great apes. Moreover, the human
sample, with the exception of one skull, derives from an industrial population in
which masticatory constraints underlying morphological correlations may have
been relaxed (see e.g. Spoor et al., 2005). Nevertheless, the present findings for
the single-rooted human premolars in the sample studied here appear to hold true
more generally, given that Garn et al. (1978) found only weak correlations
between root length and crown diameters of mandibular premolars in a North
American sample.
Overall, the results obtained here suggest that in the absence of any
constraints on root length, and any advantage conferred by root spread, variation
in premolar root number may well constitute a genetic polymorphism linked
most closely to phylogenetic differences (Abbott, 1984; Wood et al., 1988;
Tobias, 1995). There remains the possibility that variation of root number may,
however, be more closely related to cuspal morphology than has been
appreciated (Kovacs, 1971). Some support for this comes from observations
made by White et al. (2000) regarding the P3s of the A. afarensis specimen
MAK-VP-1/12. This specimen is described as combining a well developed
mesio-lingual portion of the crown (metaconid) with double-bladed roots as seen
in molars. Likewise, the combination of a P4 with Tomes’ root together with a
well developed talonid in the Miocene hominin Ardipithecus ramidus (Haile-
Selassie, 2001) may suggest a potential link between crown and root
morphology. Yet another example of this can be found in both the P3 and P4 of
the Homo specimen H1 from Atapuerca. These premolars are characterised by
separate disto-lingual and mesio-buccal root components, which are in part
mirrored in the crown by an extended talonid (Bermúdez de Castro et al., 1999).
Future studies with larger sample sizes that include fossil hominins are likely to
shed more light on the issue of multirootedness in premolars.

Acknowledgements

We would like to thank the organisers for all their help and support. We are also grateful to
Simon Hillson and Andrew Chamberlain for comments and discussion concerning this project.
Thanks also to Karl Heinz Höhne for the use of VOXEL-MAN, Nathan Jeffery for help with the
RMA statistics and to Simon Chaplin, Helen Chatterjee and Don Reid for allowing access to
specimens and to staff at the Hammersmith Hospital London and London Bridge Hospital London.
The authors acknowledge The Leverhulme Trust for funding the CT scans through a grant to
MCD.
 Premolar Root Number Variation in Hominoids 267

REFERENCES

Abbott S. A. (1984) A comparative study of tooth root morphology in the great apes, modern man
and early hominids. PhD thesis, University of London.
Bennejeant C. (1936) Anomalies et variations dentaires chez les primates. PhD thesis, Université
de Clermont-Ferrand.
Bermúdez de Castro J. M., Rosas A., Nicolás M. E. (1999) Dental remains from Atapuerca-TD6
(Gran Dolina site, Burgos, Spain). J. Hum. Evol., 37, 523–566.
Brunet M., Guy F., Pilbeam D., Mackaye H. T., Likius A., Ahounta D., Beauvilain A., Blondel C.,
Bocherens H., Boisserie J. R., De Bonis L., Coppens Y., Dejax J., Denys C., Duringer P.,
Eisenmann V., Fanone G., Fronty P., Geraads D., Lehmann T., Lihoreau F., Louchart A.,
Mahamat A., Merceron G., Mouchelin G., Otero O., Pelaez C. P., Ponce de León M. S., Rage
J. C., Sapanet M, Schuster M., Sudre J., Tassy P., Valentin X., Vignaud P., Viriot L., Zazzo A.,
Zollikofer C. (2002) A new hominid from the Upper Miocene of Chad, Central Africa. Nature,
418, 145–151.
Garn S. M., Van Alstine W. L., Cole P. E. (1978) Relationship Between Root Lengths and Crown
Diameters of Corresponding Teeth. J. Dent. Res., 57, 636.
Goh S. W. (1957) Variations in the Morphology of Mandibular Premolar Roots. Brit. Dent. J.,
102, 311–314.
Haile-Selassie Y. (2001) Late Miocene hominids from the Middle Awash, Ethiopia. Nature, 412,
178–181.
Hillson S., FitzGerald C., Flinn H. (2005) Alternative Dental Measurements: Proposals and
Relationships With Other Measurements. Am. J. Phys. Anthropol., 126, 418–426.
Höhne K.-H., Pflesser B., Pommert A., Riemer M., Schiemann T., Schubert R., Tiede U. (1996) A
new representation of knowledge concerning human anatomy and function. Nature Medicine,
1, 506–511.
Kovacs I. (1971) A Systematic Description of Dental Roots. In Dental Morphology and Evolution.
A. A. Dahlberg (ed.), Chicago and London: The University of Chicago Press. pp. 211–256.
Kupczik K. (2003) Tooth root morphology in primates and carnivores. PhD thesis, University of
London.
Leakey M. G., Spoor F., Brown F. H., Gathogo P. N., Kiarie C., Leakey L. N., McDougall I.
(2001) New Hominin genus from eastern Africa shows diverse middle Pliocene lineages.
Nature, 410, 433–440.
Levy A. R., Wright W. H. (1978) The Relationship Between Attachment Height and Attachment
Area of Teeth Using a Digitizer and a Digital Computer. J. Periodontol., 49, 483–485.
Macho G. A., Spears I. R. (1999) Effects of Loading on the Biomechanical Behavior of Molars of
Homo, Pan, and Pongo. Am. J. Phys. Anthropol., 109, 211–227.
Pommert A., Höhne K.-H., Pflesser B., Richter E., Riemer M., Schiemann T., Schubert R.,
Schumacher U., Tiede U. (2001) Creating a high-resolution spatial/symbolic model of the inner
organs based on the Visible Human. Medical Image Analysis, 5, 221–228.
Remane A. (1960) Zähne und Gebiß. In Primatologia. Handbuch der Primatenkunde. Vol. 3. H.
Hofer, A. H. Schultz and D. Starck (eds.), Basel, New York: S. Karger. pp. 637–846.
Senyürek M. S. (1953) A study of the pulp cavities and roots of the lower premolars and molars of
Prosimii, Ceboidea and Cercopithecoidea. Belleten, 17, 321–365.
Smith B. H. (1986) Development and Evolution of the Helicoidal Plane of Dental Occlusion. Am.
J. Phys. Anthropol. 69, 21–35.
268 Kornelius Kupczik et al.

Spencer M. A. (2003) Tooth-Root Form and Function in Platyrrhine Seed-Eaters. Am. J. Phys.
Anthropol., 122, 325–335.
Sperber G. E. (1974) The morphology of the cheek teeth of early South African hominids. PhD
thesis, University of the Witwatersrand.
Spoor F., Leakey M. G., Leakey L. N. (2005) Correlations of cranial and mandibular prognathism
in extant and fossil hominids. Transactions of the Royal Society of South Africa. In press.
Tobias P. V. (1995) Root number in the maxillary third premolars: a very ancient polymorphism.
In Aspects of Dental Biology: Palaeontology, Anthropology and Evolution. J. Moggi-Cecchi
(ed.), Florence: International Institute for the Study of Man, pp. 283–290.
Turner II C. G. (1981) Root Number Determination in Maxillary First Premolars for Modern
Human Populations. Am. J. Phys. Anthropol., 54, 59–62.
Ward S. C., Johanson D. C., Coppens Y. (1982) Subocclusal Morphology and Alveolar Process
Relationships of Hominid Gnathic Elements From The Hadar Formation: 1974–1977
Collections. Am. J. Phys. Anthropol., 57, 605–630.
White T. D., Suwa G., Simpson S., Asfaw B. (2000) Jaws and Teeth of Australopithecus afarensis
From Maka, Middle Awash, Ethiopia. Am. J. Phys. Anthropol., 111, 45–68.
Wood B. A., Engleman C. A. (1988) Analysis of the dental morphology of Plio-Pleistocene
hominids. V. Maxillary postcanine tooth morphology. J. Anat., 161, 1–35.
Wood B. A., Abbott S. A., Uytterschaut H. (1988) Analysis of the dental morphology of Plio-
Pleistocene hominids. IV. Mandibular postcanine root morphology. J. Anat., 156, 107–139.