Intragroup variation in the Pre-Columbian Cuba population: A perspective from cranial morphology
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Pre-Columbian Cuba population cranial morphology
Taisiya Syutkina et al.
Anthropological Review • Vol. 84(3), 233–255 (2021)
Intragroup variation in the Pre-Columbian
Cuba population: A perspective from cranial
morphology
Taisiya Syutkina1, Mario Juan Gordillo Pérez2,
Silvia Teresita Hernández Godoy3,4, Carlos Arredondo Antúnez5,
Armando Rangel Rivero6
1
Center for Human Ecology, the Russian Academy of Sciences N.N. Miklouho-Maklay
Institute of Ethnology and Anthropology
2
Department of Biology, University of Oriente - Santiago de Cuba
3
Grupo de Investigación y Desarrollo de la Dirección Provincial de Cultura de Matanzas
4
Departamento de Estudios Socioculturales, Universidad de Matanzas
5
Montané Anthropological Museum, Biological Faculty, University of Havana
6
Montané Anthropological Museum, Biological Faculty, University of Havana
Abstract: The paper aims to study intragroup variation inside the two pre-Columbian Cuban populations:
the aceramic Archaic and the ceramic Taino groups, based on their cranial morphology. The latter applied
artificial cranial deformation to all its members, so the groups are referred to as “non-deformed” and
“deformed” samples here. Studies across different disciplines suggest evidence of cultural and biological
diversity inside the non-deformed group, while local variations of applying the deforming device can be
responsible for shape variation across the deformed group. Cranial metrics and non-metric cranial traits of
the 92 crania of Cuban origin were analyzed, although the sample size varied between the analyses due to
the incompleteness of the crania. Geometric morphometrics was applied to the deformed crania to study
the shape variation across the sample. Three deformed crania from the Dominican Republic were ana-
lyzed together with the deformed Cuban sample to test the variability of the practice between the islands.
Principal component analysis and the Mantel test did not reveal any geographic differences in the cranial
metric traits. No morphological differences associated with the antiquity of materials could be seen either
based on the available data. The principal component analysis of the Procrustes coordinates of the cranial
vault outline in the lateral norm revealed continuous variability of cranial shapes from the ones with more
flattened frontal and occipital bones to the more curved outlines, which is probably explained by individual
variation. Non-metric traits variation revealed bilateral asymmetry in the expression of the occipito-mas-
toidal ossicles among the deformed crania. In conclusion, the study did not support assumptions about
morphological diversity inside the studied samples or proved the impossibility of available craniological
data to reflect possible intragroup differentiation at the moment.
Key words: biological anthropology, cranial metrics, non-metric traits, geometric morphometrics, anthro-
pology of Cuba, artificial cranial deformation, The Greater Antilles
Original Research Article Received: June 25, 2021; Revised: August 23, 2021; Accepted: August 27, 2021
DOI: 10.2478/anre-2021-0021
© 2021 Polish Anthropological Society
234 Taisiya Syutkina et al.
Introduction ticular type requires both deformed and
non-deformed crania from the same pop-
Prior to European arrival, Cuba is known ulation, and the results are only valid for
to have been inhabited by an indigenous the population in question (Rhode and
population, which is generally divided Arriaza 2006). Possible intragroup varia-
into two groups: the aceramic Archaic tion can be explained by multiple origins,
population, which is believed to have ar- microevolutionary processes, different
rived around 5000 years ago (Napolitano lifestyles for the non-deformed sample
et al. 2019:7–8) and the ceramic agricul- and by local traditions in applying the
tural Taino population, which inhabited deforming apparatus for the deformed
the island since approximately 500 AD sample. Apart from the spatial dimen-
(Torres Etayo 2006:35–36) until deci- sion, there is a temporal one, which
mation by the newcomers from the Old could reflect possible changes through-
World. out the long history of the island popu-
There has been much variation in the lation. The assumption that these factors
ways the aboriginal population of Cuba might have impacted the morphology of
is referred to. To avoid confusion caused the studied groups follows from the re-
by changing perspectives due to new ar- sults of numerous studies across differ-
chaeological, linguistic and biological ev- ent disciplines.
idence, we will stick to strictly morpho- Two recent genetic studies of the
logical criteria of “non-deformed” and pre-contact Caribbean populations, in-
“deformed” crania, which are associated cluding the Cuban ones (Nägele et al.
with the aceramic (Archaic) and ceramic 2020; Fernandes et al. 2021), confirmed
(Taino) population groups respectively. a single genetic profile for Ceramic-age
This study focuses on the intragroup (referred to as “deformed” in this study)
variation of the Cuban pre-Columbian individuals, which suggests a single mi-
population basing on cranial morphol- gration or successive migrations from
ogy of the deformed and non-deformed the same region for this group. Howev-
samples. Between-group variation is not er, the studies produced different results
being addressed, and the samples are an- considering the homogeneity of the Ar-
alyzed separately for two reasons. First, chaic (“non-deformed”) population. This
there has been extensive evidence that inconsistency is most likely explained by
the two groups did not have recent com- different sampling: the Cueva del Perico
mon genetic ancestry, so comparing their individual, shown to have different (pre-
cranial morphologies would not yield sumably North American) ancestry by
new information. Second, a large body the work of Nägele et al., was not includ-
of evidence suggests that artificial cranial ed in the second genetic study, neither
deformation affects facial and basal areas was it available for the present work.
of the cranium (Antón 1989; Cheverud An extensive odothological study has
et al. 1992; Friess and Baylac 2003), so identified two migratory waves that led
the direct comparison of deformed and to populating the Circum-Caribbean re-
non-deformed crania of different popula- gion, one connected with the aceramic
tions would have produced spurious re- Archaic groups, the other one bringing
sults. Detection of the regions that stay the ceramic Taíno population (Coppa et
unaffected by the deformation of a par- al. 2008). The deformed Cuban sample
Pre-Columbian Cuba population cranial morphology 235
in the study was represented by the sam- to a more differentiated approach to the
ple coming from a single cemetery, Chor- diversity of pre-Columbian communities
ro de Maíta, while the non-deformed in Cuba.
samples came from multiple sites and Archaeological data also suggested a
still clustered together according to their possible division of the non-deformed
dental morphology. This finding is espe- population into two periods: hunt-
cially remarkable as one of the samples er-gathering and protoagricultural (ac-
analyzed by Coppa et al. came from Cue- tual names may vary from author to au-
va del Perico, an outlier in the genetic thor), the latter one characterized by the
study discussed above. evidence of simple agriculture and crude
Archaeological data suggest complex ceramics at the corresponding sites
population interconnections within the (Tabío 1988:64–65; Guarch Delmon-
island at different periods (Cooper 2007; te 1990; Pérez Carratalá 2014:78–80).
Chinique de Armas et al. 2020), while, It has also been shown based on stable
on the other hand, some sites stayed isotope analysis that subsistence strat-
isolated from the surrounding groups egies differed throughout the territory
throughout long periods. For example, of the island even between contempora-
Canímar Abajo in Matanzas (populat- neous communities: groups inhabiting
ed by the non-deformed group) bears Canímar Abajo site have been shown to
evidence of long-standing cultural tra- rely on horticulture while Cueva Calero
ditions not present at other contempo- and Cueva del Perico inhabitants appar-
raneous Cuban sites (Alarie and Rock- ently relied on hunting-fishing-gather-
sandic 2016). This long isolation could ing strategy (Chinique de Armas et al.
have formed the basis for possible mi- 2018).
croevolutionary processes within some Classification of the archaeological
communities, which might have affected and skeletal materials is also an essential
their cranial morphology. issue for discussing the groups’ homo- or
Previously the non-deformed group heterogeneity. Cooper (2007:232) noted
was often divided into two archaeologi- that ceramic vessel fragments had been
cal complexes (Herrera Fritot 1964; Po- found at some sites classified as pre-agri-
spisil and Rivero de la Calle 1964; Tabio cultural in the Western part of the island,
and Rey 1966): Complex I and Complex and the cultural attribution relied upon
II (sometimes called as Guayabo Blan- the absence of artificial cranial deforma-
co and Cayo Redondo or Guanahatabey tion in the human remains. He argues
and Siboney). The division was based on that it challenges the whole system of
mortuary practices registered at two dif- classification of the findings, which au-
ferent sites, but some researchers tend- tomatically considers any non-deformed
ed to see two biologically distinct pop- crania belonging to the pre-agricultural
ulations behind them (Alonso Alonso population. This, in turn, means that one
1995:99). Later a more profound consid- has no reason to assume that all non-de-
eration of ancient cultures made for un- formed crania belonged to a morpholog-
derstanding that the interplay between ically, genetically, or culturally uniform
the groups that occupied different sites population.
was more complex than a simple dichot- As far as the deformed sample is con-
omy of two cultural variants, which led cerned, Herrera Fritot (1964:105) noted
236 Taisiya Syutkina et al.
that the deformed crania could be divid- Cranial samples from Cuba avail-
ed into two groups: those with moder- able for this study suffer from several
ate deformation concentrated mainly in problems: most of them lack contextu-
the frontal and basal occipital areas and al information, zone of origin is often
those with more pronounced deforma- known approximately, secure dating is
tion which affected the whole vault and only available for several sites. Moreover,
in most cases produced a bilobate shape. the humid climate does not favor the
He deduced that there must have existed long conservation of the bone remains,
different designs of deforming devices, so their state of preservation is often far
some of them having a sagittal band in from perfect. With that in mind, this ar-
addition to the transverse ones. ticle addresses the following questions:
The results of all the studies men- (1) Did possible multiple origins or mi-
tioned above suggest grounds for ex- croevolutionary processes contribute to
pecting a certain degree of heterogeneity morphologically detectable intragroup
both inside the deformed and non-de- differentiation inside the non-deformed
formed samples. There is evidence that sample? (2) Did local differences in ar-
some local non-deformed groups relied tificial cranial deformation contribute
upon different subsistence strategies, to intragroup differentiation inside the
were culturally diverse and had lived in deformed sample? and (3) Can analysis
isolation for a long time (Chinique de of the available craniological data con-
Armas and Rocksandic 2019). Consid- tribute to our understanding of the rela-
ering that they had been inhabiting the tionships between different groups that
island for several millennia, these factors inhabited the island before European
might have led to significant differences colonization?
between the local or temporal groups. Cranial morphology is widely used
Given that the non-deformed sample can to address prehistorical issues due to
include individuals with different ances- the conservative nature of its variation,
try and that there is no certainty about its underlying genetic component, and
them being a single group, a study of strong geographic patterning (Bunak
intragroup morphological variability can 1959; Howells 1989; Pietrusewsky 2014).
provide insights into the anthropological Metric and non-metric cranial traits were
structure of the island’s early popula- used in this study to explore the variabil-
tion. While not much genetic-based in- ity of cranial morphology in each sample.
tragroup morphological variation can be For the deformed group, two-dimension-
expected for the deformed sample, little al geometric morphometrics was applied
is known about the deforming practice to study and visualize the shape variation
itself and whether it was performed in across the sample. 2D shape analysis is
the same manner throughout the whole a robust tool for studying morphologi-
island. The objective of the present study cal variation but must be used with cau-
is to analyze the intragroup variation in tion with 3D objects due to the so-called
cranial morphology of the two pre-Co- “three-to-two-dimension error” (Cardini
lumbian Cuban samples and see wheth- 2014; Buser et al. 2018), and its perfor-
er individual differentiation inside them mance has been shown to deteriorate
correlates with territorial, cultural, or when applied to small samples belong-
temporal factors. ing to one taxon (Cardini and ChiapelliPre-Columbian Cuba population cranial morphology 237
2020). However, it has been proven ef- de la Calle (1964) questioned this idea:
fective for the studies of artificial cranial according to them, it was difficult to sep-
deformation (Perez 2007; Gonzalez et al. arate the first two complexes both by ar-
2011) and thus will be only applied to chaeological and anthropological data. A
the deformed group in this paper. direct comparison between the crania did
not reveal any differences, so it was sug-
A review of previous cranial gested that the preceding authors might
have come to an opposite conclusion due
studies of the Cuban pre- to the inclusion of both male and female
Columbian population crania in their samples.
Soviet anthropologist V. Ginzburg,
Various scholars attempted to study the whose measurements of the Cuban cra-
internal structure of the Cuban pre-Co- nia were included in the present paper,
lumbian population basing on cranial concluded that the non-deformed and
data, which has always been scarce. The deformed groups were not directly re-
first scholar to study ancient human re- lated, although both must have had
mains was Luis Montané Dardé, who South-American biological affinities. He
studied the non-deformed crania from also suggested two morphological sub-
the site of Boca del Purial and noted their types of the non-deformed crania, one
anthropological heterogeneity (Mon- being wider- and lower-faced than the
tané 1908 cited in Hernández Godoy other (Ginzburg 1967:189). His conclu-
2003:192). Later studies were conducted sions were disputed by V. Alexeev, who
by Rivero de la Calle, who examined the suggested that the differences between
deformed crania and identified this type the groups could be entirely explained
of deformation as “tabular oblique” ac- by the effect of cranial deformation
cording to Dembo and Imbelloni’s clas- (Alexeev 1986:20). By studying the vari-
sification (1938) and detected one crani- ation of non-metric traits in the two pop-
um that had “tabular erect” type, which ulations, Rivero de la Calle (2009:179–
he explained as a possible defect of ap- 180) concluded that their divergence is
plication of the deforming device (1949 greater than the one between compara-
cited in Herrera Fritot 1964:82–83). tive samples from distant locations of
One of the first works summarizing all the American continent (like Peru and
the known cranial data was made by René British Columbia). Recent genetic stud-
Herrera Fritot (1964), who developed ies reaffirmed that both groups had se-
the craniotrigonometric method and ap- curely different origins (Lalueza-Fox et
plied it to the study of Cuban aboriginal al. 2003; Nägele et al. 2020; Fernandes
crania. He investigated them through the et al. 2021).
prism of archaeological classification of After that, the focus of major archae-
the three archaeological “complexes”– ological and anthropological works based
Complex I or Guayabo Blanco (non-de- on Cuban pre-Columbian materials
formed), Complex II or Cayo Redondo shifted from studying the population in
(non-deformed), and Complex III or general to studying local cultural, ecolog-
Taíno (deformed), and saw individuals ical, nutritional, health, and other issues
from the respective sites as characteris- (Crespo-Torres et al. 2013:437–439).
tic for these “types”. Pospisil and Rivero This turned to be a promising approach;238 Taisiya Syutkina et al.
biological anthropology managed to shed the pre-Columbian Cuban population us-
new light on mortuary (La Rosa Cor- ing a multimethodological approach and
zo 2003), weaning (Chiniqie de Armas taking into account findings reported in
2017) and cultural practices (Alarie and recent studies from other disciplines and
Roksandic 2016), subsistence strategies new perspectives on the complex rela-
(Chinique de Armas et al. 2015), nature tions of the groups inhabiting the island.
of co-existence with the Europeans at
the later sites (Valcárcel Rojas 2016), pa- Material and methods
thologies (Armstrong et al. 2013).
Some recent attempts to approach in- Data acquisition. The total craniometric
tragroup diversity based on morpholog- sample consists of 95 crania: 92 male and
ical data were proposed by students of female crania of Cuban origin and three
The University of Havana. Bolufé Torres deformed male crania from the Domin-
(2015) found differences in the orbital ican Republic to test the variability of
breadth and stature between non-de- cranial morphology between the islands
formed individuals from different loca- (see Table 1 for more information on the
tions in the West of the island. Although sites and Supplement Table 1 for the full
the author concluded that the results composition of the samples). The Cuban
might indicate the different origin of the sample is comprised of the 57 non-de-
individuals, they should be treated with formed (male = 40, female = 17) and
caution due to the small sample sizes 39 deformed (male = 27, female = 12)
available. Valdés Pi (2009) managed to crania. However, not all of them could
compose a comparatively large sample of be included in all the multivariate anal-
both non-deformed and deformed crania. yses due to their incompleteness. The
A set of cluster analyses based on facial sample size for each analysis is reported
craniometric variables identified possi- separately in this section. Cranial mea-
ble sub-clusters both inside the non-de- surements have been standardized by
formed and deformed groups, but no both the Biometrika school and Martin
patterns (geographical, cultural, chrono- and Saller (1957) in the version applied
logical) that could explain the identified in Soviet/Russian craniometry (Alexeev
division were suggested. García Méndez and Debets 1964). The craniometric
carried out a geometric morphometric sample is composed of two parts: crania
analysis of the outlines characterizing measured by Vulf Ginzburg (n = 18) in
the cranial shapes in the frontal and lat- 1964 (Ginzburg 1967) and crania mea-
eral norms in a sample of non-deformed sured by one of the authors (TS) in
individuals from the East and the West 2017 following the same methodology
of the country. She concluded there were (n = 77). Ginzburg’s sample consists
no significant differences between them of the crania that could not be found or
(García 2018; García et al. in press). Fi- accessed in 2017 and only contains cra-
nally, preliminary analyses of the metric niometric variables. Non-metric traits
data studied here have been previously were observed by TS and MGP for the
published in Russian by the first author same 77 individuals that were measured
(Syutkina 2017; 2018). by TS. The traits used in the study are a
This paper is the first attempt to an- combined list of traits described in Berry
alyze a comprehensive cranial sample of and Berry (1967), Krenzer (2006), andPre-Columbian Cuba population cranial morphology 239
Fig. 1. Map showing location of the sites discussed in the article. Generated with the use of raster package
in R (Hijmans 2021). The sites from the zone of Baracoa are represented as a single point, more infor-
mation on exact origins is available in the Supplement Table 1
Table 1. Cranial samples from Cuba and Dominican Republic
Site Male Female Total
Non-deformed Unknown 2 0 2
Baracoa 16 7 23
Punta del Este 1 0 1
Pinar del Río 1 0 1
Cueva Calero, Matanzas 0 3 3
Boca del Purial, Sancti Spíritus 1 3 4
Canímar Abajo (Younger Cemetery), Matanzas 10 1 11
La Santa, Habana 0 1 1
Cueva Florencio, Matanzas 1 0 1
Ciénaga de Zapata, Matanzas 2 0 2
Guayabo Blanco, Matanzas 2 0 2
Puerto de Santa María, Camagüey 3 1 4
Laguna del Tesoro, Matanzas 0 1 1
Total non-deformed 39 17 56
Deformed Unknown 3 3 6
Baracoa 18 6 24
Guardalavaca 1 0 1
Cayo Salinas, Sancti Spíritus 2 1 3
Banes 0 2 2
Cueva de Yasica, Dominican Republic 1 0 1
La Caleta, Dominican Republic 2 0 2
Total deformed 27 12 39240 Taisiya Syutkina et al.
Movsesyan (2005) (see Table 2 for the era, skulls placed in lateral view, camera
full list and references). All traits were positioned at 30 cm from the auriculare
registered per side to study the symme- point. Two landmarks (the deepest point
try of their expression and per cranium on the frontal bone after glabella and the
to calculate their frequencies in each intersection of the line continuing the di-
group (Czarnetzki 1971). Only adult cra- rection of the zygomatic process of the
nia were included (based on the fusion of temporal bone with the outline of the
the spheno-occipital synchondrosis). Sex skull in its posterior part) and 23 semi-
and age for the 77 crania were estimat- landmarks between them were placed
ed according to general methodological along the sagittal curve (Figure 2). These
guidelines (Buikstra and Ubelaker 1994). landmarks were chosen to analyze the
The shape of the deformed crania outline avoiding the glabella and mastoid
in the lateral norm was studied using regions, which would otherwise add un-
two-dimensional geometric morpho- desirable sex-related variation. The land-
metrics, following standard procedures marks were placed, and their coordinates
(Bookstein 1991; Zelditch and Swiderski were obtained using tpsDIG2 2.31 soft-
2018; Vasiliev et al. 2018). Images were ware (Rohlf 2015). The semilandmarks
taken with a Nikon Coolpix P100 cam- were then slid according to the minimum
Table 2. Non-metric traits used in the study and their frequencies in the non-deformed and deformed
samples
Trait Reference Non-deformed n (N) Deformed n (N)
Metopism Berry and Berry 1967 0 (40) 0 (33)
Saggital ossicles Krenzer 2006 0 (29) 0.129 (31)
Ossicle at the lambda Berry and Berry 1967 0.071 (28) 0.188 (32)
Inca bone Krenzer 2006 0 (28) 0.125 (32)
Palatine torus Berry and Berry 1967 0 (33) 0.103 (29)
Supraorbital foramen Krenzer 2006 0.538 (39) 0.406 (32)
Supraorbital notch Krenzer 2006 0.795 (39) 0.906 (32)
Accessory infraorbital foramen Berry and Berry 1967 0.216 (37) 0.133 (30)
Infraorbital suture Krenzer 2006 0.649 (37) 0.600 (30)
Trochlear spine Movsesyan 2005 0.091 (33) 0.000 (28)
Multiple zygomatic foramina Krenzer 2006 0.784 (37) 0.607 (28)
Stellate pterion Movsesyan 2005, Murphy 1956 0.029 (35) 0.034 (29)
Fronto-temporal articulation Berry and Berry 1967 0.000 (35) 0.034 (29)
Epipteric bone Berry and Berry 1967 0.000 (35) 0.207 (29)
Coronal ossicles Berry and Berry 1967 0.057 (35) 0.032 (31)
Lambdoid ossicles Berry and Berry 1967 0.148 (27) 0.533 (30)
Occipito-mastoidal ossicles Krenzer 2006 0.348 (23) 0.533 (30)
Ossicle at the asterion Berry and Berry 1967 0.074 (27) 0.367 (30)
Squamoparietal ossicles Krenzer 2006 0.032 (31) 0.414 (29)
Parietal foramen Berry and Berry 1967 0.767 (30) 0.793 (29)
Foramen of Huschke Berry and Berry 1967 0.414 (24) 0.310 (29)
Posterior condylar canal Berry and Berry 1967 0.962 (26) 0.880 (25)
Condylar facet double Berry and Berry 1967 0.188 (16) 0 (20)Pre-Columbian Cuba population cranial morphology 241
male and female samples and compared
with average world figures (Alekseev and
Debets 1964:123–25) where available.
SSD was only calculated for traits that
could be measured in no less than ten
individuals in the smaller female sample
(Evteev 2008:9).
Principal component analysis was
performed to study the patterns of mor-
phological variation within each sam-
ple. Metric variables were used for the
non-deformed sample: the set of variables
Fig. 2. Landmarks (squares) and semilandmarks
to be included was composed in a manner
(circles) used in the study that would allow maximizing the sample
size and cumulative proportion of the
bending energy criterion, and both land- variance explained by the first two PCs.
marks and semilandmarks were aligned For the same reason, neurocranial and
using a generalized Procrustes analysis facial variables were studied separately
(Bookstein 1991). for the non-deformed sample, as com-
bining them led to a drastic decrease of
Statistical analyses the sample size to a very small number of
individuals. Seven facial measurements,
All craniometric values were compared summarizing overall facial, nasal, and or-
with the worldwide variation of the bital dimensions (FMB, ZMB, UFH, NLB,
studied traits and categorized as “very NLH, OBH, OBB) and ten measurements
small”, “small”, “medium’, “large”, describing cranial vault (GOL, XCB, BBH,
“very large” (Alexeev and Debets AUB, FRA, PAA, OCA, FRC, PAC, OCC)
1964:114–22). The data were tested for were included in the principal compo-
outliers using boxplots (not presented). nent analysis. Only the male sample was
For the non-deformed crania, descriptive analyzed to prevent sexual dimorphism
statistics were calculated with and with- from affecting the result, and the female
out outliers. In the deformed sample, sample was too small to be analyzed in-
the outlying values cannot be regarded dependently. 24 and 14 individuals could
as random deviations, and the variabil- be analyzed in the facial and neurocranial
ity should be considered the result of analyses, respectively.
artificial deformation. Only individual The deformed crania were analyzed
number 1047, which is believed to have using Procrustes coordinates to study
a different type of deformation, was ex- the shape variation across the sample;
cluded prior to the repeated calculation male and female samples were pooled for
to check whether it was responsible for this analysis. It included 34 crania with
extreme variability of some traits in the preserved cranial vaults. Individual num-
male sample. ber 1047 from Baracoa was considered
Sexual size dimorphism (SSD) for all an outlier and excluded from the sample
the traits was calculated as the difference for this analysis for its markedly different
between the mean value of a trait in the type of deformation.242 Taisiya Syutkina et al.
Additionally, a two-way Mantel test ed with Asian ancestry. The standard
(Mantel 1967; Sokal and Rohlf 1995) deviation of most variables in the male
was used to test the association between sample falls within average standard de-
the matrices of Euclidean craniometric viations for given morphological traits
and geographical distances for the male except WFB, XFB, ASB, ALL, ALB, FCD
part of the non-deformed sample. As in and ALA. In the female sample, these are
previous analyses, morphological vari- ASB, ZMA, XFB, FCD, ALL and ALB.
ables were divided into two subsets to However, after identifying and exclud-
avoid a drastic reduction of the sample. ing outliers (one for each variable), only
The neurocranial subset consisted of six FCD and alveolar arch dimensions stayed
variables (GOL, XCB, WFB, XFB, AUB, highly variable within both samples.
ASB) and 20 individuals, while the facial Despite the female’s sample higher
subset – of seven variables (FMB, ZMB, variability and lower representativeness,
UFH, OBB, OBH, NLB, NLH) and 23 all the values fall in the same general cat-
individuals. egories (small/medium) as in the male
All calculations were performed in sample. The sexual size dimorphism was
R version 4.0.3 (R Core Team 2020) found to be low among the non-deformed
with RStudio interface version 1.4.1103 crania: only five traits (XFB, PAA, ZMB,
(RStudio Team 2020). The Mantel test WNB and FCD) proved to have higher
was carried out in R Studio using the than average rates of SSD in the sample.
function mantel.rtest from package ade4
(Thioulouse et al. 2018), the number of Deformed sample
permutations was set at 5000. The geo- In general, both male and female crania
graphic distances calculations were per- from the deformed sample (Table 4) have
formed in R using the haversine method short, very low, and very wide vaults; the
from the package “geodist” (Padgham frontal bone is flat and wide, it is steeply
and Sumner 2021). The geometric mor- inclined and often bears a mark left by
phometric analysis was carried out by the deforming device. Forehead profile
means of geomorph package in R (Ad- angles values are far beyond the lowest
ams et al. 2021). average limits for physiologically normal
crania. All neurocranial and facial widths
Results are very large, so are orbital and facial
dimensions (although orbital and nasal
Metric variation widths tend to be smaller in the female
sample compared to female worldwide
Summary statistics average values). Horizontal profiling of
Non-deformed sample the face is moderate both at nasomalar
Both male and female non-deformed cra- and zygomaxillary levels.
nia (Table 3) are characterized by medi- The male sample proved to be very
um-small facial and neurocranial dimen- heterogeneous: standard deviations of
sions on a worldwide scale (Alekseev more than a third of variables exceed
and Debets 1964:114–122), including the world average: XCB, BBH, BPL, XFB,
zygomaxillary and nasomalar facial an- PAA, OCA, FRC, PAC, OCC, UFH, ALL,
gles, thus suggesting limited expression FCD, FAN and FAG. Testing for outliers
of cranial metric traits usually associat- revealed numerous outlying values forPre-Columbian Cuba population cranial morphology 243
the same cranium, 1047, which is be- 83). Standard deviations decreased after
lieved to be the only one having “tabular removing it from the sample at this step
erect” deformation type (Rivero de la Cal- but stayed over the world average for the
le 1949 cited in Herrera Fritot 1964:82– same variables as before except XFB and
Table 3. Summary statistics for the non-deformed sample
Male Female
Abbreviation Measurement
n Mean sd n Mean sd
GOL Glabello-occipital length 31 172.7 6.5 14 167.2 5.5
XCB Maximum cranial breadth 27 135.3 3.8 15 130.9 5.1
BBH Basion-bregma height 18 133.9 5.2 10 128.2 4.4
BNL Basion-nasion length 18 97.3 3.8 10 93.5 2.8
BPL Basion-prosthion length 14 93.6 2.8 9 92.1 3.6
WFB Minimum frontal breadth 32 92.8 5.0 14 87.7 4.2
XFB Maximum frontal breadth 28 111.9 5.6 14 108.6 4.8
AUB Biauricular breadth 26 124.0 4.9 14 116.4 4.4
ASB Biasterionic breadth 25 106.0 5.4 13 103.2 4.7
FRA Frontal arc 26 119.0 5.7 9 116.2 4.7
PAA Parietal arc 22 125.3 5.8 10 119.7 4.7
OCA Occipital arc 20 109.9 7.4 6 108.7 2.5
FRC Frontal chord 27 106.7 4.7 10 104.0 2.9
PAC Parietal chord 22 110.1 4.4 10 105.8 4.7
OCC Occipital chord 20 94.5 3.9 6 93.5 2.4
FMB Bifrontal breadth 33 104.3 3.0 13 98.7 3.2
ZYB Bizygomatic breadth 23 132.9 4.2 12 124.5 6.1
ZMB Bimaxillary breadth 28 95.7 3.9 16 90.8 4.8
UFH Upper facial height (n-alv) 30 63.1 3.6 14 61.6 1.8
OBB Orbital breadth from maxilofrontale 33 41.3 1.8 17 40.0 1.8
OBH Orbital height 34 33.1 1.3 17 33.1 1.5
NLB Nasal breadth 32 23.5 1.7 16 22.9 0.9
NLH Nasal height 32 49.3 3.2 17 47.4 2.3
ALL Alveolar length 14 49.8 3.6 5 49.3 3.9
ALB Alveolar breadth 15 59.2 4.6 5 56.5 3.4
SIS Simotic subtense 31 2.7 0.7 14 2.3 0.4
WNB Simotic chord 31 8.6 1.5 14 7.5 2.0
MFS Maxillofrontal subtense 30 5.7 1.2 14 5.0 1.0
MFC Maxillofrontal chord 30 19.3 2.2 15 17.2 2.0
FCD Fossa canina depth 29 3.8 2.0 15 3.8 2.2
FAN Forehead profile angle from nasion 19 82.0 3.8 13 84.5 2.8
FAG Forehead profile angle from glabella 19 76.2 3.5 13 78.2 3.2
GFA General facial angle 18 83.1 2.3 13 82.4 2.9
ALA Alveolar angle 14 65.9 6.5 9 65.5 6.0
NPA Nasal protrusion angle 20 21.2 6.6 13 16.9 4.1
NMA Nasomalar angle 29 144.0 4.4 12 144.4 4.1
ZMA Zygomaxillary angle 24 127.2 4.8 13 127.7 6.8244 Taisiya Syutkina et al.
FRC. Female crania seem to be consider- frontal/facial vertical angles (FAN, FAG,
ably more uniform than the male sample: GFA, ALA). SSD in the deformed sam-
standard deviations exceed the world av- ple is not high in terms of metric traits
erage for traits XCB, XFB, PAA, FCD and (observed index in the sample was high-
Table 4. Summary statistics for the deformed sample
Male Female
Abbreviation Measurement
n mean sd n mean sd
GOL Glabello-occipital length 23 172.0 5.6 12 160.6 4.2
XCB Maximum cranial breadth 22 154.5 5.5 12 148.0 5.4
BBH Basion-bregma height 17 125.6 5.4 9 122.3 3.2
BNL Basion-nasion length 17 97.2 4.3 9 91.0 2.1
BPL Basion-prosthion length 16 99.0 5.9 9 92.8 3.1
WFB Minimum frontal breadth 20 96.8 2.6 12 94.2 4.4
XFB Maximum frontal breadth 20 122.5 6.2 10 118.4 5.7
AUB Biauricular breadth 21 136.2 4.2 10 130.0 3.5
ASB Biasterionic breadth 20 112.5 3.4 12 109.1 4.4
FRA Frontal arc 23 117.9 6.1 10 113.0 2.9
PAA Parietal arc 23 107.6 9.7 11 103.4 8.2
OCA Occipital arc 18 113.5 11.7 11 108.5 6.9
FRC Frontal chord 23 110.0 5.4 10 104.8 3.4
PAC Parietal chord 23 95.8 7.0 11 92.5 4.8
OCC Occipital chord 18 97.4 7.6 11 92.5 4.2
FMB Bifrontal breadth 21 109.6 3.3 12 104.8 2.3
ZYB Bizygomatic breadth 17 142.5 4.6 10 132.2 3.7
ZMB Bimaxillary breadth 20 102.1 4.5 11 98.9 2.5
UFH Upper facial height (n-alv) 21 70.4 5.1 11 67.4 2.2
OBB Orbital breadth from maxilofrontale 22 43.3 1.9 11 41.3 1.8
OBH Orbital height 22 36.1 1.4 11 35.8 0.9
NLB Nasal breadth 22 26.2 1.6 11 25.0 1.6
NLH Nasal height 22 53.8 3.1 11 51.8 1.8
ALL Alveolar length 15 52.4 4.0 7 49.6 2.7
ALB Alveolar breadth 15 64.3 4.1 8 61.2 2.9
SIS Simotic subtense 21 3.6 0.8 11 3.1 0.6
WNB Simotic chord 21 10.4 1.6 11 10.1 1.0
MFS Maxillofrontal subtense 22 6.1 1.2 10 5.8 0.6
MFC Maxillofrontal chord 22 19.4 1.7 10 18.4 2.5
FCD Fossa canina depth 22 3.4 1.4 11 3.1 1.3
FAN Forehead profile angle from nasion 16 66.2 5.2 10 67.2 5.9
FAG Forehead profile angle from glabella 16 56.4 5.4 10 58.8 5.8
GFA General facial angle 15 81.1 2.9 11 80.5 3.3
ALA Alveolar angle 15 69.8 5.0 11 68.7 6.5
NPA Nasal protrusion angle 16 21.6 4.6 10 18.0 3.0
NMA Nasomalar angle 21 141.2 4.5 11 143.0 3.4
ZMA Zygomaxillary angle 20 127.5 4.2 11 126.5 3.9Pre-Columbian Cuba population cranial morphology 245
er than average maximum only for GOL, plot; moreover, pairs of morphologically
XCB, OCA, OCC and WNB), but male similar representatives of both groups are
and female crania are well distinguish- not infrequent (e.g., individuals number
able visually. 18 and 42, 5 and 24, 13 and 25). We can
also note the central position of the is-
PCA land individual from Punta del Este, the
Non-deformed sample marginal position of an individual with
Facial variables. The details about the unknown, but presumably Cuban aborig-
percentage of variance explained and inal origin, the relative proximity of two
loadings are plotted in Figure 3. As the crania which come from the same cave in
loading plots indicate, both PCs sum- Baracoa (individuals number 39 and 14).
marize differences in facial dimensions Neurocranial variables. The first two
and nasal height, with nasal breadth and PCs account for 78% of the total varia-
orbital dimensions being the least sig- tion of the sample based on neurocranial
nificant. The results of the PCA do not metric traits (Figure 4). The first PC ex-
reveal any territorial differences in the plains 49% of the variation and is a size
non-deformed sample. Crania from the variable as it is negatively correlated with
largest territorial groups – Canímar Aba- all the variables, length and height of the
jo and Baracoa (which is not a “group” in vault and occipital arc being the most
a strict sense) are found in all parts of the significant ones. The second PC explains
Fig. 3. PCA score plot (A) and loadings (B) of seven facial variables. Non-deformed sample. Numeration
as in Supplement Table 1246 Taisiya Syutkina et al.
Fig. 4. PCA score plot (A) and loadings (B) of ten neurocranial variables. Non-deformed sample. Numera-
tion as in Supplement Table 1
29% of the variation and is mainly cor- respectively, both being from the zone of
related with PAA (negatively) and OCA Baracoa. The PCA based on indices rather
(positively). Although the number of in- than metric variables yielded very similar
dividuals analyzed has decreased drasti- results and is not presented here. Indi-
cally due to poorer preservation of crani- viduals from different geographical loca-
al vault in the sample, it is interesting to tions appear to be randomly distributed
notice that some trends revealed by the in these plots, thus not suggesting any
PCA for the facial variables could also be morphological pattern connected with
observed for the neurocranial data. For geography in the non-deformed group.
example, individuals from Cueva Fría
(Baracoa) maintain their proximity to Mantel test
each other despite being tightly clustered To further explore whether there is an
with four individuals from four different association between morphology and
locations. So do individuals number 13 location of the studied crania from the
and 7 also from Maisí (more detailed lo- non-deformed sample, the Mantel test
cation is unknown), individuals number was employed over the matrix of Euclid-
16 and 17 from Canímar Abajo. Punta del ean distances between the morphological
Este individual maintains its central po- variables and the matrix of geographical
sition, while individuals number 3 and 2 distances between all the sites. No cor-
now hold the most marginal positions in relation was found between the matrices
terms of the first PC and the second PC, of neurocranial variables and geograph-Pre-Columbian Cuba population cranial morphology 247 ical distances (r = 0.04, p = 0.3). This the Mantel test did not reveal any con- set of variables (GOL, XCB, WFB, XFB, nection between the craniometric data AUB, ASB) accounted for the length and and geographic distances for the non-de- width dimensions of the crania, but not formed sample. their height, due to generally poor pres- ervation of the bases of the crania. Inclu- Deformed sample sion of the cranial height in the subset Shape change analysis reduced the sample to 12 individuals, Figure 5 shows shape variation for the and although such an analysis cannot first two components. Most specimens be considered reliable, it was carried in the deformed sample come from the out to ensure that a possible associa- same zone of Baracoa, several come tion was not overlooked. The correlation from other Cuban locations, five are of remained negligible and insignificant, unknown but presumably Cuban origin, though (r = 0.03, p = 0.3). Finally, the and three crania come from two sites in Mantel test on the facial variables yielded the Dominican Republic. The latter do a similar result: r = −0.1, p = 0.9. Thus, not appear to cluster together and are Fig. 5. Shape variation of the deformed crania in the lateral view. Numeration as in Supplement Table 1
248 Taisiya Syutkina et al.
scattered across the main cloud of obser- cantly (p = 0,04) more frequent on the
vations. right side.
The plot shows a continuous varia-
tion of the deformed crania between the Discussion and conclusions
shapes at the extremes of the PC1 and
PC2 axes, depicted at the sides of the The goal of the present paper was to see
plot. PC1, which accounts for 51% of the whether we could expand our under-
variation, marks the direction of change standing of the morphological variation
from a more severely affected vault with of the Cuban pre-Columbian population
flattened frontal and occipital bones and on the intragroup level by applying var-
a marked depression after the coronal su- ious methods of studying cranial mor-
ture on the left to a more rounded and phology.
close to physiological norm vault on the We acknowledge that the skeletal ma-
right. The crania of both sexes are dis- terial available for the present study has
tributed randomly across the plot. not become substantially larger since the
early papers reviewed above. While es-
Non-metric traits variation timates of the total number of individu-
Frequencies of the non-metric traits are als from archaeologically excavated sites
shown in Table 2. Both groups have zero often numbers into hundreds (Valcárcel
frequencies of the metopic suture, low Rojas 2016; Rodríguez Hernández 1998;
or zero frequencies of sagittal and cor- Chinique de Armas and Rodríguez Su-
onal ossicles, ossicles at lambda, Inca arez 2012), the remains themselves are
bone, palatine tori, pterion types other usually very fragmentary. The present
than sphenoparietal (although epipteric study could not cover all existing collec-
bones occur in the 20% of the deformed tions for various reasons, but much more
individuals), trochlear spine, double con- representative samples than existing at
dylar facet. Both groups also have high the moment would be needed for more
rates of the supraorbital notch, multiple objective conclusions. Relethford (2002)
zygomatic foramina, parietal foramina, has shown that more than 80% of the to-
posterior condylar canals, and infraor- tal craniometric variation is found within
bital sutures. At the same time, there is local populations, i.e., among individ-
a marked contrast in the frequencies of uals. According to Kozintsev (2016:3),
lambdoid, occipito-mastoidal, squamo- more than 120 specimens in a sample are
parietal ossicles, and ossicles at the aste- required to trace its composite nature.
rion between the two groups: all of these At the same time, if morphological pat-
traits are very common in the deformed terns are parallelized by some indepen-
group and rather rare in the non-de- dent criteria, e.g., varying archaeological
formed one. context, the antiquity of the burials etc.,
No significant differences between it is justified to suggest a really existing
sexes were found in the frequency of intragroup differentiation.
non-metric traits in both samples. When How right are we to pool all non-de-
symmetry was considered, the only sig- formed crania in one sample? On the
nificant difference was seen in wormian one hand, our analysis did not reveal
bones in the masto-occipital suture in any evident clustering of the individu-
the deformed sample: they were signifi- als in the pooled sample, and we do notPre-Columbian Cuba population cranial morphology 249
possess any certain information about from Cave 4 in Punta del Este, a much
them representing different populations. younger site. Similarly, the absence of
On the other hand, the opposite is also clear patterns of clustering and scarcity
true: there is no evidence of them be- of materials does not allow a comparison
ing a single population. Some scholars between sites with different archaeolog-
(Ulloa Hung and Valcárcel Rojas 2019, ical settings. Bolufé Torres (2015) at-
González Herrera 2008) have argued that tempted to study morphological patterns
“Archaic” or “Siboney” in the Caribbean among skeletal remains from four sites
is a historical and archaeological con- situated in the western part of the island
struct, based on first European written and found significant differences in the
sources and supported by the need for orbital breadth between Guayabo Blanco
classification and systematization of ac- individuals on the one hand and Canímar
cumulated archaeological materials and Abajo and Cueva Calero – on the oth-
cultures, while in fact this concept masks er. A geometric morphometric analysis
the actual diversity of groups that form revealed variation in the cranial shape
this artificial “community”. among the three samples. Although the
In this study, the non-deformed cra- author concludes that the results might
nia proved to be rather homogeneous, indicate the different origin of the in-
with the alveolar region being the most dividuals, the results should be treated
variable one. Principal component analy- with caution due to the small sample
sis and Mantel test did not reveal any ter- sizes (1–8 individuals in each sample).
ritorial differences in the cranial metric The present study, based on multivariate
traits. No patterns resulting from time analyses of cranial metrics, does not sup-
differences among the samples could be port this conclusion.
seen either based on the data available The deformed sample is significant-
for the study (see Table 5 for available ly more variable, especially its male
dates). For the non-deformed sample, in- part. The principal component analy-
dividuals from the single site of Canímar sis of the Procrustes coordinates of the
Abajo were found scattered throughout cranial vault outline in the lateral norm
the PCA plot, while the individual com- revealed continuous variability of crani-
ing from Guayabo Blanco, the oldest site al shapes from the ones with more flat-
among the ones that could be included tened frontal and occipital bones to the
in the analysis, was found in the mid- more curved outlines. In our study, most
dle of the distribution next to the one of the deformed crania come from the
Table 5. Radiocarbon dates for some of the sites where the materials of the study come from. All dates are
in calibrated years BP
Site Culture Date Reference
Canímar Abajo (Younger cemetery) Archaic 1570–7200 Nägele et al. 2020
Guayabo Blanco Archaic 2530–1700 Nägele et al. 2020
Cueva Calero Archaic 1500–1100 Nägele et al. 2020
La Caleta (Dominican Republic) Ceramic 1307–3320 Fernandez et al. 2021
Punta del Este, Cueva 4 Archaic 969–675 Ernesto Tabio, Estudio de las culturas
Antiguas en Cuba
El Purial Archaic 3644–2780 Cooper 2007250 Taisiya Syutkina et al.
same region of Baracoa (in many cases suture inside the deformed sample. The
without any additional information), and only available written evidence of the
three individuals from the Dominican deforming apparatus in the Caribbean
Republic do not tend to cluster togeth- (although a long time after the contact
er, so geography cannot account for this and not directly in Cuba) is provided by
variation of the deformed shapes. More- Leblond, who evidenced the practice in
over, a highly specific individual number the late 18th century (1813 cited in van
1047 with the deformation type that has Duijvenbode 2017:80) and describes a
been described as “tabular erect” instead procedure of placing two wooden boards
of “tabular oblique” comes from the at the front and the back of the skull of a
same zone of Baracoa. Sexual differences baby. The asymmetry found in our study
do not explain the variation either, which might result from more pressure applied
contradicts the conclusion of Ginzburg to the right side of the deforming ban-
(1964:202), who reported more mod- dage when a right-handed person fixes
erate deformation of the female crania. it.
Thus, it can be assumed that there were Thus, the study revealed relative ho-
no differences in the application of the mogeneity of both non-deformed and de-
deforming device to male and female ba- formed samples and the absence of sys-
bies. The lack of contextual information tematic differences inside each sample or
and radiocarbon dates for the deformed proved the impossibility of available cra-
sample make it impossible to test wheth- niological data to reflect possible intra-
er the variation of the deformed cranial group differentiation. In other words, by
shapes is due to changes of the tradition now, there are no grounds for identify-
over time. Our results do not support ing sub-groups inside the non-deformed
Herrera Fritot’s theory of deforming de- and deformed samples based on cranial
vices of two types: if this were the case, morphology.
we would expect to see two separate Studies of skeletal remains are inevi-
clusters of crania rather than continuous tably limited to dealing with samples that
variation. Therefore, the most reason- do not necessarily reflect the properties
able explanation for the observed trend of the population it represents, but the
is an individual variation of the resulting more numerous the sample is and the
shapes. more information about it is available,
Variation of non-metric traits in each the more credible the conclusions can
sample found in this study is totally in be. It is probable that as the pool of ma-
accordance with earlier works (Rivero terials increases, the trends identified in
de la Calle 1983:180–181). The high the present study will be proved errone-
frequency of sutural ossicles in the de- ous, and new patterns not evident in the
formed sample can be both due to the analysis of such limited material will be
effect of deformation or manifesta- revealed instead. We hope that this paper
tion of a particular genetic complex – a will serve as a reference point for future
non-deformed sample of the same or- studies that will confirm or refute the
igin would be required to make a defi- preliminary conclusions offered here and
nite conclusion. An interesting result is continue to explore relations between
an asymmetry seen in the expression of different groups populating the island
wormian bones in the mastoid-occipital throughout its pre-Columbian history.Pre-Columbian Cuba population cranial morphology 251
Acknowledgments References
Adams D, Collyer M, Kaliontzopoulou A, Bak-
The authors received no financial sup-
en E. 2021. Geomorph: Software for geo-
port for the research, authorship, or metric morphometric analyses. R pack-
publication of this article. The authors age version 3.3.2. Available at: https://
are grateful to Plinio Luis Gainza García cran.r-project.org/package=geomorph
for providing access to cranial collections [Accessed 12 June 2020].
in Baracoa, to the staff members of all Alarie K, Roksandic M. 2016. A Pre-Colum-
hosting institutions for their help during bian Dental Modification Complex at the
the data collection, to Vladimir Kufterin Site of Canímar Abajo, Matanzas, Cuba.
and Nadezhda Dubova for valuable com- In: Roksandic I, editor. Cuban Archaeolo-
gy in the Caribbean. Gainesville: Universi-
ments on the draft version of the manu-
ty Press of Florida. 106–24.
script and anonymous reviewers for their Alexeev VP, Debets GF. 1964. Kraniometriya:
helpful suggestions and efforts towards Metodika antropologicheskikh issledo-
improving our manuscript. vaniy. [Craniometry: methods of anthro-
pological research]. Moscow: Nauka.
The Authors’ contribution Alexeev VP. 1986. K antropologicheskoi
kharakteristike aborigennogo naseleniia
TS: organization of work, writing the Kuby [To the anthropological characteris-
draft and final version, measurements tic of the aboriginal population of Cuba].
In: Vasilievskiy RS, editor. Arkheologiia
and statistical analyses; MJGP: writing
Kuby [Archaeology of Cuba]. Novosi-
the draft and final manuscript, data ac- birsk: Nauka. 15–22.
quisition, performing of the data visu- Alonso Alonso AM. 1995. Fundamentos para
alization; STHG: writing final version of la Historia del Guanahatabey de Cuba. La
the work, data acquisition and data cura- Habana: Editorial Academia.
tor; CAA: co-author of the final version, Antón SC. 1989. Intentional cranial vault de-
work and data curator; ARR: co-author formation and induced changes of the cra-
of the final version of the work and data nial base and face. Am J Phys Anthropol
curator. 79(2):253–67.
Armstrong S, Cloutier L, Arredondo C, Rok-
sandic M, Matheson C. 2013. Spina bifida
Conflict of interest in a pre-Columbian Cuban population:
a paleoepidemiological study of genetic
The authors declare that there is no con- and dietary risk factors. Int J Paleopathol
flict of interest. 3(1):19–29.
Supplementary materials available on Berry CA, Berry RJ. 1967. Epigenetic vari-
request from the authors due to privacy/ ation in the human cranium. J Anat
ethical restrictions. 101(2):361–79.
Bolufé Torres R. 2015. Variaciones de la es-
tatura y las características morfométricas
Corresponding author
craneofaciales en las poblaciones pesca-
doras recolectoras cazadoras del occiden-
Taisiya Syutkina, Checherskiy proezd- te de Cuba: Cueva del Perico I, Canímar
24-130, Moscow, Russian Federation Abajo, Cueva Calero y Guayabo Blanco.
e-mail: syuttaya@gmail.com Tesis en opción al grado de Licenciado en
Biología. La Habana: Universidad de La
Habana.252 Taisiya Syutkina et al.
Bookstein FL. Morphometric Tools for Land- tions of the Caribbean. Int J Osteoarchae-
mark Data: Geometry and Biology. Cam- ol 28(5):492–509.
bridge: Cambridge University Press, 1991. Chinique de Armas Y, Rodríguez Suárez R.
Buikstra JE, Ubelaker DH editors. 1994. Stan- 2012. Cambios en las actividades subsis-
dards for Data Collection from Human tenciales de los aborígenes del sitio arque-
Skeletal Remains. Fayetteville: Arkansas ológico Canímar Abajo, Matanzas, Cuba.
Archeological Service. Cuba Arqueológica 5(2):30–48.
Bunak VV. 1959. Cherep cheloveka y stadii Chinique de Armas Y, Roksandic M, Nikitović
iego formirovaniia u iskopaiemykh liudei D, Rodríguez Suárez R, Smith D, Kanik
i sovremennykh ras [Human skull and the N et al. 2017. Isotopic reconstruction of
stages of its formation in fossil people and the weaning process in the archaeolog-
modern races]. Moscow: USSR Academy ical population of Canímar Abajo, Cuba:
of Sciences Publishing House. A Bayesian probability mixing model ap-
Buser TJ, Sidlauskas BL, Summers AP. 2018. proach. PloS one 12(5):e0176065.
2D or not 2D? Testing the utility of 2D Chinique de Armas Y, Roksandic M. 2019.
vs. 3D landmark data in geometric mor- Assessing the Biological and Cultural Di-
phometrics of the sculpin subfamily Oli- versity of Archaic Age Populations from
gocottinae (Pisces; Cottoidea). Anat Rec Western Cuba. Smithson. Contr. An-
301(5):806–18. thropol 51:161–71.
Cardini A, Chiapelli M. 2020. How flat can a Cooper J. 2007. Island interaction in the
horse be? Exploring 2D approximations of prehistoric Caribbean: an archaeological
3D crania in equids. Zoology 139:125746. case study from northern Cuba. Doctoral
Cardini A. 2014. Missing the third dimension dissertation. London: University College
in geometric morphometrics: how to as- London.
sess if 2D images really are a good proxy Coppa A, Cucina A, Hoogland ML, Lucci M,
for 3D structures? Hystrix 25(2):73–81. Luna Calderón F, Panhuysen RGAM et
Cheverud J., Kohn LA, Konigsberg LW, Leigh al. 2008. New evidence of two different
SR. 1992. Effects of fronto-occipital artifi- migratory waves in the circum-Caribbe-
cial cranial vault modification on the cra- an area during the pre-Columbian period
nial base and face. Am J Phys Anthropol from the analysis of dental morphological
88(3):323–45. traits. In: Hofman CL, Hoogland ML and
Chinique de Armas Y, Buhay WM, Rodríguez AL van Gijn, editors. Crossing the bor-
Suárez R, Bestel S, Smith D, Mowat SD et ders: new methods and techniques in the
al. 2015. Starch analysis and isotopic evi- study of archaeological materials from the
dence of consumption of cultigens among Caribbean. Tuscaloosa: University of Ala-
fisher–gatherers in Cuba: the archaeologi- bama Press. 195–213.
cal site of Canímar Abajo, Matanzas. J Ar- Crespo-Torres EF, Mickleburgh HL, Valcárcel
chaeol Sci 58:121–32. Rojas R. 2013. In: Keegan WF, Hofman
Chinique de Armas Y, González Herrera UM, CL and R Rodríguez Ramos, editors. The
Buhay WM, Yero Masdeu JM, Viera San- Oxford handbook of Caribbean archaeol-
fiel LM, Burchell M et al. 2020. Chronol- ogy. New York: Oxford University Press.
ogy of the archaeological site of Playa Del 436–51.
Mango, Rio Cauto, Granma, Cuba. Radio- Czarnetzki A. 1971. Epigenetische Skelett-
carbon 62(5):1221–36. merkmale im Populationsvergleich: I.
Chinique de Armas Y, Pestle W. 2018. Assess- Rechts-links-Unterschiede bilateral ange-
ing the association between subsistence legter Merkmale. Z Morphol Anthropol
strategies and the timing of weaning 2:238–54.
among indigenous archaeological popula- Dembo A, Imbelloni J. 1938. Deformaciones
Intencionales del Cuerpo Humano de Car-You can also read
