Co a ia ion be ween he c anium and he ce ical e eb ae in
hominids
Mikel A legi
a
,
b
,
*
, Ana Pan oja-P
e ez
c
, Ch is ine Veschamb e-Cou u e
d
,
Asie G
omez-Oli encia
e
,
,
c
a
Ins i u Ca al
a de Paleoecologia Humana i E oluci
o Social (IPHES-CERCA), Zona Educacional 4, Campus Sescelades URV (Edifici W3), 43007 Ta agona,
Spain
b
Uni e si a Ro i a i Vi gili, Depa men d'His
o ia i His
o ia de l'A , A inguda de Ca alunya 35, 43002 Ta agona, Spain
c
Cen o UCM-ISCIII de In es igaci
on sob e E oluci
on y Compo amien o Humanos, A da. Mon o e de Lemos 5 (Pabell
on 14), 28029 Mad id, Spain
d
UMR 5199 PACEA, Uni e si
e de Bo deaux, All
ee Geo oy Sain Hilai e, B^
a imen B8, CS 50023, 33615, Pessac Cedex, F ance
e
Depa amen o de Geología, Facul ad de Ciencia y Tecnología, Uni e sidad del País Vasco-Euskal He iko Unibe si a ea (UPV/EHU), Ba io Sa iena S/n,
48940 Leioa, Spain
Sociedad de Ciencias A anzadi, Zo oagagaina 11, 20014 Donos ia-San Sebas i
an, Spain
a icle in o
A icle his o y:
Recei ed 30 June 2020
Accep ed 26 Oc obe 2021
A ailable online 8 Decembe 2021
Keywo ds:
Co a ia ion
Phylogeny
Neck
Apes
abs ac
The analysis o pa e ns o in eg a ion is c ucial o he econs uc ion and unde s anding o how
mo phological changes occu in a axonomic g oup h oughou e olu ion. These pa e ns a e ela i ely
cons an ; howe e , bo h pa e ns and he magni udes o in eg a ion may a y ac oss species. These
di e ences may indica e mo phological di e sifica ion, in some cases ela ed o unc ional adap a ions o
he biomechanics o o ganisms. In his s udy, we analyze pa e ns o in eg a ion be ween wo unc ional
and de elopmen al s uc u es, he c anium and he ce ical spine in hominids, and we quan i y he
amoun o di e gence o each ana omical elemen h ough phylogeny. We applied hese me hods o
h ee-dimensional da a om 168 adul hominid indi iduals, summing a o al o mo e han 1000 ce ical
e eb ae. We ound he a las (C1) and axis (C2) display he lowes co a ia ion wi h he c anium in
hominids (Homo sapiens,Pan oglody es,Pan paniscus,Go illa go illa,Go illa be ingei,Pongo pygmaeus).
H. sapiens show a ela i ely di e en pa e n o c anioce ical co ela ion compa ed wi h chimpanzees
and go illas, especially in a iables implica ed in main aining he balance o he head. Finally, he a las
and axis show lowe magni ude o shape change du ing e olu ion han he es o he ce ical e eb ae,
especially hose loca ed in he middle o he subaxial ce ical spine. O e all, esul s sugges ha di -
e ences in he pa e n o c anioce ical co ela ion be ween humans and go illas and chimpanzees
could eflec he pos u al di e ences be ween hese g oups. Also, he s onge c anioce ical in eg a ion
and la ge magni ude o shape change du ing e olu ion shown by he middle ce ical e eb ae sugges s
ha hey ha e been selec ed o play an ac i e ole in main aining head balance.
©2022 The Au ho s. Published by Else ie L d. This is an open access a icle unde he CC BY-NC-ND
license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/).
1. In oduc ion
The ela ionship be ween o m and unc ion is p esen in many
biological s uc u es (e.g., P euscho , 2004;E coli e al., 2012;
Hu chinson, 2012). In human e olu ion, one o he mos explo ed
opics is he s udy o mo phological changes ha may ha e
occu ed as an adap a ion o bipedal locomo ion (Robinson, 1972;
B amble and Liebe man, 2004;Sockol e al., 2007;Lo ejoy e al.,
2009a,b,c;Wa ene e al., 2015;Ryan and Sukhdeo, 2016;Ryan
e al., 2018). Those s udies la gely ocused on ana omical ele-
men s di ec ly ela ed o locomo ion, such as he pel is and lowe
limbs (S e n, 2000;Pon ze e al., 2009;G abowski e al., 2011;
G abowski and Roseman, 2015). Howe e , in ecen yea s, he
numbe o s udies ega ding he e eb al column has inc eased,
wi h mos s udies ocusing on he lumba egion and gi ing special
a en ion o di e ences in he deg ee o lo dosis bo h be ween
sexes and ac oss hominin species (Whi come e al., 2007;Been
e al., 2012,2014;G
omez-Oli encia e al., 2017) and some on he
ho acic egion (e.g., Bas i e al., 2014,2017;Been e al., 2017;
G
omez-Oli encia e al., 2018).
*Co esponding au ho .
E-mail add ess: [email p o ec ed] (M. A legi).
Con en s lis s a ailable a ScienceDi ec
Jou nal o Human E olu ion
jou nal homepage: www.else ie .com/loca e/jhe ol
h ps://doi.o g/10.1016/j.jhe ol.2021.103112
0047-2484/©2022 The Au ho s. Published by Else ie L d. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/).
Jou nal o Human E olu ion 162 (2022) 103112
The li e a u e ega ding he ce ical egion in p ima es was,
un il ecen ly, ela i ely sca ce compa ed wi h he o he spinal
egions (e.g., Schul z, 1942,1961;Slijpe , 1946;F ancis, 1955a,b;
Toe ien, 1957,1961;Jenkins, 1969). Mo e ecen ly, he in e es in
his egion has inc eased, in pa icula wi h s udies analyzing he
mo pho unc ional in e ac ions wi h pos u e and locomo ion
(Man eda e al., 2006;Mi e oecke e al., 2007;Been e al., 2014;
Nalley and G ide -Po e , 2017;A legi e al., 2017,2018;Meye e al.,
2018) and he ela ionship wi h he c anium (Nalley and G ide -
Po e , 2015,2019;Villamil, 2018). These (and o he ) s udies ha e
used se e al pe spec i es o e alua e his unc ional ela ionship
no only in p ima es bu also in o he mammal g oups: o example,
app oaches based on he biomechanical analysis o he c anium
(Demes, 1985); kinema ic analyses o his complex in he wild and
in cap i i y (B amble, 1989;S ai and Ross, 1999;Dunba and
Badam, 2000;C omwell e al., 2001;Choi e al., 2003;Dunba
e al., 2008;Zubai e al., 2019); analyses om adiog aphs, pho-
og aphs, elec onic senso s, and dissec ions (Vidal e al., 1986;G a
e al., 1995a,b;Benoi e al., 2020;Jo issen e al., 2020); app oaches
based on he mo phological co ela ion and in eg a ion among
ai s (Nalley and G ide -Po e 2015;Villamil, 2018); and muscu-
loskele al analyses o define modules in he c anioce ical complex
(Diogo e al., 2008,2017;Diogo and Wood, 2011;Es e e-Al a a
e al., 2015;A nold e al., 2017a;Powell e al., 2018;Boyle e al.,
2020).
B oadly, hese s udies e ealed ha despi e pos u al di e ences
in mammals, no subs an ial di e ences exis among axa in main-
aining c anial balance agains g a i y a es (Vidal e al.,1986;G a
e al.,1995a,b). In his passi e pos u e, he mammal neck adop s an
s-shaped e ical posi ion o minimize he dis ance be ween he
mass o he head and he weigh -bea ing ce ico ho acic junc ion,
hus educing do sal neck muscula s ess (Vidal e al., 1986;G a
and Wilson, 1989). In quad upeds, he e ical posi ion o he
neck equi es high do siflexion o he ce ico ho acic a icula ion
(C6eT2) and hype ex ension o he a lan o-occipi al join , which
also allows he adjus men o head o ien a ion and gaze (Vidal
e al., 1986;Whi e and Panjabi, 1990;G a e al., 1995a;Nalley
and G ide -Po e , 2019). The impo an unc ional ole played by
he c anial and caudal ce ical modules con as s wi h he mid-
ce ical module, which is mainly ci cumsc ibed o axial o a ion
and does no show unc ional specializa ions (G a e al., 1995a,b;
A nold, 2020). Howe e , species ha do no display a comple e
quad upedal pos u e, such as some p ima es, and especially mod-
e n humans, show a mo e limi ed ange o mo ion in he a lan o-
occipi al a icula ion; hus, hey ci cumsc ibe mos o he c anio-
ce ical mo ions in he midsagi al plane o he ce ico ho acic
a icula ion (Whi e and Panjabi, 1990;G a e al., 1995a,b). The
ela i ely sligh di e ences obse ed a es be ween quad upedal
and nonquad upedal mammals inc ease du ing locomo ion. The
o me eo ien s he neck ho izon ally du ing exe ion, whe eas
humans and plausibly o he up igh mammals do no (Vidal e al.,
1986;G a e al.,1995a,b;Dunba and Badam,1998;S ai and Ross,
1999).
The concep s o in eg a ion and modula i y e e o he deg ee
o in e ac ion be ween he cha ac e s o one o mo e ana omical
s uc u es (Olson and Mille , 1958). Bo h concep s ha e been
defined as impo an in he pheno ypic e olu ion o o ganisms
om a de elopmen al, gene ic, and/o unc ional poin o iew
(e.g., Olson and Mille , 1958;Che e ud, 1996;Wagne , 1996;
Goswami e al., 2014). Mo phological in eg a ion desc ibes high
deg ee o co ela ion wi hin subse s o mo phological ai s, which
may esul in long- e m coe olu ion (Che e ud, 1996). Modula i y
e e s o he ela i e independence o ai s ha a e pa o di e en
de elopmen al o unc ional egions. In e olu iona y s udies, hese
concep s a e c ucial o he econs uc ion and unde s anding o
how mo phological changes occu in o ganisms because hey can
acili a e o es ic he e olu ion o hei cha ac e s in specific di-
ec ions (Wagne , 1996;Hallg ímsson e al., 2007;Goswami and
Polly, 2010;G
omez-Robles and Polly, 2012).
Because in eg a ion can enhance o cons ain mo phological
e olu ion, es ablishing how pa e ns o in eg a ion ha e e ol ed
concu en ly wi h he mo phology can help de e mine he e olu-
ion o a g oup (Wagne , 1988;G abowski e al., 2011). In gene al
e ms, i has been p oposed ha in eg a ion pa e ns a e ela i ely
cons an in species (Goswami, 2006;Po o e al., 2009;Ba dua
e al., 2019;Wa anabe e al., 2019). Howe e , bo h he pa e ns
and he magni udes o in eg a ion may a y ac oss species (Ma oig
and Che e ud, 2001;Ma oig e al., 2009;Po o e al., 2009;
Goswami and Polly, 2010). De ec ing po en ial di e ences in in e-
g a ion pa e ns is c i ical because hese changes may indica e
mo phological di e sifica ion as a esul o possible adap a ion o
selec ion p essu es a ec ing e olu iona y ajec o ies, likely ela ed
o unc ional e ec s o he mechanics o o ganisms (Wagne and
Schwenk, 2000).
The numbe o e eb ae in he e eb al column is egula ed by
he exp ession o he Hox genes, and hose o he pa alog g oups 4
and 5 con ol he o ganiza ion o he ce ical egion (Kessel and
G uss, 1991;Bu ke e al., 1995;Galis, 1999a;Wellik and Capecchi,
2003). In mammals, almos all species p esen a fixed numbe o
se en ce ical e eb ae (Ba eson, 1894;Johnson and O'Higgins,
1996;Galis, 1999b;Na i a and Ku a ani, 2005;Va ela-Lashe as
e al., 2011;Buchhol z, 2014;B€
ohme , 2017;B€
ohme e al., 2018),
which, a he same ime, a e in e nally o ganized in o h ee unc-
ional and de elopmen al modules: uppe (C1eC2), middle
(C3eC5), and lowe ce ical (C6eC7; A nold e al., 2016;Randau
e al., 2017). O he s udies p oposed a sligh ly di e en subdi i-
sion o he ce ical spine, which includes ei he he c anial base
(CB) as pa o he uppe module (i.e., CBeC1) o he c anium and
ho acic spine o some species (A nold e al., 2017a;Villamil, 2018).
The ce ical spine is a ansi ional unc ional egion be ween he
head and he es o he e eb al column, whe e each ce ical
module plays a di e en unc ional ole (G a e al., 1995b;A nold,
2020). In he synapsid/mammal ansi ion, hese unc ional mod-
ules did no e ol e a he same ime; hey s a ed wi h he
appea ance o he uppe ce ical module (an ea ly a las-dens-axis
join ) and finished by he consolida ion o a lowe ce ical mod-
ule (Buchhol z e al., 2012;A nold, 2020 and e e ences he ein).
Rega ding he mo phological link be ween he c anium and he
ce ical egion in p ima es, some esea che s ha e ound ce ain
co ela ions be ween specific cha ac e s, which could ha e impli-
ca ions o pos u e and locomo o beha io s (S ai and Ross, 1999;
Nalley and G ide -Po e , 2017;Villamil, 2018). In gene al, axa wi h
mo e ho izon al necks show cha ac e is ic ce ical ai s o a oid
e eb al a icula displacemen (e.g., co onal o ien a ion o he
a icula ac s) and o he inse ion o he la ge epaxial muscula-
u e ha suppo s he long load a m ha esul s om his pos u e
(e.g., la ge ans e se and spinous p ocesses; Adams and Moo e,
1975;Eb aheim e al., 2008;Nalley and G ide -Po e , 2015;
A legi e al., 2017). Villamil (2018) ound ha he CB and he ce -
ical e eb ae in hominoids, especially he C1 and he cen al
ce ical e eb ae (C3eC5), a e s ongly in eg a ed. Howe e , i was
concluded ha body pos u e and locomo ion a e ela i ely weak
selec ion p essu es in he mo phology o ce ical e eb ae.
S udies using ana omical ne wo k analysis also sea ched o
unc ional, e olu iona y, and/o de elopmen al modules in he
head-neck by combining he analyses o ha d (bone/ca ilage) and
so issues (muscles; Diogo e al., 2008,2017;Diogo and Wood,
2011;Es e e-Al a a e al., 2015;Powell e al., 2018;Boyle e al.,
2020). These show ha he head and neck muscles in p ima es,
compa ed wi h o he ana omical egions, a e a be e ma ch o he
M. A legi, A. Pan oja-P
e ez, C. Veschamb e-Cou u e e al. Jou nal o Human E olu ion 162 (2022) 103112
2
mos ecen molecula ee (Diogo and Wood, 2011;Boyle e al.,
2020) and ha his complex is mainly defined by unc ion
(Es e e-Al a a e al., 2015). They u he s ess ha mode n
humans ha e mo e head and neck muscles han any o he p ima e
(Powell e al., 2018), and, in e es ingly, his leads o he leas
complex and mos de i ed musculoskele al sys em in his egion
among hominoids (Diogo and Wood, 2011,2017;Powell e al.,
2018).
The main objec i e o his s udy is o quan i y and analyze
pa e ns o co ela ion and he magni udes o in eg a ion be ween
he mo phology o he en i e c anium and he ce ical e eb ae in
hominids. Mo eo e , ac o s ha influence in eg a ion on s uc-
u es in he e olu iona y p ocess, such as he e ec o size and
phylogeny, will be conside ed. He e, we es he ollowing
hypo heses:
1) The mo phological di e ences among unc ional and de elop-
men al ce ical modules will be eflec ed in he deg ee o c a-
nioce ical in eg a ion (e.g., Villamil, 2018).
2) In eg a ion is expec ed o be s onge in species wi h la ge
musculoskele al ea u es in he do sal neck, which would indi-
ca e g ea e mechanical bending loads in he muscles espon-
sible o coun e balancing he g a i a ional o ces ac ing on he
head (Adams and Moo e, 1975).
3) Despi e pa e ns o in eg a ion being ela i ely cons an in
mammals, we expec ce ain di e ences in he pa e n o c a-
nioce ical co ela ion among g oups ha show di e en
pos u al and locomo o epe oi es (G abowski e al., 2011;
G abowski and Roseman, 2015).
4) The ce ical e eb ae in hominids p esen ce ain e olu iona y
dispa i ies (e.g., mo e di ec ional o s abilizing selec ion), and
hese di e ences would co espond wi h he in e nal modula
di ision o he ce ical spine (A nold, 2020 and e e ences
he ein).
5) Highly in eg a ed se s o ai s a e mainly aligned along he axis
o size- ela ed a ia ion (Ma oig e al., 2009;Po o e al., 2009);
he e o e, emo ing he e ec o size will esul in a educ ion in
he le els o in eg a ion (e.g., Po o e al., 2013;A legi e al., 2018).
2. Ma e ials and me hods
The sample s udied in his wo k comp ises he c anium and he
se en ce ical e eb ae o 160 adul indi iduals o he amily
Hominidae, o aling 160 c ania and 1071 ce ical e eb ae: 43
Homo sapiens,46Pan oglody es,12Pan paniscus,45Go illa go illa,
8Go illa be ingei g aue i,3Go illa be ingei, and 3 Pongo pygmaeus
(Supplemen a y Online Ma e ial [SOM] Table S1). The su ace o
each ana omical elemen was scanned in wo iews, c anial and
caudal, using a Go!SCAN 20 (wi h a esolu ion o 0.1 mm o he
e eb ae and 0.4 mm o he c ania) and la e i ually assembled
in o a single h ee-dimensional (3D) objec using VXelemen s
so wa e . 6.3 (C ea o m Inc., L
e is).
Bo h geome ic mo phome ic (GM) and adi ional mo pho-
me ic (TM) me hods we e used o analyze co a ia ion be ween he
c anium and he ce ical e eb ae in Hominidae. GM me hods
we e used o pe o m he analyses a he in e specific le el, and TM
me hods we e used a he in aspecific le el. Be o e pe o ming he
analyses, landma ks ha could no be cap u ed due o damage in
he bone we e, i possible, es ima ed using bila e al symme y and
o he wise calcula ed using pa ial leas squa es (PLS; Books ein
e al., 1990;Rohl and Co i, 2000). O e all, less han 2% o he o-
al landma ks we e es ima ed. All s a is ical analyses we e pe -
o med in R . 4.0.2 (R Co e Team, 2020), and, mo e specifically, o
GMs using he package ‘geomo ph’ . 3.2.1 (Adams e al., 2020).
2.1. Da a collec ion
Th ee-dimensional landma ks ep esen ing he mo phology o
each ana omical elemen we e i ually cap u ed om he c ea ed
3D scan models: 33 landma ks in he c anium, 27 in he a las (C1),
33 in he axis (C2), and 34 in he subaxial ce ical e eb ae
(C3eC7; see SOM Tables S2eS5 o landma k defini ions; Figs.1 and
2) using Viewbox 4 so wa e . 4.5.0 (dHAL so wa e, Kifissia). A
high pe cen age o C3eC5 e eb ae o H. sapiens p esen wi h
bi ube cula i y o he ip o he spinous p ocess (G
omez-Oli encia
e al., 2013). In hese cases, he landma ks co esponding o he
mos do sal poin o he spinous p ocess we e placed i ually in
he midsagi al line. Be o e pe o ming he analyses, hose land-
ma ks ha could no be cap u ed owing o damage in he bone
we e es ima ed using bila e al symme y i possible and o he wise
we e calcula ed using PLS eg ession (Books ein e al., 1990;Rohl
and Co i, 2000). O e all, less han 2% o he o al landma ks
we e es ima ed. Owing o he di ficul y o ob aining indi iduals in
he collec ions ha included bo h he c anium and all se en ce -
ical e eb ae, hose ha had he c anium and a minimum o fi e
ou o he se en ce ical e eb ae we e selec ed. O he s udied
indi iduals, 88% p esen ed he eigh ana omical elemen s exam-
ined he e (i.e., c anium and se en ce ical e eb ae) and he
emaining 12% (n¼20) only lack one o wo ana omical elemen s.
2.2. Geome ic mo phome ics
All analyses using GM echniques we e pe o med a he in e -
specific le el including all axa. Be o e pe o ming he s a is ical
analyses, we conduc ed a gene alized P oc us es analysis (GPA;
Rohl and Slice, 1990) om he aw 3D coo dina es o each ce ical
e eb a and he c anium sepa a ely including all axa o emo e
he in o ma ion ela ed o size, posi ion, and o ien a ion.
In eg a ion and pai wise compa isons We quan ified he deg ee o
mo phological in eg a ion be ween shape a iables ha desc ibe
he di e en elemen s o he c anium and ce ical e eb ae. We
used he wo-block PLS eg ession me hod, using he
‘in eg a ion. es ’ unc ion o he package ‘geomo ph’(Adams
e al., 2020). This me hod uses he decomposi ion o he
be ween-block co a iance ma ix and sea ches o pai s o new
axes ha ep esen he maximum amoun o co a iance be ween
he wo blocks. As each axis o each block (e.g., axis 1 o block 1)
only co ela es wi h he co esponding axis o he o he block
(e.g., axis 1 o block 2), he co a iance pa e n can only be
analyzed by a pai o PLS axes a a ime (Books ein e al., 1990;
Klingenbe g and Zaklan, 2000;Rohl and Co i, 2000). The esul s
o his p ocedu e a e app op ia e because i yields alues ha a e
una ec ed by sample size o by he numbe o a iables (Adams
and Collye , 2016). The significance was calcula ed by compa ing
he ob ained PLS co ela ion (
PLS
) alues wi h hose esul ing
om a andom pe mu a ion o he indi iduals in one pa i ion
ela i e o hose in he o he (Books ein e al., 2003;Adams and
Collye , 2016). Then, o es whe he each ce ical e eb a
p esen ed a significan ly di e en magni ude o in eg a ion wi h
he c anium compa ed wi h he o he e eb ae, we calcula ed
he e ec sizes o each pai wise PLS analysis using he
‘compa e.pls’ unc ion o ‘geomo ph’(Collye e al., 2015;Adams
and Collye , 2016). This analysis, a he han using he PLS
co ela ion coe ficien , pe o ms wo-sample Z- es s, which a e
obus o di e ences in sample size and he numbe o landma ks.
Phylogene ic signal Phylogene ically ela ed species sha e an
e olu iona y his o y and hus end o display simila ai alues
due o hei common ances y (Felsens ein, 1985). To e alua e
shape di e gence o e he e olu ion o he c anium and he
M. A legi, A. Pan oja-P
e ez, C. Veschamb e-Cou u e e al. Jou nal o Human E olu ion 162 (2022) 103112
3
ce ical e eb ae, we calcula ed he phylogene ic signal o ou da a
using he phylogene ic in o ma ion o hominids om he 10kT ees
P ojec pla o m (A nold e al., 2010) and he P oc us es mean
a e age o each species o each elemen independen ly. To
es ima e he phylogene ic signal, we used he unc ion ‘physignal’
o ‘geomo ph,’which is based on he mul i a ia e e sion o he
K-s a is ic (K
mul
:Adams, 2014). This me hod es ima es he deg ee
o phylogene ic signal in a da a se ela i e o wha is expec ed
unde a B ownian mo ion o e olu ion. Con e sely o p e ious
me hods, K
mul
is app op ia e o highly dimensional mul i a ia e
da a a oiding ype I e o s. Significance was es ima ed ia shape
da a pe mu a ion among he ips o he phylogeny (Adams, 2014;
Adams and Collye , 2019). Finally, we explo ed pa e ns o c anial
and ce ical e eb ae shape e olu ion by p ojec ing he
phylogeny on o he mo phological mo phospace ep esen ed by
he fi s wo p incipal componen s (PC1 and PC2).
Phylogene ic PLS analysis To accoun o ela edness among he
hominid species in ou sample, we calcula ed c anioce ical in e-
g a ion while accoun ing o he phylogene ic ela ionships among
axa using phylogene ic PLS analyses. This app oach, implemen ed
in he unc ion ‘phylo.in eg a ion’o ‘geomo ph,’also displays
app op ia e ype I e o a es and cons an le els o in eg a ion
i espec i e o he numbe o species o ai dimensions (Adams
and Felice, 2014). The significance o he analyses was pe o med
ia pe mu a ion analysis in he same way as o he PLS analysis
(discussed ea lie ).
2.3. T adi ional mo phome y
All analyses om he T adi ional mo phome y sec ion we e
pe o med a he in aspecific le el, ha is, including only he h ee
species wi h he g ea es sample sizes (i.e., H. sapiens,G. go illa, and
P. oglody es). He e, we analyzed co a ia ion be ween he c anium
and he ce ical e eb ae using linea measu emen s om bo h
aw da a se s and size-adjus ed ( emo ing he influence o size).
Fi s , linea measu emen s we e de i ed om he 3D coo dina es
using he ‘in e lmkdis ’ unc ion o ‘geomo ph.’These linea a i-
ables we e selec ed om a se ies o s anda d measu emen s ha
bes ep esen he mo phology o he c anium and each ce ical
e eb a (SOM Tables S6 and S7;McCown and Kei h, 1939;Ma in
and Salle , 1957;Howells, 1973;B €
aue , 1988). Six a iables we e
selec ed o he a las (C1), nine o he axis (C2), and ano he nine
o each o he C3eC7 e eb ae (Fig. 3). Co ela ion analyses be-
ween wo elemen s equi e ha hey bo h ha e he same numbe
o a iables, so he nine a iables selec ed o he c anium we e
educed o six o he analyses wi h he a las (SOM Table S8). A e
ha , o es whe he sex was a significan sou ce o a ia ion in he
da a se s ha needed o be emo ed, we accoun ed o he ela i e
amoun o shape a ia ion a ibu able o sex on each e eb al and
c anial elemen pe species by c ea ing a linea model and es i-
ma ing he p obabili y ia analysis o a iance. All analyses esul ed
in nonsignifican di e ences be ween sexes, and hus, we did no
co ec sex a ia ion om he da a se s.
PLS in eg a ion analysis Wi h he ob ained linea a iables om
he aw 3D coo dina es, we fi s analyzed he magni udes o in e-
g a ion be ween he c anium and he ce ical e eb ae in he h ee
species using he ‘in eg a ion. es ’ unc ion o ‘geomo ph,’ he same
me hod applied o GMs (discussed ea lie ).
Be ween a iables pai wise co ela ion es Nex , we analyzed
pa e ns o co ela ion be ween he c anium and he ce ical
e eb ae in he h ee species by quan i ying he co ela ion be-
ween pai s o a iables. To do so, we calcula ed he Pea son co -
ela ion coe ficien and significance be ween all he c anial
a iables, on he one side, and all he a iables o each e eb a, on
he o he side, using he ‘co . es ’ unc ion o he R package ‘s a s’ .
4.0.2 (R Co e Team, 2020).
Figu e 1. Landma ks ( ed do s) and linea measu emen s (blue lines) aken o he c anium in his s udy. Landma ks: uppe le , c anium in le la e al iew; lowe le , in caudal
iew; uppe cen e , in en al iew. Linea measu emen s: lowe cen e , in le la e al iew; lowe cen e , in en al iew. No e ha no all landma ks a e isible in all iews. See
SOM Tables S2 and S6 o landma k and linea measu emen s defini ion. (Fo in e p e a ion o he e e ences o colo in his figu e legend, he eade is e e ed o he Web e sion
o his a icle.)
M. A legi, A. Pan oja-P
e ez, C. Veschamb e-Cou u e e al. Jou nal o Human E olu ion 162 (2022) 103112
4
Influence o size Nex , we assessed he po en ial influence o size
on in eg a ion. To do so, we calcula ed he geome ic mean o each
ana omical elemen (c anium and ce ical e eb ae) using hei
linea measu emen alues (six o he a las and nine o each o he
o he elemen s), and we used i as a p oxy o size (Da och and
Mosimann, 1985;Junge s e al., 1995). To ob ain ‘size-co ec ed’
da a se s, we di ided he aw alues o he linea measu emen s
by he geome ic mean o each e eb a and he c anium
(Coleman, 2008;Pablos e al., 2013;A legi e al., 2017), he ea e
size-adjus ed. Then, we epea ed he wo p e ious analyses (i.e.,
‘in eg a ion. es ’and ‘co . es ’) using he ob ained size-adjus ed
da a se s. In addi ion, we analyzed he amoun o ce ical shape
a ia ion explained by c anial size (i.e., allome y) and he
di e ences in he allome ic pa e n among species. To do so, we
pe o med a eg ession analysis using he linea measu emen s
om he aw da a se s as dependen a iables and he c anial
size ep esen ed by he geome ic mean as he independen
a iable using he ‘p ocD.lm’ unc ion o ‘geomo ph’and he
‘pai wise’ unc ion o ‘RRPP’ . 0.6.2 (Collye and Adams, 2018,
2021) o calcula e he angle be ween he male and emale
eg ession ec o s and i s significance.
2.4. Repea abili y o he da a se s
Finally, o asce ain he eliabili y o ou esul s, and ollowing
Melo e al. (2016), we es ed he epea abili y o he aw da a se s
(e.g., C3 G. go illa,C4P. oglody es) by boo s apping each da a se
10,000 imes and compa ing he o iginal and he ob ained co-
a iances ma ix using andom skewe s analysis (mean ¼0.880,
median ¼0.881; SOM Table S9).
3. Resul s
3.1. Geome ic mo phome ics
In e specific in eg a ion and pai wise compa isons All analyses
yielded high and significan esul s (Table 1), wi h c anioce ical
magni udes o in eg a ion anging be ween
PLS
¼0.503
(c anium/a las) and
PLS
¼0.832 (c anium/C6). The a las and axis
e ealed he lowes alues o c anioce ical co a ia ion ollowed
by he C7 and C3. Thus, cen al ce ical e eb ae (i.e., C4eC6)
showed highe alues o co a ia ion wi h he c anium han hose
loca ed mo e pe iphe ally in he ce ical spine. The al as and axis
yielded significan ly di e en alues o magni udes o
c anioce ical in eg a ion compa ed wi h he es o he
e eb ae (Table 2). The es o he pai wise compa ison did no
e eal significan di e ences among hem; howe e , i is wo h
ema king ha he lowes di e ences we e ob ained among
middle ce ical e eb ae (i.e., C4eC6). These esul s ela i ely
suppo ou hypo hesis ha he deg ee o c anioce ical
in eg a ion would be di e en be ween e eb ae om di e en
ce ical modules.
Figu e 2. Landma ks used in his s udy in he ce ical e eb ae. No e ha no all landma ks a e isible in all iews. See SOM Table S3 o landma k defini ion. The figu ed e eb ae
belong o Pan oglody es.
M. A legi, A. Pan oja-P
e ez, C. Veschamb e-Cou u e e al. Jou nal o Human E olu ion 162 (2022) 103112
5
Phylogene ic signal The esul s o he phylogene ic analyses a e
shown in Table 3. They e ealed ha only he c anium, C3, and C4
e eb ae shapes exhibi ed significan ( hough weak) phylogene ic
signal, indica ing ha c anial and ce ical e eb ae shapes om
hese axa esemble each o he less han expec ed unde he
B ownian mo ion model o e olu ion. The a las and axis exhibi ed
he lowes alues (K
mul
¼0.052 and 0.058, espec i ely) and he
c anium he highes (K
mul
¼0.337), indica ing lowe and highe
magni udes o shape change du ing e olu ion, espec i ely. The
subaxial ce ical e eb ae displayed a end ha inc eases om
C3 o C4 (maximum alue, K
mul
¼0.296) and hen dec eases
owa d C7 (minimum alue, K
mul
¼0.112). This suppo s ou
hypo hesis ha ce ical e eb ae in hominids p esen a
di e gence in he magni udes o e olu iona y a ia ion.
Phylomo phospaces Visual in o ma ion o he phylomo phospaces
is shown in Figu e 4. I he magni udes o phylogene ic signal
(discussed ea lie ) indica es he magni ude and he di ec ion o
shape di e gence in he p ocess o e olu ion (Klingenbe g and
Gidaszewski, 2010), he phylomo phospaces allow one o
isualize he his o y o mo phological di e sifica ion o a clade
(Sidlauskas, 2008). In Figu e 4, we can obse e ha he c anium
and he subaxial ce ical e eb ae show a simila e olu iona y
Figu e 3. Linea measu emen s o he ce ical e eb ae used in his s udy. The numbe s indica e he landma ks om which he linea measu emen s ha e been calcula ed.
Table 1
In eg a ion alues (
PLS
), e ec sizes, and s anda d e o s (SE) be ween he c anium
(C ) and he ce ical e eb ae (C1eC7) a he in e specific le el (Hominidae).
a
PLS
pE ec size SE
C /C1 0.503 <0.001 4.026 0.004
C /C2 0.594 <0.001 5.874 0.003
C /C3 0.772 <0.001 7.200 0.003
C /C4 0.825 <0.001 7.587 0.004
C /C5 0.776 <0.001 6.014 0.004
C /C6 0.832 <0.001 7.506 0.004
C /C7 0.707 <0.001 5.544 0.004
a
Significan alues a e indica ed in bold (p- alue <0.05).
Table 2
Pai wise di e ences in PLS e ec sizes compa ing le els o c anioce ical mo pho-
logical in eg a ion in hominids.
a
C /C1 C /C2 C /C3 C /C4 C /C5 C /C6 C /C7
C /C1 0.062 <0.001 <0.001 <0.001 <0.001 <0.001
C /C2 1.867 0.003 <0.001 0.001 <0.001 0.059
C /C3 4.697 2.926 0.314 0.625 0.359 0.357
C /C4 5.522 3.832 1.006 0.602 0.928 0.063
C /C5 5.113 3.376 0.488 0.521 0.666 0.168
C /C6 5.455 3.756 0.917 0.090 0.432 0.075
C /C7 3.632 1.891 0.921 1.861 1.380 1.779
a
Significan alues a e indica ed in bold (p- alue <0.05). Values a e ep esen ed
in he lowe diagonal, and p- alues in he uppe diagonal.
Table 3
Phylogene ic signal o he c anium and ce ical e eb ae in hominids ep esen ed
in K
mul
alues.
Phylogene ic signal pE ec size
C anium 0.337 0.033 1.667
C1 0.052 0.479 0.043
C2 0.058 0.239 0.557
C3 0.186 0.028 1.757
C4 0.296 0.020 1.691
C5 0.283 0.064 1.796
C6 0.226 0.063 1.762
C7 0.112 0.085 1.323
Significan alues a e indica ed in bold (p- alue <0.05).
M. A legi, A. Pan oja-P
e ez, C. Veschamb e-Cou u e e al. Jou nal o Human E olu ion 162 (2022) 103112
6
end. Hominine g oups a e sepa a ed along, and almos in pa allel,
he axis o he PC1, showing a s ong di e gence in opposi e
di ec ions o go illas and humans. The genus Pongo di e ges
om he es o he amily, mos ly in he di ec ion o he PC2. In
he a las, P. paniscus and H. sapiens on he one hand, and
P. oglody es and go illas on he o he hand, show sho e
dis ances be ween hem han wi h espec o hei common
ances o , indica ing a ce ain homoplasy in shape. Finally, i is
di ficul o asce ain he e olu iona y a io based only on he
phylomo phospaces; howe e , excep o he a las, he longe
b anches o humans indica e as e di e gence in his species
han hose o he o he axa.
Phylogene ic PLS The esul s o he phylogene ic PLS analyses a e
shown in SOM Table S10. All pai wise es s yielded high and
significan alues (
PLS
>0.965). In con as o he phylogene ic
unco ec ed analyses, none o he e eb ae showed significan
di e ences in he
PLS
alues compa ed wi h he es o he
e eb ae.
3.2. T adi ional mo phome ic analyses o aw da a
This sec ion p esen s he in eg a ion analyses be ween he
c anium and he ce ical e eb ae based on linea a iables a he
in aspecific le el. Fo his pu pose, only he h ee species wi h
la ge sample sizes (H. sapiens,P. oglody es, and G. go illa) we e
assessed.
PLS in eg a ion analysis The esul s o he PLS analysis om aw
da a a e p esen ed in Table 4 and Figu e 5. They show ha all
ce ical e eb ae, excep o he a las in humans and
chimpanzees, we e significan ly in eg a ed wi h he c anium o
he h ee species. The h ee g oups showed lowe alues in he
a las, which significan ly di e om he es o he e eb ae,
excep o he C3 in go illas (SOM Table S11). In gene al e ms,
go illas possessed he highes alues o co a ia ion, ollowed by
chimpanzees and humans. O e all, he pa e n o changes in he
magni udes o co a ia ion mo ing down he e eb al column
was simila in di e en species, bu wi h a di e en magni ude o
Figu e 4. Phylomo phospaces o he c anium and he se en ce ical e eb ae in Hominidae. The phylogene ic ee o his amily was supe imposed on o he fi s wo p incipal
componen s o each ana omical elemen . The ed do indica es he oo o he ee, and he numbe s on he b anches (in e nal nodes) ep esen he las common ances o s o he
espec i e pai s o lineages. (Fo in e p e a ion o he e e ences o colo in his figu e legend, he eade is e e ed o he Web e sion o his a icle.)
M. A legi, A. Pan oja-P
e ez, C. Veschamb e-Cou u e e al. Jou nal o Human E olu ion 162 (2022) 103112
7
in eg a ion, excep o he C3 in humans, which depa ed he
pa e n ollowed by chimpanzees and go illas (Fig. 5). Pai wise
compa isons among he h ee species e ealed ha go illas
significan ly di e om humans and chimpanzees in he
c anioce ical magni ude o in eg a ion in he se en ce ical
e eb ae (SOM Table S12). In con as , humans and chimpanzees
only displayed significan di e ences in hose e eb ae loca ed
in he midlowe ce ical spine (i.e., C4, C5, and C6). These esul s
suppo he hypo hesis ha species wi h la ge musculoskele al
ea u es in he do sal neck display a highe deg ee o in eg a ion.
Be ween a iables pai wise co ela ion es The esul s e ealed
ha he co ela ion be ween he c anium and he ce ical e e-
b ae in he h ee species occu ed a di e en deg ees and among
di e en ai s (Table 5 and Fig. 6;SOM Table S13). In gene al e ms,
in go illas and chimpanzees, he highes c anioce ical co ela ions
we e shown be ween ei he he leng h o he CB o he leng h o
he ace in he c anium and he maximum do so en al o
ans e sal diame e ai s in he ce ical e eb ae (Table 5). In
humans, in con as , a clea co ela ion pa e n did no exis
be ween c anioce ical a iables. Indeed, in his axon, he
s onge co ela ion alues we e displayed in ai s ela ed o
maximum c anial leng h and e eb al body heigh , ans e se,
and do so en al diame e (i.e., M1, M5, and M8). Also, we can
obse e how, besides magni ude o in eg a ion (highe alues in
go illas), go illas and chimpanzees p esen a mo e simila pa e n
o co ela ion among c anioce ical ai s h oughou he ce ical
spine compa ed wi h humans. They show he highes
co ela ions in he uppe le o each diag am, whe e a iables
ela ed o he midsagi al plane a e loca ed (Fig. 6). Con e sely, in
humans, he dis ibu ion o he highes co ela ions is no
concen a ed in ha a ea o he diag am and shows a mo e
he e ogeneous pa e n h oughou he ce ical egion. These
Table 4
In eg a ion alues (
PLS
), e ec sizes (E-size), and s anda d e o s (SE) om he co a ia ion analysis be ween he c anium and he ce ical e eb ae a he in aspecific le el
based on linea measu emen s om aw and size-adjus ed (wi hou he influence o size) da a se s.
Homo sapiens Pan oglody es Go illa go illa
PLS
pE-size SE
PLS
pE-size SE
PLS
pE-size SE
Raw
C /C1 0.206 0.856 0.999 0.012 0.257 0.488 0.143 0.010 0.552 0.001 4.324 0.012
C /C2 0.589 0.009 2.733 0.012 0.774 0.001 5.671 0.010 0.861 0.001 8.562 0.011
C /C3 0.537 0.021 2.278 0.011 0.575 0.001 3.609 0.010 0.716 0.001 6.182 0.011
C /C4 0.549 0.010 2.739 0.011 0.722 0.001 5.709 0.010 0.862 0.001 8.274 0.011
C /C5 0.533 0.018 2.319 0.011 0.707 0.001 5.388 0.010 0.880 0.001 8.792 0.011
C /C6 0.534 0.026 2.042 0.012 0.708 0.001 4.975 0.010 0.881 0.001 8.503 0.011
C /C7 0.566 0.008 2.745 0.011 0.712 0.001 5.246 0.010 0.866 0.001 8.036 0.011
Size-adjus ed
C /C1 0.364 0.612 0.318 0.014 0.433 0.190 0.894 0.012 0.620 0.004 3.363 0.013
C /C2 0.548 0.217 0.768 0.014 0.589 0.068 1.489 0.012 0.664 0.001 3.997 0.012
C /C3 0.413 0.678 0.514 0.013 0.511 0.045 1.885 0.010 0.571 0.006 2.787 0.012
C /C4 0.450 0.632 0.358 0.013 0.602 0.004 2.905 0.011 0.702 0.001 4.559 0.011
C /C5 0.446 0.446 0.102 0.013 0.516 0.090 1.376 0.011 0.704 0.001 4.636 0.011
C /C6 0.464 0.306 0.479 0.013 0.407 0.663 0.450 0.011 0.707 0.001 4.613 0.011
C /C7 0.434 0.499 0.075 0.013 0.421 0.526 0.106 0.011 0.687 0.001 4.347 0.011
Significan alues a e indica ed in bold (p- alue <0.05).
Figu e 5. Co ela ion analysis be ween he ce ical e eb ae and he skull om aw da a (le ) and size-adjus ed ( igh ) in Homo sapiens (g een), Pan oglody es (blue), and Go illa
go illa (ligh yellow). The nume ical alues a e ep esen ed in Table 4. The ce ical e eb ae ep esen ed in he figu e belong o an indi idual o he species P. oglody es.
Abb e ia ion:
PLS
¼Pa ial leas squa e co ela ion. (Fo in e p e a ion o he e e ences o colo in his figu e legend, he eade is e e ed o he Web e sion o his a icle.)
M. A legi, A. Pan oja-P
e ez, C. Veschamb e-Cou u e e al. Jou nal o Human E olu ion 162 (2022) 103112
8
esul s indica e ce ain di e ences in he c anioce ical co ela ion
pa e n be ween species ha p esen di e en pos u al and
locomo o epe oi es; howe e , we ha e no di ec ly es ed he
ela ionship be ween hem he e. Thus, o confi m ou hypo hesis
ha ce ain di e ences in pos u al and locomo o epe oi es
would be eflec ed in hei pa e ns o c anioce ical in eg a ion,
u he analysis wi h a mo e specific app oach is needed.
Allome y The esul s showed ha all he ce ical e eb ae,
excep he a las in chimpanzees and humans, exhibi ed a signifi-
can amoun o shape a ia ion explained by c anial size (Table 6).
Go illas yielded he la ges pe cen ages o a ia ion ollowed by
chimpanzees. Also, he C2eC7 e eb ae p esen ed posi i e
allome ic ends; in con as , he a las showed a nega i e end
(Fig. 7). Pai wise analyses yielded simila allome ic pa e n o
he h ee species in he fi s (a las) and las (C7) ce ical
e eb ae (Table 7). Humans di e om chimpanzees in he
C4eC5 e eb ae and om go illas in he C2eC6. Go illas and
chimpanzees di e in he allome ic pa e n in he C2 and C5
e eb ae.
3.3. TM analyses o size-adjus ed da a
Compa ed wi h he p e ious sec ion on aw da a, he size-
adjus ed da a gi e he e in o ma ion on shape wi hou he influ-
ence o size.
PLS in eg a ion analysis The esul s o he PLS analysis om size-
adjus ed da abases a e p esen ed in Table 4 and Figu e 5.In
compa ison wi h hose esul s ob ained om aw da a se s,
excep o he a las, all he ce ical e eb ae ob ained lowe
co a ia ion alues wi h he c anium. Go illas displayed significan
PLS
alues in all he co a ia ion analyses (i.e., c anium wi h
C1eC7) and chimpanzees in C3 and C4 e eb ae, and humans
did no show significan esul s in any c anioce ical co a ia ion
analyses (Table 4). In consequence, compa ed wi h he esul s
ob ained om he aw da a se , his esul s in a educ ion in he
di e ences in he alues o co a ia ion among e eb ae. Also,
pai wise compa isons e ealed ha once he size ac o is
emo ed, di e ences be ween e eb ae and species a e also
educed (SOM Tables S15e16). Finally, hese esul s pa ially
suppo he hypo hesis ha emo ing he e ec o size would
esul in a educ ion o he magni ude o in eg a ion. Con e sely
o he es o he e eb ae, he a las inc eased i s alues.
Be ween- a iables pai wise co ela ion es These esul s also
confi med his educ ion o he le els o co ela ion om size-
adjus ed da a se s compa ed wi h hose ob ained om aw da a
se s, excep in he a las o H. sapiens and P. oglody es (SOM
Table S16 and Fig. S1). Also, in gene al e ms, he highes
c anioce ical co ela ion alues a e no ela ed o a iables
ep esen ing maximum leng h, ei he midla e al (e.g., MaxT Di,
ST D, o M8) o do so en al (e.g., MaxDVDi, M5, o M13), bu
a iables ela ed o he e eb al o amen (SOM Table S17).
4. Discussion
4.1. C anium-ce ical co a ia ion: di e ences be ween species
The ce ical column o hominoids has been p oposed o be
highly in eg a ed wi h he c anial base, wi h no subs an ial di e -
ences be ween species (Villamil, 2018). Howe e , ou esul s indi-
ca e di e ences exis bo h in he pa e ns o co ela ion and
magni udes o in eg a ion o he ce ical e eb ae wi h he c a-
nium as a whole (Fig. 6). Among hominines, go illas display he
highes magni ude o c anioce ical in eg a ion in all he e eb ae,
ollowed by chimpanzees. I he magni ude o in eg a ion indica es
he deg ee o in e ac ion be ween he c anium and he ce ical
e eb ae, he pa e n e eals how his is p oduced (G abowski
e al., 2011). Ou esul s showed ha humans di e om bo h go-
illas and chimpanzees in he ai s ha link he c anium and he
ce ical spine. In bo h go illas and chimpanzees, he s onges
co ela ion occu s be ween a iables/ ai s ep esen ing p ogna-
hism and leng h o he c anial base and maximum leng h and
wid h o he e eb ae. In con as , in humans, his is p oduced
Table 5
Highes co ela ion alues be ween he c anium and ce ical a iables in Homo sapiens,Pan oglody es, and Go illa go illa om aw da a se s.
H. sapiens Highes co ela ion p2nd Highes co ela ion p
C /C1 Basion-P os hion/MaxD Di 0.251 0.113 Zygo-Zygo/ST D 0.237 0.148
C /C2 Zygo-Zygo/ST D 0.528 <0.001 Glabella-Inion/M1a 0.518 0.001
C /C3 Glabella-Inion/M5 0.513 <0.001 Opis hion-Basion/M11 0.470 0.001
C /C4 Glabella-Inion/M8 0.547 <0.001 Nasion-P os hion/ST D 0.536 <0.001
C /C5 Glabella-Inion/M5 0.549 <0.001 Opis hion-Basion/M11 0.525 <0.001
C /C6 Nasion-P os hion/M1 0.546 <0.001 Glabella-Inion/MaxT Di 0.466 0.002
C /C7 Glabella-Inion/ST D 0.538 <0.001 Nasion-P os hion/MaxT Di 0.516 <0.001
P. oglody es Highes co ela ion p2nd Highes co ela ion p
C /C1 Opis hion-Basion/M11 0.306 0.051 Opis hion-Basion/MaxT Di 0.281 0.076
C /C2 Basion-P os hion/MaxT Di 0.653 <0.001 Basion-P os hion/MaxD Di 0.648 <0.001
C /C3 Basion-P os hion/MaxT Di 0.647 <0.001 Opis hion-Basion/M10 0.516 <0.001
C /C4 Basion-P os hion/MaxD Di 0.672 <0.001 Nasion-Basion/M5 0.579 <0.001
C /C5 Basion-P os hion/MaxD Di 0.631 <0.001 Nasion-Basion/S D 0.600 <0.001
C /C6 Basion-P os hion/MaxD Di 0.674 <0.001 Nasion-Basion/M5 0.603 <0.001
C /C7 Basion-P os hion/MaxD Di 0.718 <0.001 Nasion-Basion/M5 0.659 <0.001
G. go illa Highes co ela ion p2nd Highes co ela ion p
C /C1 Glabella-Inion/MaxT Di 0.621 <0.001 Glabella-Inion/MaxD Di 0.586 <0.001
C /C2 Nasion-Basion/MaxT Di 0.868 <0.001 Nasion-Basion/MaxD Di 0.839 <0.001
C /C3 Nasion-Basion/MaxT Di 0.847 <0.001 Basion-P os hion/MaxT Di 0.798 <0.001
C /C4 Nasion-Basion/MaxT Di 0.846 <0.001 Basion-P os hion/MaxD Di 0.842 <0.001
C /C5 Basion-P os hion/MaxD Di 0.865 <0.001 Nasion-Basion/MaxD Di 0.858 <0.001
C /C6 Basion-P os hion/M13 0.862 <0.001 Nasion-Basion/MaxD Di 0.851 <0.001
C /C7 Nasion-Basion/ST D 0.852 <0.001 Basion-P os hion/MaxD Di 0.851 <0.001
Significan alues a e indica ed in bold (p- alue <0.05).
The e eb al dimensions ollow Figu e 6 and SOM Table S13.
M. A legi, A. Pan oja-P
e ez, C. Veschamb e-Cou u e e al. Jou nal o Human E olu ion 162 (2022) 103112
9
Lo ejoy, C.O., Sewa, G., Spu lock, L., As aw, B., Whi e, T.D., 2009b. The pel is and
emu o A dipi hecus amidus: The eme gence o up igh walking. Science 326,
71e1e71e6.
Lo ejoy, C.O., Simpson, S.W., Whi e, T.D., As aw, B., Suwa, G., 2009c. Ca e ul climbing
in he Miocene: The o elimbs o A dipi hecus amidus and humans a e p imi-
i e. Science 326, 70e70e8.
Lyce , S.J., Colla d, M., 2005. Do homologies impede phylogene ic analyses o he
ossil hominids? An assessmen based on ex an papionin c anioden al
mo phology. J. Hum. E ol. 49, 618e642.
Man eda, E., Mi e oecke , P., Books ein, F.L., Schae e , K., 2006. Func ional
mo phology o he fi s ce ical e eb a in humans and nonhuman p ima es.
Ana . Rec. 289B, 184e194.
Ma oig, G., Che e ud, J.M., 2001. A compa ison o pheno ypic a ia ion and
co a ia ion pa e ns and he ole o phylogeny, ecology, and on ogeny du ing
c anial e olu ion o New Wo ld monkeys. E olu ion 55, 2576e2600.
Ma oig, G., Shi ai, L.T., Po o, A., de Oli ei a, F.B., De Con o, V., 2009. The e olu ion
o modula i y in he mammalian skull II: E olu iona y consequences. E ol. Biol.
36, 136e148.
Ma in, R., Salle , K., 1957. Leh buch de An h opologie. Gus a Fische , S u ga .
McCown, T.D., Kei h, A., 1939. The S one Age o Moun Ca mel. The Fossil Human
Remains om he Le alloiso-Mous e ian. Cla endon P ess, Ox o d.
Melo, D., Ga cia, G., Hubbe, A., Assis, A.P., Ma oig, G., 2016. E olQR: An R package
o e olu iona y quan i a i e gene ics. F1000Resea ch 4, 1e25.
Meye , M.R., Woodwa d, C., Tims, A., Bas i , M., 2018. Neck unc ion in ea ly hom-
inins and suspenso y p ima es: Insigh s om he uncina e p ocess. Am. J. Phys.
An h opol. 166, 613e637.
Mi e oecke , P., Books ein, F., 2007. The concep ual and s a is ical ela ionship
be ween modula i y and mo phological in eg a ion. Sys . Biol. 56, 818e836.
Mülle , M.A., Me en, L.J., B€
ohme , C., Nyaka u a, J.A., 2021. Pushing he bounda y?
Tes ing he ‘ unc ional elonga ion hypo hesis’o he gi a e’s neck. E olu ion 75,
641e655.
Nalley, T.K., G ide -Po e , N., 2015. Func ional mo phology o he p ima e head and
neck. Am. J. Phys. An h opol. 156, 531e542.
Nalley, T.K., G ide -Po e , N., 2017. Func ional analyses o he p ima e uppe ce -
ical e eb al column. J. Hum. E ol. 107, 19e35.
Nalley, T.K., G ide -Po e , N., 2019. Ve eb al mo phology in ela ion o head pos u e
and locomo ion I: The ce ical spine. In: Been, E., G
omez-Oli encia, A., K ame , P.A.
(Eds.), Spinal E olu ion: Mo phology, Func ion, and Pa hology o he Spine in
Hominoid E olu ion. Sp inge In e na ional Publishing, Cham, pp. 35e50.
Na i a, Y., Ku a ani, S., 2005. E olu ion o he e eb al o mulae in mammals: A
pe spec i e on de elopmen al cons ain s. J. Exp. Zool. Pa B Mol. De . E ol.
304B, 91e106.
Olson, E.C., Mille , R.L., 1958. Mo phological In eg a ion. The Uni e si y o Chicago
P ess, Chicago.
Pablos, A., Ma ínez, I., Lo enzo, C., G acia, A., Sala, N., A suaga, J.L., 2013. Human
alus bones om he Middle Pleis ocene si e o Sima de los Huesos (Sie a de
A apue ca, Bu gos, Spain). J. Hum. E ol. 65, 79e92.
Pon ze , H., Holloway, J.H., Raichlen, D.A., Liebe man, D.E., 2009. Con ol and unc ion
o a m swing in human walking and unning. J. Exp. Biol. 212, 523e534.
Po o, A., de Oli ei a, F.B., Shi ai, L.T., De Con o, V., Ma oig, G., 2009. The e olu ion
o modula i y in he mammalian skull I: Mo phological in eg a ion pa e ns
and magni udes. E ol. Biol. 36, 118e135.
Po o, A., Shi ai, L.T., de Oli ei a, F.B., Ma oig, G., 2013. Size a ia ion, g ow h
s a egies, and he e olu ion o modula i y in he mammalian skull. E olu ion
67, 3305e3332.
Powell, V., Es e e-Al a a, B., Molna , J., Villmoa e, B., Pe i , A., Diogo, R., 2018.
P ima e modula i y and e olu ion: Fi s ana omical ne wo k analysis o p ima e
head and neck musculoskele al sys em. Sci. Rep. 8, 1e10.
P euscho , H., 2004. Mechanisms o he acquisi ion o habi ual bipedali y: A e
he e biomechanical easons o he acquisi ion o up igh bipedal pos u e?
J. Ana . 204, 363e384.
P euscho , H., Klein, N., 2013. To sion and bending in he neck and ail o sau opod
dinosau s and he unc ion o ce ical ibs: Insigh s om unc ional
mo phology and biomechanics. PLoS One 8, e78574.
P ofico, A., Pi as, P., Buzi, C., Di Vincenzo, F., La a ini, F., Melchionna, M.,
Veneziano, A., Pasquale, R., Manzi, G., 2017. The e olu ion o c anial base and
ace in Ce copi hecoidea and Hominoidea: Modula i y and mo phological
in eg a ion. Am. J. P ima ol. 79, e22721.
R Co e Team, 2020. R: A language and en i onmen o s a is ical compu ing. R
Founda ion o S a is ical Compu ing, Vienna, Aus ia. URL. h ps://www.R-
p ojec .o g/.
Randau, M., Cu , A.R., Hu chinson, J.R., Pie ce, S.E., Goswami, A., 2017. Regional
di e en ia ion o elid e eb al column e olu ion: A s udy o 3D shape a-
jec o ies. O g. Di e s. E ol. 17, 305e319.
Robinson, J.T., 1972. Ea ly Hominid Pos u e and Locomo ion. The Uni e si y o
Chicago P ess, Chicago.
Rohl , F.J., Slice, D., 1990. Ex ensions o he P oc us es me hod o he op imal su-
pe imposi ion o landma ks. Sys . Biol. 39, 40e59.
Rohl , F.J., Co i, M., 2000. Use o wo-block pa ial leas -squa es o s udy co a ia-
ion in shape. Sys . Biol. 49, 740e753.
Ryan, T.M., Sukhdeo, S., 2016. KSD-VP-1/1: Analysis o he pos c anial skele on
using high- esolu ion compu ed omog aphy. In: Haile-Selassie, Y., Su, D.F.
(Eds.), The Pos c anial Ana omy o Aus alopi hecus a a ensis. Sp inge , Do -
d ech , pp. 39e62.
Ryan, T.M., Ca lson, K.J., Go don, A.D., Jablonski, N., Shaw, C.N., S ock, J.T., 2018.
Human-like hip join loading in Aus alopi hecus a icanus and Pa an h opus
obus us. J. Hum. E ol. 121, 12e24.
Schlu e , D., 1996. Adap i e adia ion along gene ic lines o leas esis ance. E o-
lu ion 50, 1766e1774.
Schul z, A.H., 1942. Condi ions o balancing he head in p ima es. Am. J. Phys.
An h opol. 29, 483e497.
Schul z, A.H., 1961. P ima ologia. Handbuch de p ima enkunde. S. Ka ge , Basel.
Sidlauskas, B., 2008. Con inuous and a es ed mo phological di e sifica ion in sis e
clades o cha aci o m fishes: A phylomo phospace app oach. E olu ion 62,
3135e3156.
Slijpe , E., 1946. Compa a i e biologic ana omical in es iga ions on he e eb al
column and spinal muscula u e o mammals. Tweede Sec . 17, 1e128.
Sockol, M.D., Raichlen, D.A., Pon ze , H., 2007. Chimpanzee locomo o ene ge ics
and he o igin o human bipedalism. P oc. Na l. Acad. Sci. USA 104,
12265e12269.
S eppan, S.J., Phillips, P.C., Houle, D., 2002. Compa a i e quan i a i e gene ics:
E olu ion o he G ma ix. T ends Ecol. E ol. 17, 320e327.
S e n J ., J.T., 2000. Climbing o he op: A pe sonal memoi o Aus alopi hecus
a a ensis. E ol. An h opol. 9, 113e133.
S ai , D.S., 2001. In eg a ion, phylogeny, and he hominid c anial base. Am. J. Phys.
An h opol. 114, 273e297.
S ai , D.S., Ross, C.F., 1999. Kinema ic da a on p ima e head and neck pos u e:
Implica ions o he e olu ion o basic anial flexion and an e alua ion o
egis a ion planes used in paleoan h opology. Am. J. Phys. An h opol. 108,
205e222.
Toe ien, M.J., 1957. No e on he ce ical e eb ae o he La Chapelle man. S. A . J.
Sci. 53, 447e449.
Toe ien, M.J., 1961. The leng h and inclina ion o he p ima e ce ical spinous
p ocesses. T ans. R. Soc. Sou h A ica 36, 95e105.
Vande Linden, A., Campbell, K.M., B ya , E.K., San ana, S.E., 2019a. Head- u ning
mo phologies: E olu ion o shape di e si y in he mammalian a laseaxis
complex. E olu ion 73, 2060e2071.
Vande Linden, A., Hed ick, B.P., Kamila , J.M., Dumon , E.R., 2019b. A las
mo phology, scaling and locomo o beha iou in p ima es, oden s and ela-
i es (Mammalia: Eua chon ogli es). Zool. J. Linnean Soc. 185, 283e299.
Va ela-Lashe as, I., Bakke , A.J., an de Mije, S.D., Me z, J.A., an Alphen, J., Galis, F.,
2011. B eaking e olu iona y and pleio opic cons ain s in mammals: On slo hs,
mana ees and homeo ic mu a ions. E oDe o 2, 11.
Va
on-Gonz
alez, C., Whelan, S., Klingenbe g, C.P., 2020. Es ima ing phylogenies
om shape and simila mul idimensional da a: Why i is no eliable. Sys . Biol.
69, 863e883.
Villamil, C.I., 2018. Pheno ypic in eg a ion o he ce ical e eb ae in he Homi-
noidea (P ima es). E olu ion 72, 490e517.
Villamil, C.I., 2021. The ole o de elopmen al a e, body size, and posi ional
beha io in he e olu ion o co a ia ion and e ol abili y in he c anium o
s epsi hines and ca a hines. J. Hum. E ol. 151, 102941.
Vidal, P.P., G a , W., Be hoz, A., 1986. The o ien a ion o he ce ical e eb al
column in un es ained awake animals. Exp. B ain Res. 61, 549e559.
on C amon-Taubadel, N., Smi h, H.F., 2012. The ela i e cong uence o c anial and
gene ic es ima es o hominoid axon ela ionships: Implica ions o he
econs uc ion o hominin phylogeny. J. Hum. E ol. 62, 640e653.
Wagne , G.P., 1988. The influence o a ia ion and o de elopmen al cons ain s on
he a e o mul i a ia e pheno ypic e olu ion. J. E ol. Biol. 1, 45e66.
Wagne , G.P., 1996. Homologues, na u al kinds and he e olu ion o modula i y. Am.
Zool. 36, 36e43.
Wagne , G.P., Schwenk, K., 2000. E olu iona ily s able configu a ions: Func ional
in eg a ion and he e olu ion o pheno ypic s abili y. E ol. Biol. 31, 155e217.
Wa ene , A.G., Lew on, K.L., Pon ze , H., Liebe man, D.E., 2015. A wide pel is does
no inc ease locomo o cos in humans, wi h implica ions o he e olu ion o
childbi h. PLoS One 10, e0118903.
Wa anabe, A., Fab e, A.C., Felice, R.N., Maisano, J.A., Mülle , J., He el, A.,
Goswami, A., 2019. Ecomo phological di e sifica ion in squama es om
conse ed pa e n o c anial in eg a ion. P oc. Na l. Acad. Sci. USA 116,
14688e14697.
Wellik, D.M., Capecchi, M.R., 2003. Hox10 and Hox11 genes a e equi ed o globally
pa e n he mammalian skele on. Science 301, 363e367.
Whi e, A.A., Panjabi, M.M., 1990. Clinical Biomechanics o he Spine. J. B. Lippinco ,
Philadelphia.
Whi come, K.K., Shapi o, L.J., Liebe man, D.E., 2007. Fe al load and he e olu ion o
lumba lo dosis in bipedal hominins. Na u e 450, 1075e1078.
Zeldi ch, M.L., 1988. On ogene ic a ia ion in pa e ns o pheno ypic in eg a ion in
he labo a o y a . E olu ion 42, 28e41.
Zubai , H.N., Chu, K.M., Johnson, J.L., Ri e s, T.J., Belooze o a, I.N., 2019. Gaze
coo dina ion wi h s ides du ing walking in he ca . J. Physiol. 597,
5195e5229.
M. A legi, A. Pan oja-P
e ez, C. Veschamb e-Cou u e e al. Jou nal o Human E olu ion 162 (2022) 103112
16
