ArticlePDF Available

Abstract and Figures

Understanding the dynamics of the Late Quaternary Caribbean mammal extinction event is complicated by continuing uncertainty over the taxonomic status of many species. Hispaniola is one of the few Caribbean islands to retain native non-volant mammals; however, there has been little consensus over past or present levels of diversity in Hispaniolan hutias (Capromyidae: Plagiodontinae). Craniodental measurement data from modern hutia specimens, previously classified as both Plagiodontia aedium and P. hylaeum, displaymorphological differences between Hispaniola’s northern and southern palaeo-islands using MANOVA and PCA. Although attempts to amplify mitochondrial DNA from the holotype of P. aedium were unsuccessful, this specimen is morphometrically associated with southern palaeo-island specimens. The mandibular size distribution of recent Plagiodontia specimens is unimodal, but the Late Quaternary mandibular size distribution is multimodal and displays much broader measurement spread, representing multiple extinct species. Finite Mixture Analysis was used to assess the best fit of different taxonomic hypotheses to the fossil mandibular size distribution. All retained FMA models include living hutias and P. spelaeum as distinct taxa; PCA further demonstrates that levels of morphological variation between modern hutia populations are lower than levels between living hutias and P. spelaeum, so that living hutias are interpreted as the single species P. aedium. Taxonomic differentiation for larger-bodied hutias is less well defined, but most retained models show only one larger species, for which the only available name is P. velozi. ‘Plagiodontia’ araeum is morphologically distinct from other species and is reassigned to Hyperplagiodontia. Hispaniola’s plagiodontine fauna has lost its largest and smallest representatives; similar trends of body size selectivity in extinction risk are shown more widely across the Caribbean mammal fauna, possibly due to different regional anthropogenic threats (invasive mammals, hunting) affecting small-bodied and large-bodied mammals during the recent past. This apparent pattern of extinction selectivity is named the ‘Goldilocks Hypothesis’.
Content may be subject to copyright.
This article was downloaded by: [The Zoological Society of London], [Samuel T. Turvey]
On: 19 December 2012, At: 05:53
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Systematics and Biodiversity
Publication details, including instructions for authors and subscription information:
Taxonomy-testing and the ‘Goldilocks Hypothesis’:
morphometric analysis of species diversity in living and
extinct Hispaniolan hutias
James Hansford a b , José M. Nuñez-Miño c , Richard P. Young c d , Selina Brace e , Jorge L.
Brocca f & Samuel T. Turvey b
a Department of Biology, University of York, Wentworth Way, York YO10 5DD, UK
b Institute of Zoology, Zoological Society of London, Regent's Park, London, NW1 4RY, UK
c Durrell Wildlife Conservation Trust, Les Augrès Manor, Trinity, Jersey JE3 5BP, Channel
d Department of Biology and Biochemistry, University of Bath, Bath BA2 7AY, UK
e School of Biological Sciences, Royal Holloway University of London, Egham TW20 0EX, UK
f Sociedad Ornitológica de la Hispaniola, Parque Zoologico Nacional, Avenida de la Vega
Real, Arroyo Hondo, Santo Domingo, Dominican Republic
To cite this article: James Hansford , José M. Nuñez-Miño , Richard P. Young , Selina Brace , Jorge L. Brocca & Samuel T.
Turvey (2012): Taxonomy-testing and the ‘Goldilocks Hypothesis’: morphometric analysis of species diversity in living and
extinct Hispaniolan hutias, Systematics and Biodiversity, 10:4, 491-507
To link to this article:
Full terms and conditions of use:
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to
anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should
be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,
proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in
connection with or arising out of the use of this material.
Systematics and Biodiversity (2012), 10(4):491507
Research Article
Taxonomy-testing and the ‘Goldilocks Hypothesis: morp h o m e t r i c
analysis of species diversity in living and extinct Hispaniolan hutias
1Department of Biology, University of York, Wentworth Way, York YO10 5DD, UK
2Institute of Zoology, Zoological Society of London, Regent’s Park, London NW1 4RY, UK
3Durrell Wildlife Conservation Trust, Les Augr`
es Manor, Trinity, Jersey JE3 5BP, Channel Islands
4Department of Biology and Biochemistry, University of Bath, Bath BA2 7AY, UK
5School of Biological Sciences, Royal Holloway University of London, Egham TW20 0EX, UK
6Sociedad Ornitol´
ogica de la Hispaniola, Parque Zoologico Nacional, Avenida de la Vega Real, Arroyo Hondo, Santo Domingo,
Dominican Republic
(Received 5 October 2012; revised 6 November 2012; accepted 8 November 2012)
Understanding the dynamics of the Late Quaternary Caribbean mammal extinction event is complicated by continuing
uncertainty over the taxonomic status of many species. Hispaniola is one of the few Caribbean islands to retain native
non-volant mammals; however, there has been little consensus over past or present levels of diversity in Hispaniolan hutias
(Capromyidae: Plagiodontinae). Craniodental measurement data from modern hutia specimens, previously classified as
both Plagiodontia aedium and P.hylaeum,displaymorphologicaldifferencesbetweenHispaniolasnorthernandsouthern
palaeo-islands using MANOVA and PCA. Although attempts to amplify mitochondrial DNA from the holotype of
P.aedium were unsuccessful, this specimen is morphometrically associated with southern palaeo-island specimens. The
mandibular size distribution of recent Plagiodontia specimens is unimodal, but the Late Quaternary mandibular size
distribution is multimodal and displays much broader measurement spread, representing multiple extinct species. Finite
Mixture Analysis was used to assess the best fit of different taxonomic hypotheses to the fossil mandibular size distribution.
All retained FMA models include living hutias and P.spelaeum as distinct taxa; PCA further demonstrates that levels of
morphological variation between modern hutia populations are lower than levels between living hutias and P.spelaeum,so
that living hutias are interpreted as the single species P.aedium. Taxonomic differentiation for larger-bodied hutias is less
well defined, but most retained models show only one larger species, for which the only available name is P.velozi.
Plagiodontiaaraeum is morphologically distinct from other species and is reassigned to Hyperplagiodontia.Hispaniolas
plagiodontine fauna has lost its largest and smallest representatives; similar trends of body size selectivity in extinction risk
are shown more widely across the Caribbean mammal fauna, possibly due to different regional anthropogenic threats
(invasive mammals, hunting) affecting small-bodied and large-bodied mammals during the recent past. This apparent
pattern of extinction selectivity is named the ‘Goldilocks Hypothesis’.
Key words: Dominican Republic, extinct mammal, extinction risk, Finite Mixture Analysis, Haiti, Hyperplagiodontia,
The Late Quaternary land mammal fauna of the insular
Caribbean is characterized by endemism and evolution-
ary radiation, and contains several ancient higher-order
clades not represented in the adjacent continental regions
of North or South America. The most species-rich en-
demic Late Quaternary Caribbean mammal family is the
Capromyidae, which comprised approximately 30 species
Correspondence to: Samuel Turvey. E-mail: samuel.turvey@ioz.
of medium- to large-bodied rodents known as hutias
(Turvey, 2009). However, the Caribbean land mammal
fauna experienced an extremely severe series of extinc-
tions during the Late Quaternary, and it is estimated that
over 1 00 endemic species of sloths, primates, eulipoty phlan
insectivores, bats and rodents may have died out during
this interval, largely due to prehistoric and historical-era
human impacts (MacPhee & Flemming, 1999; MacPhee,
2009; Turvey, 2009; Turvey & Fritz, 2011). Non-volant land
mammals were particularly badly affected by this extinction
event, and only eight currently recognized hutia species are
ISSN 1477-2000 print / 1478-0933 online
!2012 The Natural History Museum
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
492 J. Hansford et al.
Bond’s Line
palaeo-island divide
Massif de la Selle/
Sierra de Bahoruco
de la Hotte
Fig. 1. Map of Hispaniola, showing geotectonic boundaries and localities indicated in the text. Capital cities indicated with stars. Key:
1, Roseaux; 2, Miragoˆ
ane; 3, Trou Zombie, ˆ
Ile de la Gonˆ
ave; 4, St. Michel de l’Atalaye; 5, Trouing Marassa; 6, Rancho de la Guardia;
7, Cueva de Col´
on; 8, Cueva de la Virgen.
probably still extant (Borroto-P´
aez & Mancina, 2011;
IUCN, 2011).
Reconstructing the dynamics and drivers of the
Caribbean mammal extinction event has obvious implica-
tions for understanding wider patterns of global mammalian
biodiversity loss and past human impacts on island systems
(MacPhee & Flemming, 1999). However, this understand-
ing has been complicated by continued uncertainty over true
levels of Late Quaternary Caribbean mammalian diversity,
and the taxonomic status of many of the region’s extinct
and extant mammal species and populations. In addition
to the ongoing description of recently extinct taxa based
on new discoveries from the Caribbean Late Quaternary
fossil record, numerous putative species have not been for-
mally described and remain in open nomenclature, whereas
taxonomic revisions have increasingly synonymized sev-
eral species in groups such as hutias and nesophontid
island-shrews (D´
ıaz-Franco, 2001; Condis Fern´
andez et al.,
2005; Silva Taboada et al., 2007; Borroto-P´
aez & Mancina,
2011). Further synonyms no doubt remain to be docu-
mented within other Caribbean species groups, but in the
absence of formal taxonomic revision using modern quan-
titative morphometric or molecular analyses (Berovides &
Condis Fern´
andez, 2002; Milishnikov et al., 2010), it re-
mains essentially up to the arbitrary discretion of different
authors to decide how many species are probably valid.
Hispaniola, divided politically into the Dominican Re-
public and Haiti, is a large Caribbean island with a di-
verse biota containing numerous endemic species radia-
tions. Many Hispaniolan taxa show congruent patterns of
intra-island endemism that represent historical allopatry
driven by the island’s complex geotectonic history (Gifford
et al., 2004; Townsend et al., 2007; Gifford & Larson, 2008;
Sly et al., 2011; Fig. 1). Hispaniola consists of indepen-
dent northern and southern palaeo-islands which docked
in the late Miocene (Mann et al., 1991; Iturralde-Vinent
& MacPhee, 1999), but which remained separated by the
Neiba Valley, a prominent depression that was periodically
or permanently inundated by a narrow seaway until the late
Pleistocene (Maurrasse et al., 1982; Graham, 2003). The
southern palaeo-island or Presqu’ˆ
ıle du Sud is further sub-
divided into two major physiographic provinces (the Massif
de la Hotte and the Massif de la Selle–Sierra de Bahoruco)
separated by the Jacmel–Fauch ´
the peninsula and which was also inundated by a sea chan-
nel during some or all of the Plio-Pleistocene (Maurrasse
et al., 1982). All three of these regions are biogeograph-
ically distinct and are characterized by substantial levels
of endemism in many terrestrial vertebrates, invertebrates
and plants (e.g. Williams, 1961; Schwartz, 1980; Hedges,
1999), with the Jacmel–Fauch´
e depression referred to as
‘Bond’s Line’ after the ornithologist James Bond (Latta
et al., 2006). However, some Hispaniolan taxa show no
evidence of lineage divergence associated with historical
intra-island allopatry (Sly et al., 2011).
Hispaniola is one of the few Caribbean islands to retain
native capromyid rodents, referred to the endemic genus
Plagiodontia Cuvier, 1836 and the endemic subfamily Pla-
giodontinae. This group is generally interpreted as a valid
monophyletic clade (e.g. Woods, 1989), although the cladis-
tic analysis of Caribbean rodents by Woods et al.(2001)
also nested the Puerto Rican heptaxodontid Elasmodonto-
mys within the clade when echimyids were used as the out-
group. Hispaniolan hutias have been considered rare and
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
Species diversity in Hispaniolan hutias 493
Fig. 2. Original scientific illustration of Plagiodontia aedium from Cuvier (1836), showing mounted holotype specimen.
in serious danger of extinction since their original discov-
ery and description in the early nineteenth century (Cuvier,
1836); other than an animal exhibited in London Zoo in
the 1850s (Flower, 1929), living individuals were not found
again by researchers until 1911 (Woods, 1981), and were
widely thought to be extinct until the mid-twentieth century
(Allen, 1942; Fisher & Blomberg, 2011). They are currently
listed as Endangered by the IUCN (2011).
Different researchers have reached little agreement
over either past or present levels of species diversity
in Plagiodontia,andthetaxonomichistoryofHispan-
iolan hutias has been confusingly complex. The type
species, P. a e d i u m Cuvier, 1836, was described on the
basis of an animal collected in 1826 from an unknown
locality on Hispaniola (Miller, 1928; Fig. 2). A second
species, P. h y l a e u m Miller, 1928, was later established
to describe hutias from the northern Dominican Repub-
lic (northern palaeo-island) on the basis of perceived sub-
tle soft-tissue differences from the holotype. Subsequent
researchers have variously considered that aedium and
hylaeum were sympatric species (Miller, 1929b,1930;
Mohr, 1939; Tate, 1948; R´
ımoli, 1976), allopatric species
(Johnson, 1948; Woods, 1989), allopatric subspecies
(Anderson, 1965; Woods et al., 2001; Wilson & Reeder,
2005) or morphologically indistinguishable synonyms
(Woods & Howland, 1979; Woods, 1981). Although the
collection locality of the holotype of P. a e d i u m remains un-
known, the name has generally been attached to the south-
ern Haitian population (southern palaeo-island) following
apparent soft-tissue similarities between the holotype and
an animal from Miragoˆ
ane reported by Johnson (1948).
This confusion over the identity and relationship of living
hutia populations has led some authors to refer to living
Hispaniolan hutias simply as ‘Plagiodontia sp.’ (Salazar,
Recently, Brace et al. (2012) used mitochondrial DNA
analysis of modern and historical Plagiodontia samples
from across Hispaniola to confirm that distinct hutia sister
taxa occur allopatrically in different regions of the island.
The genus was shown to comprise three distinct lineages
biogeographically congruent with the island’s geotectonic
history, with a primary phylogenetic division into north-
ern and southern lineages that diverged c.0.6Maand
a further near-complete subdivision of the southern pop-
ulation across Bond’s Line into eastern and western lin-
eages. However, in the absence of consistent morphometric
variation known to differentiate Plagiodontia populations,
Brace et al.(2012)conservativelyrecognizedasingleliving
Hispaniolan hutia species containing two phylogenetically
distinct subspecies, P. a e d i u m a e d i u m (southern lineage,
following usage of previous authors) and P. a e d i u m h y -
laeum (northern lineage), but recommended that all three
allopatric Plagiodontia populations should be treated as
distinct evolutionary units for conservation management.
Brace et al.(2012)didnotsampletheholotypeofP.
aedium in their study, so that the validity of the species
names aedium and hylaeum as applied to the two main
clades in this proposed subspecies-level taxonomy remains
Hispaniola’s Late Quaternary fossil and zooarchaeolog-
ical records also contain several other extinct endemic
rodents, which include both plagiodontines and other
non-plagiodontine capromyid, echimyid and heptaxodon-
tid genera (Brotomys,Hexolobodon,Isolobodon,Quemisia;
ımoli, 1976; Woods, 1989). However, wider levels of pla-
giodontine hutia species and genus diversity in the pre-
human Late Quaternary mammal fauna of Hispaniola are
also extremely poorly understood. Holocene zooarchaeo-
logical material recovered from Amerindian kitchen mid-
dens in the northern Dominican Republic during the nine-
teenth and early twentieth centuries was initially assigned
to P. a e d i u m (Miller, 1916a,1916b;deBooy,1919),adding
to the confusion over the identity of extant hutia popu-
lations as these skeletal remains were larger than living
individuals from nearby localities that were differentiated
as P. h y l a e u m (Miller, 1928). Hutia remains identified as
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
494 J. Hansford et al.
representing both P. a e d i u m and P. h y l a e u m were reported
from Late Quaternary cave deposits in both the Dominican
Republic and Haiti by Miller (1929a,1929b,1930),butthe
larger remains were subsequently redescribed as a distinct
extinct species, P. i p n a e u m Johnson, 1948. Smaller fossil
material from caves near St. Michel de l’Atalaye, Haiti,
was also described as another extinct species, P. s p e l a e u m
Miller, 1929; however, although this species was recognized
by some subsequent authors (e.g. Johnson, 1948; R´
1976; Woods, 1989), today it is generally ignored or re-
garded as a junior synonym of P. a e d i u m (MacPhee &
Flemming, 1999; Wilson & Reeder, 2005). Two further
putative large-bodied extinct hutia species, P. c a l e t e n s i s
ımoli, 1976 and P. v e l o z i R´
ımoli, 1976, were established
on the basis of re-examination of Miller’s fossil material
and new zooarchaeological collections, but these have since
been regarded as junior synonyms of P. i p n a e u m by most au-
thors except Woods (1989), Woods & Ottenwalder (1992)
and Woods et al. (2001), who supported the validity of
P. v e l o z i .Theconfusionoverboththevalidityanddiag-
nostic characteristics of this large series of extinct species
has again led many recent authors to identify fossil ma-
terial of Plagiodontia simply as ‘Plagiodontia sp.’ (e.g.
Trias et al., 1997; MacPhee et al., 2000; McFarlane et al.,
All of these putative species of Plagiodontia are mor-
phologically very similar to one another, especially in the
pattern of oblique enamel folds on the occlusal surface of
the cheek teeth, and vary mainly in overall size. Other pla-
giodontine taxa recorded from the Late Quaternary fossil
record of Hispaniola are more morphologically distinct. An
unusual large isolated upper cheektooth (left DP4) from a
cave near Rancho de la Guardia, Dominican Republic, was
provisionally assigned to Plagiodontia and described as the
new species P.? araeum Ray, 1964. Although this species
was recovered as the sister taxon to P. a e d i u m in the cladis-
tic analysis of Woods et al.(2001),whichdidnotsample
any other Plagiodontia species, Ray (1964) considered that
it would probably require a new genus when more material
became available. However, subsequent taxonomic treat-
ments of the Plagiodontinae have retained P. a r a e u m within
Plagiodontia with little further consideration of the taxo-
nomic significance of its morphology (R´
ımoli, 1976; Woods
et al., 2001; Wilson & Reeder, 2005), whereas morpholog-
ically similar material (originally assigned to P. a e d i u m by
Miller, 1929a)wasdescribedasanewgenusandspecies,
Hyperplagiodontia stenocoralis R´
ımoli, 1976. Hyperpla-
giodontia stenocoralis has been interpreted as a junior syn-
onym of P. a r a e u m by some recent authors (e.g. Wilson &
Reeder, 2005), but is still provisionally used as a valid taxon
by others (e.g. Trias et al., 1997). A further extinct genus
and species, Rhizoplagiodontia lemkei Wood s, 19 89, w as
established on the basis of extensive, morphologically dis-
tinctive and well-described fossil material from the Massif
de la Hotte.
In order to clarify the currently confused state of Hispan-
iolan hutia taxonomy and determine the true number of liv-
ing and extinct species within Plagiodontia,weconducteda
series of quantitative morphometric analyses that included
measurement data from almost all of the modern, historical,
zooarchaeological and fossil specimens variously assigned
to the genus that are available in worldwide museum col-
lections. We test previously proposed taxonomic hypothe-
ses and present the first rigorous assessment of intraspe-
cific variation and species boundaries within Plagiodontia,
to provide new insights for understanding ecological pat-
terns and processes of mammal extinction within the insular
Materials and methods
Specimens and measurements
One hundred and eighty-one modern and Late Quaternary
zooarchaeological and fossil crania and/or mandibles (ei-
ther associated specimens or isolated elements) of adult
hutia individuals (defined as individuals showing complete
dental eruption) from Hispaniola previously assigned to
Plagiodontia and Hyperplagiodontia were studied from
the following collections: Grant Museum of Zoology,
London (LDUCZ), 14 modern specimens and one fossil
specimen; Museum National d’Histoire Naturelle, Paris
(MNHN), one modern specimen; United States National
Museum (USNM), 11 modern specimens and 20 fossil
specimens; Museum of Comparative Zoology, Harvard
University (MCZ), one fossil specimen; Florida Museum
of Natural History, Gainesville (UF), 27 modern speci-
mens and 88 fossil specimens; Museo Nacional de Historia
Natural, Santo Domingo, Dominican Republic (MHND),
13 fossil specimens; private collection of Renato R´
Santo Domingo, Dominican Republic (ROR), five fossil
specimens (see R´
ımoli, 1976) (Text S1, S2 and Fig. S1
see supplementary material, which is available on the
Supplementary tab of the article’s Taylor & Francis On-
line page at http://dx.doi/10.1080/14772000.2012.748697).
These specimens include the holotypes of all previously
erected living and extinct species of Plagiodontia and Hy-
perplagiodontia, including the skull of the holotype of P.
aedium (MNHN 2M-MO-1982-894), which was previously
reported as being lost (Miller, 1928) but which has recently
been relocated in the MNHN collections (C. Callou, pers
comm., 2010). Much of this material has been previously
assigned to a given species of Plagiodontia or Hyperpla-
Plagiodontia fossil material (see Woods, 1989; Woods &
Ottenwalder, 1992; MacPhee et al., 2000) has been cata-
logued in open nomenclature. However, much of the zooar-
chaeological material from the northern Dominican Repub-
lic collected during the late nineteenth and early twentieth
centuries, as well as the undescribed craniodental material
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
Species diversity in Hispaniolan hutias 495
reported for P.araeum by Woods et al.(2001)andWilson
Sixty-seven standard craniodental measurements (40 cra-
nial measurements and 27 mandibular measurements) were
taken from complete specimens (Text S1, S2 and Fig. S1,
see supplementary material, which is available on the Sup-
plementary tab of the article’s Taylor & Francis Online page
at http://dx.doi/10.1080/14772000.2012.748697). As many
measurements as possible were also taken from incomplete
specimens, although fossil crania and mandibles were typi-
cally dissociated and broken. All measurements were taken
using dial callipers accurate to 0.02 mm.
Statistical analyses
Morphometric data from modern/historical Plagiodontia
specimens of known provenance (11 specimens from the
northern palaeo-island, 22 specimens from the southern
palaeo-island) were analysed using Multivariate Analysis
of Variance (MANOVA), Analysis of Variance (ANOVA),
and Principal Component Analysis (PCA) in R 2.10.1 (R
Development Core Team, 2011), to investigate whether the
primary genetic division between northern and southern
palaeo-island hutia populations identified by Brace et al.
(2012) is supported by congruent morphological differ-
entiation of craniodental characteristics, and to determine
the magnitude and pattern of such morphological differ-
entiation. Morphometric data for both populations were
also compared with measurements for the holotype skull
of P.aedium.OnlyonemodernPlagiodontia mandible
(LDUCZ-Z2716) is currently available from the eastern
Massif de la Selle–Sierra de Bahoruco region of the south-
ern palaeo-island, and so it was not possible to further in-
vestigate patterns of morphological differentiation across
Bond’s Line.
Genetic analysis was carried out on a small bone
sample obtained from the skull of the holotype of P.
aedium in a dedicated ancient DNA laboratory at Royal
Holloway University of London, to attempt to assign this
specimen to one of the three phylogenetically distinct
extant Plagiodontia lineages identified by Brace et al.
(2012). For details on extraction and PCR amplification
methods and the sequences of the six primer pairs used in
this analysis, see Brace et al.(2012).
Morphometric analysis of fossil material was further
conducted to investigate the taxonomic significance of pat-
terns of morphological variation shown by Late Quaternary
Plagiodontia specimens in comparison to the morpholog-
ical variation shown across living hutia populations. Di-
rect comparison with modern Plagiodontia specimens us-
ing multivariate statistical techniques cannot be conducted
easily due to the incomplete and fragmentary nature of
the majority of available fossil specimens and the varying
availability of potential measurement landmarks. The most
complete set of measurements for fossil specimens was
mandibular alveolar toothrow length (78 available spec-
imens). Although it can be difficult to assign individual
specimens to species accurately on the basis of single mea-
surements, as multiple measurements are required to create
non-overlapping separation at the species level, this large
measurement series should permit data to conform to nor-
mal distributions reflecting natural species size ranges.
Finite Mixture Analysis (FMA) using χ2tests with
the MIXDIST package in R version 2.10.1 (Macdonald
ent proposed taxonomic hypotheses to the distribution of
mandibular sizes shown by the available fossil measure-
ment data. Means and standard deviations of mandibular
alveolar toothrow lengths were calculated for each putative
species of Plagiodontia using the respective type series,
except for extant populations (aedium and hylaeum)where
more extensive and accurately identified modern specimen
measurement datasets were used. In order to incorporate
type series data for all putative fossil taxa into our analysis,
estimates of mandibular toothrow length for type specimens
established on the basis of skulls or maxillae with no asso-
ciated mandibles (P.caletensis,P.ipnaeum and P.velozi)
were generated from the close linear relationship between
mandibular and maxillary alveolar length in Plagiodon-
tia (mandible alveolar length =0.963 maxillary alveolar
length +0.566). For P.velozi,whichhasonlyoneavail-
able (and transformed) mandibular toothrow measurement
in the type series, no species-specific standard deviation
could be calculated, and so a proxy measure of standard
deviation was calculated from modern Plagiodontia spec-
imens (n=38). A series of 16 hypothetical taxonomies
(representing mixed distributions of multiple overlapping
normally distributed populations) with varying levels of
species diversity and synonymy were generated by separat-
ing the overall measurement dataset into different combi-
nations and numbers of putative species, with a cumulative
separation of progressively larger-bodied species (Table 2).
The combined fitted distributions that did not deviate signif-
icantly from the observed data were retained. Plagiodontia
araeum and Hyperplagiodontia stenocoralis were excluded
from this analysis and were evaluated separately, because
these taxa display distinctive craniodental characters that
easily differentiate them from the more morphologically
similar representatives of Plagiodontia sensu stricto (see
Following ident ific ati on of well-defined fossil Pla-
giodontia species, craniodental measurement data for mod-
ern/historical hutia samples were then reanalysed in order
to determine whether morphological variation between liv-
ing northern and southern palaeo-island hutia populations
is similar in magnitude to that between different Late Qua-
ternary species, or whether it should be better interpreted
as subspecies-level variation. This analysis was again con-
ducted using MANOVA and PCA, but this time using a
restricted measurement dataset available for fossil taxa.
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
496 J. Hansford et al.
Fig. 3. Principal component analysis for craniodental measure-
ments of northern palaeo-island (triangles) and southern palaeo-
island (circles) Plagiodontia aedium.Percentagevariationex-
plained by PCA: axis 1, 61.43%; axis 2, 8.50%.
Body mass estimates for extinct plagiodontines and other
rodents were calculated using the ‘all species’ regression
equation for mandibular toothrow length against body mass
in rodents given by Hopkins (2008, p. 235), using alveolar
toothrow lengths for all species.
Morphometric and genetic analysis of
living hutia populations
significant differences in craniodental characters between
living northern and southern palaeo-island hutia popula-
tions (F =25.802, P=0.155), and there is no sepa-
ration of the two populations in PCA using the entire
dataset (Fig. 3). Individual ANOVAs were used to iden-
tify and eliminate craniodental characters that were not
significantly different between northern and southern pop-
ulations, and a second MANOVA using the reduced subset
of 14 characters that differed between these populations
(Table 1) found a significant separation of the two popula-
tions (F =4.597, P=0.004).
Attempts to amplify mitochondrial DNA from the holo-
type of P. a e d i u m were unsuccessful. This is likely to be
due to the degraded nature of the DNA in this old mu-
seum sample. However, the holotype skull of P.aedium
is morphometrically closest to the southern palaeo-island
population in 12 of the 14 craniodental measurements that
differ significantly between the northern and southern pop-
ulations (Table 1).
Morphometric analysis of fossil hutia
Whereas the mandibular size distribution of mod-
ern/historical Plagiodontia specimens is unimodal, the
mandibular size distribution of Late Quaternary Plagiodon-
tia specimens is multimodal and displays a much broader
measurement spread (Fig. 4). Because both male and fe-
male Plagiodontia individuals are included in the uni-
modal modern/historical sample, the multimodal distribu-
tion shown by the fossil material must therefore represent
multiple species rather than sexual dimorphism. The χ2
analysis demonstrates that four of the hypothetical taxon-
omy FMA models fitted the observed fossil mandibular size
distribution, as their χ2goodness-of-fit tests are not sig-
nificantly different (Text S2, see supplementary material,
Table 1. Craniodental measurements displaying statistically significant differences between populations of Plagiodontia aedium from
the northern and southern palaeo-islands of Hispaniola, together with measurement data for the holotype of P.aedium,showingmean
and measurement range for each character, and Pvalues for ANOVA analysis of population-level differences. Measurements for which
the holotype is morphometrically closer to the northern population are indicated with asterisks.
Northern population (n=
11): mean, range (mm)
Southern population (n=
24): mean, range (mm) Pvalue
Min. breadth of zygomatic plate 4.15 (3.24–4.86) 4.92 (4.20–5.52) 0.000 4.98
Max. alveolar width of molar row (maxilla) 6.35 (5.15–7.04) 5.66 (4.52–6.92) 0.000 4.45
External width across bony palate at PM4 14.39 (12.28–15.33) 13.09 (11.86–15.37) 0.001 13.12
Crown width of pm4 4.91 (4.52–5.98) 4.36 (4.06–5.58) 0.001 4.20
Incisive foramina length6.79 (5.20–8.40) 6.06 (4.94–7.08) 0.006 6.52
Max. internal mandible height (base to tooth crown) 16.13 (14.87–17.52) 16.93 (15.62–18.20) 0.006 18.08
Max. alveolar width of molar row (mandible) 6.37 (5.15–7.04) 5.67 (4.52–6.92) 0.007 5.22
Crown width of m1 5.68 (4.62–6.50) 5.16 (4.60–6.06) 0.007 4.92
Crown width of m2 5.48 (4.68–6.12) 5.07 (4.44–5.76) 0.01 4.64
Max. internal height of mandible (base to alveolar lip) 13.47 (11.88–14.80) 14.28 (12.66–15.48) 0.012 15.04
Postpalatal length to occipital foramen 23.08 (20.14–25.04) 24.64 (21.72–27.76) 0.014 24.31
Crown width of PM4 5.42 (4.52–5.98) 4.97 (4.06–5.58) 0.017 4.82
Max. interorbital width22.72 (18.90–24.66) 21.29 (19.55–24.73) 0.028 22.26
Crown width of M1 5.50 (4.69–6.07) 5.18 (4.74–5.86) 0.038 5.08
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
Species diversity in Hispaniolan hutias 497
Fig. 4. Probability density histograms of mandibular size distributions for modern (top) and Late Quaternary fossil/zooarchaeological
(bottom) Plagiodontia specimens. The x-axis shows 0.5 mm bins ranked from the smallest available specimen. Size distributions of
holotype specimens of all putative species of Plagiodontia sensu stricto are indicated on the lower histogram (transformed mandibular
measurements used for P.caletensis,P.ipnaeum and P.velozi): filled square, P.spelaeum; cross-hatched diamond, P.aedium; filled
triangle, P.hylaeum; open triangle, P.caletensis; cross-hatched square, P.ipnaeum;opendiamond,P.velozi.
Table 2. Alternative taxonomic hypotheses of Late Quaternary species diversity in Plagiodontia tested using Finite Mixture analysis,
showing different possible species divisions (from one to six Plagiodontia species), and type series used to calculate species-specific
mean and SD for mandible alveolar lengths (SD for P.aedium calculated using modern specimen dataset). Asterisks next to taxonomic
hypotheses indicate that these fit the observed fossil mandibular size distribution; asterisks next to mandibular lengths indicate that these
represent transformed maxillary measurements (see text for regression equation).
Specimen Mandibular alveolar
Species 1 2 3 4 5 6 7 891011 12 13 14 1516 number length (mm)
spelaeum 1111111 1 1 1 1111 1 1USNM253160 18.86
USNM 253161 18.35
USNM 253162 16.98
USNM 253163 17.85
USNM 253165 18.34
aedium 1211112 2 2 2 2222 2 2MNHN2M-
hylaeum 1221113 2 2 2 3333 2 3USNM239887 18.80
caletensis 1222113 3 2 2 4334 3 4ROR3 24.46
ROR 4 23.90
ROR 5 23.92
ipnaeum 1222213 3 3 2 4435 4 5USNM254375 25.11
USNM 254380 27.36
velozi 1222223 3 3 3 4445 5 6USNM175230 29.13
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
498 J. Hansford et al.
which is available on the Supplementary tab of
the article’s Taylor & Francis Online page at
http://dx.doi/10.1080/14772000.2012.748697). All of
these models have statistical merit. All retained models
show both P.aedium (including P.hylaeum)andP.
spelaeum as distinct taxa, so that both should be uncontro-
versially recognized as valid species. However, taxonomic
differentiation in the upper end of the mandibular size
distribution is less well defined, because of the smaller
number of specimens of large-bodied extinct Plagiodontia
individuals currently available in museum collections.
Three of the four retained models show only one larger
species, but more than one larger species is supported in
the fourth retained model.
The extinct Plagiodontia species best defined by FMA,
P.spelaeum, was then used to further investigate relative
levels of variation between living hutia populations in the
wider context of Late Quaternary morphological diversity.
It was possible to use 18 mandibular characters for the
type series of P.spelaeum (Table 3). Using these characters
alone, the living northern and southern palaeo-island pop-
ulations (n=33) could still be separated using MANOVA
(F =4.032, P<0.001), and these two populations were
also completely distinguished from the P.spelaeum sample
(n=5) (F =5.917, P<0.001). There was full separation
of P.spelaeum and partial separation of living northern and
southern hutia populations in PCA on axis 2 (Fig. 5). Of
the characters used to differentiate between P.spelaeum
and living hutias, internal toothrow height at pm4 and m3
do not show a strong loading on PCA axis 1 (eigenvector
values =0.0847 and 0.0496), and so are unlikely to be of
substantial use for species identification on their own; how-
ever, whereas other individual mandibular characters are
of greater usefulness (eigenvector values =0.273–0.225),
multivariate analysis using the entire character set will pro-
vide much greater accuracy in assigning hutia specimens to
Body mass estimates for all currently recognized extinct
and extant Hispaniolan plagiodontine hutias, together with
other large-bodied Greater Antillean rodent species, are
given in Table 4.
Revised taxonomy of plagiodontine
Although genetic analysis of the almost 200-year-old holo-
type specimen of Plagiodontia aedium unfortunately failed
to yield amplifiable DNA, morphometric analysis indicates
that the holotype skull is much closer to modern speci-
mens from the southern palaeo-island in almost all of the
craniodental characters that differ significantly between liv-
ing northern and southern populations, and it clusters with
modern southern palaeo-island specimens in the PCA anal-
ysis of P.aedium and P.spelaeum (Fig. 5). The name
Fig. 5. Principal component analysis for mandibular measure-
ments of Plagiodontia spelaeum (squares), northern palaeo-island
P.aedium (triangles), southern palaeo-island P.aedium (circles),
and holotype of P.aedium (open diamond). Percentage variation
explained by PCA: axis 1, 63.65%; axis 2, 10.24%.
Plagiodontia aedium can therefore be interpreted as the
valid name for the southern palaeo-island taxon, consistent
with the usage of previous authors. The likely collection
of the holotype specimen from the southern palaeo-island
is supported by other early reports of hutias from south-
ern Haiti, suggesting increased historical collection effort
in this region. M´
eric Moreau de Saint-M´
ery wrote of an
animal by the name ‘d’agoutis’ seen in April 1788 in a
fallen tree near Roseaux, immediately north of the Mas-
sif de la Hotte (Moreau de Saint-M´
ery, 1797), and sev-
eral other hutia specimens collected in the early twentieth
century also come from the region west of Bond’s Line
(Woods, 1981). PCA further demonstrates that levels of
morphological variation within living hutias are lower than
levels between well-defined Late Quaternary species (P.
spelaeum versus the combined sample of aedium and hy-
laeum specimens), and so although northern and southern
populations can be defined on the basis of both morphome-
tric and genetic data, it is more appropriate to interpret this
as subspecies-level variation rather than elevating northern
and southern taxa to separate species status. We therefore
follow the taxonomic conclusions of Brace et al.(2012)
by recognizing extant northern and southern subspecies of
Hispaniolan hutia.
Although one of our retained FMA models recognizes
more than one larger-bodied extinct Plagiodontia species,
most of our retained models only recognize a single
larger-bodied species. Our interpretation of taxonomic
boundaries in extinct hutias may have been influenced by
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
Species diversity in Hispaniolan hutias 499
Table 3. Craniodental measurements for Plagiodontia aedium (including both northern and southern palaeo-island taxa) and
P.spelaeum used to test for differences between these species, showing mean and measurement range for each character, and Pvalues
for ANOVA analysis of species-level differences.
Mandibular measurement
aedium (n=44): mean,
range (mm)
spelaeum (n=5): mean,
range (mm) Pvalue
Length of symphysis 24.09 (19.94–28.68) 17.32 (16.70–17.98) <0.001
Length of diastema 13.88 (9.96–16.10) 9.29 (8.58–9.76) <0.001
Length from incisor base to posterior bulb of incisor root 23.64 (19.50–28.62) 17.23 (16.34–18.10) <0.001
Crown length of molar row 19.30 (16.26–22.12) 15.80 (15.20–16.62) <0.001
Alveolar length of molar row 21.79 (18.45–24.94) 18.08 (16.98–18.86) <0.001
Maximum alveolar width of molar row 5.93 (4.08–7.04) 5.10 (4.90–5.34) 0.007
Crown length of pm4 5.71 (4.50–7.26) 4.65 (4.18–4.86) <0.001
Crown width of pm4 4.58 (3.54–5.48) 3.90 (3.57–4.58) 0.002
Crown length of m1 4.69 (3.95–5.65) 4.17 (3.47–4.86) 0.003
Crown width of m1 5.35 (3.94–6.50) 4.30 (3.94–4.60) <0.001
Crown length of m2 4.39 (3.86–5.25) 3.46 (3.18–3.77) <0.001
Crown width of m2 5.20 (4.14–6.12) 4.32 (3.88–4.60) <0.001
Crown length of m3 3.93 (3.00–4.70) 3.10 (2.64–3.54) <0.001
Crown width of m3 4.55 (3.44–5.48) 3.43 (3.24–3.62) <0.001
Maximum height of mandible (base to alveolar lip) 14.23 (11.88–16.94) 10.89 (10.58–11.08) <0.001
Maximum height of mandible (base to tooth crown) 16.95 (14.87–19.92) 13.08 (13.03–13.15) <0.001
Height of toothrow at pm4 (internal aspect) 3.06 (1.46–4.18) 2.21 (1.90–2.66) 0.004
Height of toothrow at m3 (internal aspect) 1.15 (0.50–2.00) 0.89 (0.56–1.52) 0.112
artefactually small standard deviations calculated from
the small sample sizes available for most fossil species,
but we conservatively suggest that only one larger-bodied
Plagiodontia species may have existed on Hispaniola, as
our existing fossil measurement data only provide very
weak support for the existence of multiple large-bodied
species. Although there is relatively little well-preserved
material referable to this large-bodied species available
in museum collections, it is represented by an almost
complete skull with associated mandibles (UF 225374)
Table 4. Body mass estimates and measurements for living and extinct Hispaniolan plagiodontine hutias, and comparison with body
masses of the largest rodents from other Greater Antillean islands. Asterisks indicate extant species. All data for Hispaniolan species
were calculated using the Hopkins (2008) regression equation for mandibular toothrow length against body mass, to permit direct
comparison of body masses between the living Plagiodontia aedium and other extinct species. Body mass data for the extant Capromys
pilorides was obtained from direct measurement of living individuals rather than the Hopkins (2008) regression equation.
Species Family Island
Body mass (kg):
mean, range Data source
Plagiodontia spelaeum Capromyidae Hispaniola 1.57 (1.32–1.77) Type series (USNM 253160–253163,
Capromyidae Hispaniola 1.70 (1.32–2.01) Woods (1989), n=21
Plagiodontia aediumCapromyidae Hispaniola 2.63 (1.56–3.85) Total modern sample, n=34
Capromyidae Hispaniola 4.09 (2.35–5.36) USNM 205907–205908,
UF 28036–28037, MHND: 5
unnumbered specimens
Plagiodontia velozi Capromyidae Hispaniola 5.62 (5.31–5.93) Holotype (USNM 175230,
transformed measurement),
UF 225374
Capromys piloridesCapromyidae Cuba 3.09 (1.27–6.93) Borroto-P´
aez & Mancina (2011),
means and ranges from 14
Macrocapromys acevedo Capromyidae Cuba 3.18 (2.75–3.62) Silva Taboada et al. (2007), n=2
Macrocapromys latus Capromyidae Cuba 3.66 Silva Taboada et al. (2007), n=1
Heptaxodontidae Puerto Rico 8.45 (6.73–9.56) Anthony (1918), n=11
Clidomys osborni Heptaxodontidae Jamaica 18.02 (14.92–21.50) Morgan & Wilkins (2003), n=2
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
500 J. Hansford et al.
from Trouing Lanj Genti, a cave site in the Massif de la
Hotte (Woods, 1989; Figs 10–12, 14–15).
The appropriate taxonomic name to use for the ex-
tinct large-bodied Plagiodontia species can be determined
through consideration of intraspecific size variation in the
single extant species, P.aedium,whichshowsasizerange
of 18.02–24.94 mm (n=34) for mandibular alveolar
toothrow length. All observed or transformed mandibular
alveolar toothrow lengths for the type series of P.caleten-
sis (23.90–24.47 mm; n=3) fall within the upper size
range for modern P.aedium specimens, so that this species
can be interpreted as a junior synonym of P.aedium.The
transformed maxillary toothrow length of the holotype of
P.ipnaeum (25.11 mm) falls just outside the upper range
of modern hutia toothrow lengths, but is extremely close in
size, and we consider that this specimen may therefore also
merely represent a large individual of P.aedium;theevi-
dence for P.ipnaeum representing a valid extinct species
is very weak, and we conservatively aggregate it with P.
aedium and interpret it as a possible further junior syn-
onym of the living species. The only previously described
extinct species that falls far outside the upper size distribu-
tion shown by modern P.aedium specimens is P.velozi,
the holotype of which has a transformed maxillary toothrow
length of 29.13 mm. We therefore recognize P.velozi as the
only available name that can be used for the extinct large-
bodied Plagiodontia species recognized in our morphome-
tric analysis. Further taxonomic assessment of the validity
of the various proposed extinct large-bodied Plagiodontia
species must await the collection of more complete fossil
hutia material than is currently available, and/or ancient
DNA analysis.
Our analyses recognize three distinct size-differentiated
Plagiodontia species from the Late Quaternary of His-
paniola: the extinct P.spelaeum Miller, 1929; the extant
P.aedium Cuvier, 1836 (comprising two allopatric sub-
species, P.aedium aedium Cuvier, 1836 from the south-
ern palaeo-island, and P.aedium hylaeum Miller, 1928
from the northern palaeo-island); and the extinct P.velozi
ımoli, 1976 (Figs 6–16). Despite differing significantly in
size, these three species display very similar dental mor-
phologies, and so can unquestionably be assigned to the
same genus. In contrast, although almost all of the speci-
mens of both ‘Plagiodontiaaraeum and Hyperplagiodon-
tia stenocoralis that are currently available for analysis
are mandibles, so that skull characteristics cannot be com-
pared, all specimens assigned to these taxa (Text S1, S2 and
Fig. S1, see supplementary material, which is available on
the Supplementary tab of the article’s Taylor & Francis On-
line page at http://dx.doi/10.1080/14772000.2012.748697)
display a closely similar morphology that is very distinct
from other Plagiodontia species. Representatives of Pla-
giodontia sensu stricto have cheek teeth with approxi-
mately square-shaped occlusal surfaces that are approxi-
mately as long as they are wide, and with relatively open
obliquely oriented flexids separated by broad re-entrant
folds (Figs 6–12). The cheek teeth of ‘P.’ araeum and H.
stenocoralis show the same number and orientation of flex-
ids and re-entrant folds, but these are extremely strongly
anterolabially–posterolingually compressed and the trans-
verse width of the occlusal surface is markedly greater than
the anteroposterior length, presenting a lozenge-shaped
rather than a square-shaped outline (Figs 17, 19–21). The
cheek teeth of ‘P.’ araeum and H.stenocoralis are also far
more brachydont than in representatives of Plagiodontia
sensu stricto, with the occlusal surface raised up only a
couple of millimetres above the jawline (Fig. 18). These
taxa are certainly morphologically closer to Plagiodon-
tia sensu stricto than they are to the other Late Quater-
nary Hispaniolan capromyids Hexolobodon,Isolobodon or
Rhizoplagiodontia (e.g. root of lower incisor terminates
below m1; cheek teeth unrooted and lacking external ce-
ment bands; re-entrant folds oblique; see Woods, 1989 for
further discussion); however, supporting the original sug-
gestion of Ray (1964), they are distinct at the generic level
from Plagiodontia.Theavailablenameforthisgenusis
Hyperplagiodontia R´
ımoli, 1976.
Hyperplagiodontia araeum was described on the basis of
an isolated deciduous cheektooth, and so comparison be-
tween the hypodigms of H.araeum and H.stenocoralis is
not straightforward. However, there are no consistent mor-
phological differences between any of the available spec-
imens representing Hyperplagiodontia that are available
for study, with the type series of H.stenocoralis show-
ing almost identical mandibular alveolar toothrow lengths
(25.4–26.8 mm) to specimens assigned to ‘P.’ araeum in
the UF collections (25.5–26.0 mm) (R´
ımoli, 1976; Woods,
1989). We therefore provisionally interpret these two taxa
as conspecific, with H.stenocoralis representing a junior
synonym of H.araeum, pending further fossil collection
from the type locality of H.araeum.
The results of our morphometric analysis of mod-
ern/historical Plagiodontia craniodental measurement data
are congruent with recently published genetic data (Brace
et al., 2012), and support the previous recognition of a
major biogeographic division of living hutia populations
into distinct subspecies of P.aedium associated with His-
paniola’s northern and southern palaeo-islands. It is pos-
sible that some of the observed craniodental differences
between northern and southern hutia populations may be
associated with different feeding adaptations (e.g. mas-
ticatory musculature; see Woods & Howland, 1979), as
Wood s (19 81) p r op o sed d iet ary d iff ere nce s bet wee n the se
two populations, suggesting that animals in the Massif de
la Hotte feed largely on the ground whereas animals in
the north-eastern Dominican Republic live and feed in the
canopy. However, these observed differences may instead
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
Species diversity in Hispaniolan hutias 501
Figs 6–12. 6,9,Plagiodontia aedium, MNHN 2M-MO-1982-894 (holotype); 6,undersideofskullshowingocclusalviewofmaxillary
toothrows; 9, paired hemimandibles showing occlusal view of mandibular toothrows. 78,P.spelaeum,cavenearSt.MicheldelAtalaye,
epartement de l’Artibonite, Haiti; 7, USNM 303751, maxilla with incomplete toothrows; 8, USNM 253165 (paratype), left hemimandible
showing occlusal view of mandibular toothrow. 1012,P.velozi;10, USNM 175230 (holotype), cave near St. Michel de l’Atalaye,
epartement de l’Artibonite, Haiti; maxilla with incomplete toothrows. 1112, UF 225374, Trouing Lanj Genti, Massif de la Hotte,
Haiti; 11, underside of skull showing occlusal view of maxillary toothrows; 12, left hemimandible showing occlusal view of mandibular
toothrow. Scale bar =10 mm.
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
502 J. Hansford et al.
Figs 13–16. Lateral and dorsal views of the skulls of Plagiodontia aedium (holotype, MNHN 2M-MO-1982-894) (13,16)andP.velozi
(UF 225374, Trouing Lanj Genti, Massif de la Hotte, Haiti) (14,15). Lateral view of P.velozi skull reversed for comparison with P.
aedium skull. Scale bar =10 mm.
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
Species diversity in Hispaniolan hutias 503
Figs 17–21. Hyperplagiodontia araeum.17, MCZ 7675 (holotype), left DP4, unnamed cave 2 km southeast of Rancho de la Guardia,
Provincia El´
ıas Pi˜
na, Dominican Republic. Scale bar =2mm.18,20, UF 28036, Trou Zombie, ˆ
Ile de la Gonˆ
ave, Haiti; 18, external
view of mandible showing extremely brachydont toothrow; 20, occlusal view of mandibular toothrow. Scale bar =10 mm. 19, USNM
205908 (paratype of H.stenocoralis), cave near St. Michel de l’Atalaye, D´
epartement de l’Artibonite, Haiti; occlusal view of mandibular
toothrow showing m1-3. Scale bar =10 mm. 21, MHND (unnumbered specimen), Cueva de la Virgen, Carreterra Mella, Distrito Nacional,
Dominican Republic; occlusal view of mandibular toothrow. Scale bar =10 mm.
reflect recent behavioural responses to human-caused forest
loss in Haiti rather than true evolutionary differences, and
hutias from the Massif de la Hotte have also been reported
to feed on a wide range of canopy plant species, including
tree bark, twigs, branches, leaves, buds and fruits (Woods
& Ottenwalder, 1992).
Quantitative analysis of body size differences and qual-
itative analysis of dental characters across a large sample
of Late Quaternary fossil hutia specimens provide a new
framework for species delimitation and infrageneric varia-
tion within the Plagiodontinae, identifying a different series
of valid extinct species and genera to previous taxonomic
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
504 J. Hansford et al.
treatments. Although the only extinct plagiodontine not
considered in this assessment, Rhizoplagiodontia lemkei,
is apparently biogeographically restricted to the Massif de
la Hotte (Woods, 1989), as is the recently described extinct
primate Insulacebus toussaintiana (Cooke et al., 2011), all
of the extinct species of Plagiodontia and Hyperplagiodon-
tia recognized in our analyses have been recorded from
both the northern and southern palaeo-islands of Hispan-
iola. Both P.spelaeum and P.velozi are recorded from
the northern palaeo-island from St. Michel de l’Atalaye,
Haiti (Miller, 1929a), and from the southern palaeo-island
from the Massif de la Hotte (Woods, 1989; P.spelaeum
represents the smallest size morph in the UF Plagiodon-
tia sample from this region). Hyperplagiodontia araeum is
recorded from the northern palaeo-island from Rancho de
la Guardia (Ray, 1964) and Cueva de la Virgen, Dominican
Republic (MHND collections, unpubl.) and from St. Michel
de l’Atalaye, Haiti (Miller, 1929a;R
ımoli, 1976); from the
southern palaeo-island from Cueva de Col´
on, Dominican
Republic (Trias et al., 1997) and from Trouing Marassa
(Miller, 1930; MacPhee et al., 2000) and other sites east
of Bond’s Line in Haiti (Woods, 1989); and also from Trou
Zombie, ˆ
Ile de la Gonˆ
ave (Woods, 1989) (Fig. 1). It is pos-
sible that some or all of these species may have displayed
similar biogeographic differentiation across Hispaniola at
the subspecific or population level as demonstrated for P.
aedium, but morphometric resolution is currently too poor
to test this hypothesis given theincomplete and fragmentary
nature of the available fossil material.
Our improved understanding of Late Quaternary species
diversity in Hispaniolan hutias has significant implications
for interpreting patterns of human-caused extinction in
the Caribbean land mammal fauna. Robust last-occurrence
dates are still not known for any extinct plagiodontines,
making it difficult to correlate possible causative extinc-
tion drivers within a temporal framework. However, both
P.spelaeum (as the smaller Plagiodontia size morph from
caves in the Massif de la Hotte) and R.lemkei have been
reported from recent cave deposits in stratigraphic asso-
ciation with the remains of introduced black rats Rattus
rattus (Woods, 1989; Woods & Ottenwalder, 1992), sug-
gesting that they both persisted into the European historical
era. None of these species has yet been shown to have sur-
vived well beyond first European contact. The sixteenth
century Spanish chronicler Gonzalo Fernandez de Oviedo
es reported the occurrence of a large edible rodent on
Hispaniola known locally as the ‘quemi’, which may rep-
resent P.velozi,H.araeum,oranothernow-extinctlarge-
bodied endemic capromyid (Miller, 1929b). More recently,
Wood s et al.(1985)describedaseriesofindependentre-
ports from four communities in the Sierra de Bahoruco of
tinct from P.aedium (greater body size, relatively long tail)
and locally called the ‘comadreja’ (‘weasel’ in Spanish),
which these authors considered might represent P.velozi.
The most recent of these reports dated from the early 1970s.
We con sid er it p o ss i bl e on th e bas i s of t h ese r epo rts t hat
P.velozi or another large-bodied hutia species may have
persisted into the twentieth century in this relatively inac-
cessible region, although there is unfortunately no evidence
for its continued survival.
Recognition of P.velozi as a distinct, recently extinct
species, and improved understanding of the morphology
and status of H.araeum,revealsthatthelargestrepresen-
tatives of Hispaniola’s extinct hutia fauna were comparable
in size to the largest Cuban capromyids and approached
the size of the large heptaxodontid Elasmodontomys from
Puerto Rico (Table 4), suggesting similar patterns of mam-
malian niche availability and evolutionary constraint be-
tween these three neighbouring islands. Although Jamaica’s
Late Quaternary mammal fauna included the much larger
extinct heptaxodontid Clidomys (Table 4), this different pat-
tern of rodent body size evolution may reflect competitive
release in the Jamaican rodent fauna in the absence of ter-
restrial megalonychid sloths on this island.
Our statistical support for distinct species status for
P.spelaeum further demonstrates that Hispaniola’s pla-
giodontine fauna has lost both its largest and smallest rep-
resentatives, with only the medium-sized P.aedium surviv-
ing to the present. Similar trends of body size selectivity
in extinction risk are also shown more widely across the
Caribbean land mammal fauna. All of the small-bodied ne-
sophontid island-shrews (Nesophontes;<150 g) and Cuban
and Hispaniolan spiny rats (Boromys,Brotomys;400 g)
have become extinct, as have all of the larger-bodied mega-
lonychid sloths (>7kg),andthesolenodonshavealso
lost their largest and smallest species, Solenodon arredon-
doi (990 g) and S.marcanoi (230 g), with only the two
intermediate-sized species still extant (body mass estimates
from Turvey & Fritz, 2011). Although non-volant terres-
trial Caribbean mammal species from all size classes have
become extinct during the Late Quaternary, all surviving
representatives fall within a medium weight range of c.
0.5–3.0 kg (Borroto-P´
aez & Mancina, 2011; Table 4). This
relationship between body mass and extinction risk in the
Caribbean land mammal fauna may differ from patterns
seen in other geographical areas. In particular, it has been
suggested that mammal species of intermediate body mass
(with a proposed ‘critical weight range’ of 35 g–5.5 kg)
conversely have an elevated risk of extinction in Australia
(Burbidge & McKenzie, 1989), although the validity and
generality of this proposed pattern continues to be debated
(e.g. Cardillo & Bromham, 2001; Johnson & Isaac, 2009;
Chisholm & Taylor, 2010).
The apparent increased survivorship of medium-sized
Caribbean mammals may reflect the specific combina-
tion of anthropogenic threats that have existed in the re-
gion during the historical period and recent prehistory.
Patterns of mid-Holocene exploitation of terrestrial ani-
mal resources by early Amerindian settlers remain poorly
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
Species diversity in Hispaniolan hutias 505
understood (Newsom & Wing, 2004), with multiple large-
bodied sloths (Megalocnus,Neocnus)knowntohavesur-
vived in the Greater Antilles for over a millennium fol-
lowing first human arrival in the region (Steadman et al.,
2005; MacPhee et al., 2007), and considerable exploitation
of several hutia taxa is also shown in the pre-Columbian
archaeological record (e.g. Colten et al., 2009). However,
prey-selection patterns of modern aboriginal subsistence
hunters elsewhere in the world (Lyons et al., 2004), and
possibly also wider patterns of Late Quaternary mammal
extinction (Grayson & Meltzer, 2002), suggest that larger
mammals may well have been preferentially hunted by early
Amerindian settlers. Large mammals are also known to be
at higher risk of extinction due to traits such as reproduc-
tive rate scaling negatively with body mass (Cardillo et al.,
2005), making these species more vulnerable to human im-
pacts due to their slower life histories. Conversely, smaller
mammals would have been increasingly vulnerable to com-
petition and predation by invasive mammals that reached
the Caribbean islands with the arrival of Europeans, no-
tably black rats but also more recent introductions such
as mongoose Herpestes javanicus (MacPhee & Flemming,
1999). In contrast, although several living hutia and solen-
odon species are known to be the target of recent subsis-
tence hunting activities and are also killed by feral dogs and
other invasives (e.g. Woods, 1981; Oliver, 1982; Wilkins,
2001; Turvey et al., 2008, in press), medium-sized mam-
mals are likely to have been somewhat more resilient to
these threats. We suggest that the apparent pattern of body
size selectivity in the Late Quaternary Caribbean mammal
extinction event that we have described should be called the
‘Goldilocks Hypothesis’, as the surviving members of the
Caribbean land mammal fauna are neither too large nor too
small, but instead are ‘just right’. This proposal requires
further investigation within a rigorous statistical and phylo-
genetically explicit framework, in order to quantify the true
nature of the relationship between body size and extinction
risk in Caribbean mammals.
Logistical support for fieldwork in Haiti to collect hutia
samples was provided by Soci ´
ıti, and par-
ticular thanks go to Helen Meredith, Paul Scofield, Eladio
Fer nan dez, Os ´
eus, Philippe Bayard, Jessie Haspil,
Paul Judex Ezouardin, Potau Rosevalt, Frederique Ch´
and Jos´
e Ottenwalder. We thank Jack Ashby and Mark Car-
nall (LDUCZ), Cecile Callou (MNHN), Linda Gordon and
David Bohaska (USNM), William Amaral and Catherine
Weis el (M CZ) , Ric h ar d H ulb ert a nd Ca n da c e McC affer y
(UF), Miguel Santiago Nu˜
nez Novas (MHND) and Re-
nato R´
ımoli for access to further hutia specimens. Fur-
ther assistance was provided by Ian Barnes. Funding for
this research was provided by the Natural Environment Re-
search Council (NERC Postdoctoral Research Fellowship
IP/1075/1108), the Royal Society (University Research Fel-
lowship UF080320), the University of York (MRes NERC
studentship grant), SYNTHESYS2 made available by the
European Community Research Infrastructure under FP7
(‘Synthesis of Systematic Resources’, 226506-CP-CSA-
Infra), and the UK Darwin Initiative project 17025 (‘Build-
ing evidence and capacity to conserve Hispaniola’s endemic
land mammals’).
ALLEN, G.M. 1942. Extinct and Vanishing Mammals of the West-
ern Hemisphere with the Marine Species of all the Oceans.
American Committee for International Wild Life Protection,
Special Publication 11, Washington, DC.
ANDERSON, S. 1965. Conspecificity of Plagiodontia aedium and
P.hylaeum (Rodentia). Proceedings of the Biological Society
of Washington 78,9598.
ANTHONY, H.E. 1918. The indigenous land mammals of Porto
Rico, living and extinct. Memoirs of the American Museum of
Natural History 2,331435.
ANDEZ, M.M. 2002. Valor
omico de la morfometr´
ıa del esqueleto craneal y
postcraneal en tres especies de jut´
ıas de Cuba (Rodentia:
Capromyidae). Revista Biologia 16,150158.
AEZ,R.&MANCINA, C.A., Eds. 2011 Mam´
ıferos en
Cuba. UPC Print, Vaasa, Finland.
M.G. & TURVEY, S.T. 2012. Population history of the His-
paniolan hutia Plagiodontia aedium (Rodentia: Capromyi-
dae): testing the model of ancient differentiation on a geotec-
tonically complex Caribbean island. Molecular Ecology 21,
BURBIDGE,A.A.&MCKENZIE, N.L. 1989. Patterns in the mod-
ern decline of Western Australia’s vertebrate fauna: causes
and conservation implications. Biological Conservation 50,
CARDILLO,M.&BROMHA M, L. 2001. Body size and risk of ex-
tinction in Australian mammals. Conservation Biology 15,
2005. Multiple causes of high extinction risk in large mammal
species. Science 309,12391241.
CHISHOLM,R.A.&TAY L O R , R. 2010. Body size and extinction risk
in Australian mammals: an information-theoretic approach.
Austral Ecology 35,616623.
ceramic faunal exploitation at the Las Obas site, Cuba.
Bulletin of the Peabody Museum of Natural History 50,
C. 2005. Revisi´
on taxon´
omica del g´
enero Nesophontes (In-
sectivora: Nesophontidae) en Cuba: an´
alisis de los caracteres
ostico. Monografies de la Societat d’Hist`
oria Natural de
les Balears 12,95100.
monkey from Haiti and the origins of the Greater Antillean
primates. Proceedings of the National Academy of Sciences
USA 108, 2699–2704.
Plagiodonte des habitations. Annales des Sciences Naturelles
2, 347–353.
DEBOOY, T. 1919. Santo Domingo kitchen-midden and burial
mound. Indian Notes and Monographs 1919, 107–137.
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
506 J. Hansford et al.
IAZ-FRANCO, S. 2001. Situaci
on taxon´
omica de Geocapromys
megas (Rodentia: Capromyidae). Caribbean Journal of Sci-
ence 37,7280.
FISHER,D.O.&BLOMBERG, S.P. 2011. Correlates of rediscovery
and the detectability of extinction in mammals. Proceedings
of the Royal Society B 278, 1090–1097.
FLOWER, S.S. 1929. List of vertebrated animals exhibited in the
gardens of the Zoological Society of London, 1828–1927.
Centenary edition, vol. 1. Mammals. Zoological Society of
London, London.
GIFFORD,M.E.&LARSON,A.2008.In situ genetic differentia-
tion in a Hispaniolan lizard (Ameiva chrysolaema): a multilo-
cus perspective. Molecular Phylogenetics and Evolution 49,
2004. Population structure and history of a phenotypically
variable teiid lizard (Ameiva chrysolaema) from Hispaniola:
the influence of a geologically complex island. Molecular
Phylogenetics and Evolution 32,735748.
vironments of the Caribbean region. Systematic Botany 28,
GRAYSON,D.K.&MELTZER, D.J. 2002. Clovis hunting and large
mammal extinction: a critical review of the evidence. Journal
of World Prehistory 16,313359.
HEDGES, S.B. 1999. Distribution patterns of amphibians in the
West Indies. In: DUELLMAN, W.E., Ed., Pat te rn s of D is tr i-
bution of Amphibians: A Global Perspective.JohnsHopkins
University Press, Baltimore, MD, pp. 211–254.
HOPKINS, S. S. 2008. Reassessing the mass of exceptionally large
rodents using toothrow length and area as proxies for body
mass. Journal of Mammalogy 89, 232–243.
raphy of the Caribbean region: implications for Cenozoic
biogeography. Bulletin of the American Museum of Natural
History 238,195.
2011. 2011 IUCN Red List of Threatened Species.IUCN,
Gland, Switzerland [, accessed 21 May
JOHNSON,C.N.&ISAAC, J.L. 2009. Body mass and extinction risk
in Australian marsupials: the ‘critical weight range’ revisited.
Australian Ecology 34,3540.
aedium,withasynopsisofrelatedspecies.Proceedings of the
Biological Society of Washington 61,6976.
MCFARLAND,K.&FERNANDEZ,E.2006.Birds of the Do-
minican Republic and Haiti. Christopher Helm, London.
LYO NS ,S.K.,SMITH,F.A.&BROWN,J.H.2004.Ofmice,mastodons
and men: human-mediated extinctions on four continents.
Evolutionary Ecology Research 6, 339–358.
MACDO NALD,P.&DU, J. 2010. Mixdist: finite mixture dis-
tribution models. R package version 0.5-3 [http://CRAN., accessed 21 May 2012].
MACPHEE, R.D.E. 2009. Insulae infortunatae: establishing a
chronology for Late Quaternary mammal extinctions in the
West Indies. In: HAY N E S , G., Ed., American Megafaunal Ex-
tinctions at the End of the Pleistocene. Springer, Dordrecht,
pp. 169–193.
MACPHEE, R.D.E. & FLEMMING,C.1999.Requieternam:the
last five hundred years of mammalian species extinctions. In:
MACPHEE, R.D.E., Ed., Extinctions in Near Time: Causes,
Contexts, and Consequences. Kluwer Academic/Plenum,
New York, NY, pp. 333–371.
MACPHEE, R.D.E, WHITE,J.L.&WOODS,C.A.2000.Newmega-
lonychid sloths (Phyllophaga, Xenarthra) from the Quaternary
of Hispaniola. American Museum Novitates 3303,132.
O. 2007. Prehistoric sloth extinctions in Cuba: implications of
anew“last”appearancedate.Caribbean Journal of Science
geologic and tectonic development of Hispaniola. In: MANN,
P. , D RAPER,G.&LEWIS, J.F., Eds., Geologic and Tectonic De-
velopment of the North America – Caribbean Plate Boundary
in Hispaniola.GeologicalSocietyofAmericaSpecialPubli-
cation 262, pp. 1–51.
facies distribution in the southern peninsula of Haiti and the
Barahona Peninsula, Dominican Republic, and its relations
concerning tectonic evolution of the La Selle-Baoruco block.
Caribbean Geology, Collected Contributions 9,124.
Quaternary extinct mammals from caves in Sanchez Ramirez
Province, Dominican Republic. Caribbean Journal of Science
36, 163–166.
between morphological and genetic differentiation in Cuban
hutia (Rodentia, Capromyinae): species fragmentation on the
area and the problem of excessive taxonomic status problem.
Russian Journal of Genetics 46, 191–197.
MILLER,G.S.JR.1916a. Note on the indigenous rodent of Santo
Domingo. Proceedings of the Biological Society of Washing-
ton 29,47.
MILLER, G.S. Jr. 1916b. Bones of mammals from Indian sites in
Cuba and Santo Domingo. Smithsonian Miscellaneous Con-
tributions 66(12), 1–10.
MILLER,G.S.JR. 1928. The rodents of the genus Plagiodontia.
Proceedings of the United States National Museum 72,18.
MILLER,G.S.JR. 1929a. A second collection of mammals from
caves near St. Michel, Haiti. Smithsonian Miscellaneous Con-
tributions 81(9), 1–30.
MILLER,G.S.JR. 1929b. Mammals eaten by Indians, owls, and
Spaniards in the coast region of the Dominican Republic.
Smithsonian Miscellaneous Contributions 82(5), 1–16.
Hispaniola. Smithsonian Miscellaneous Collections 82,113.
Desmarest (sens. Ampl.) und Plagiodontia Cuvier. Mitteilun-
gen aus dem Hamburgischen Zoologischen Museum und In-
stitut in Hamburg 48,48118.
ERY, M.L.E. 1797 (reprint 1958). Description
topographique, physique, civile, politique et historique de la
parte Francaise de l’Isle de Saint-Domingue. 3 vols. Soci-
ete de l’Histoire des Colonies Francaises et Librarie Larose,
(Heptaxodontidae) from a Late Quaternary cave deposit in
Jamaica. Caribbean Journal of Science 39,3441.
NEWSOM,L.A.&WING,E.S.2004.On Land and Sea: Native
American Uses of Biological Resources in the West Indies.
The University of Alabama Press, Tuscaloosa.
OLIVER, W.L.R. 1982. The coney and the yellow snake: the distri-
bution and status of the Jamaican hutia Geocapromys brownii
and the Jamaican boa Epicrates subflavus.Dodo 19,633.
EVELOPMENT CORE TEAM.2011.R: A Language and Environ-
ment for Statistical Computing. R Foundation for Statistical
Computing, Vienna.
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
Species diversity in Hispaniolan hutias 507
RAY , C.E. 1964. A new capromyid rodent from the Quaternary of
Hispaniola. Breviora 203,14.
IMOLI,R.O.1976.Roedores fosiles de la Hispaniola. Universidad
Central del Este, San Pedro de Macor´
ıs, Dominican Republic.
SALAZAR, L. 1977. Notas generales sobre Plagiodontia sp.
con commentarios sobre los especimenes obtenidos por el
SCHWARTZ, A. 1980. The herpetogeography of Hispaniola, West
Indies. Studies on the fauna of Curac¸ao and other Caribbean
Islands 61, 86–127.
Compendio de los mam´
ıferos terrestres aut ´
octonos de Cuba
vivientes y extinguidos.MuseoNacionaldeHistoriaNatural,
La Habana, Cuba.
late Quaternary sloths on continents and islands. Pro-
ceedings of the National Academy of Sciences USA 102,
LATTA,S.C.&LOV ET TE, I.J. 2011. Ancient islands
and modern invasions: disparate phylogeographic histories
among Hispaniola’s endemic birds. Molecular Ecology 20,
TATE , G.H.H. 1948. Notes on the Hispaniolan hutia, Plagiodon-
tia, and extinct related genera. Journal of Mammalogy 29,
I.J. 2007. Ancient differentiation in the single-island
avian radia tion of endemic Hispaniolan chat-tanagers
(Aves: Calyptophilus). Molecular Ecology 16, 3634–
Una campa˜
na en la Rep´
ublica Dominicana. Resultados pre-
liminares. Endins 21,6374.
TURVEY,S.T.2009.Holocene Extinctions.OxfordUniversity
Press, Oxford.
mals past: biological and geographical patterns of global
mammalian extinction across the Holocene. Philosophi-
cal Transactions of the Royal Society of London B 366,
ued survival of Hispaniolan solenodon Solenodon paradoxus
in Haiti. Oryx 42, 611–614.
useful is local ecological knowledge as a conservation tool
for small mammals in a Caribbean multicultural landscape?
Conservation Biology.
WILKINS, L. 2001. Impact of hunting on Jamaican hutia (Geo-
capromys brownii) populations: evidence from zooarchaeol-
ogy and hunter surveys. In: WOODS,C.A.&SERGILE, F.E.,
Eds., Biogeography of the West Indies: Patterns and Perspec-
WILLIAMS, E.E. 1961. Notes on Hispaniolan herpetology. 3. The
evolution and relationships of the Anolis semilineatus group.
Museum of Comparative Zoology, Breviora 136,18.
WILSON,D.E.&REEDER, D.M. 2005. Mammal Species of the
World: A Taxonomic and Geographic Reference, 3rd edition.
The Johns Hopkins University Press, Baltimore, MD.
WOODS, C.A. 1989. A new capromyid rodent from Haiti: the ori-
gin, evolution, and extinction of West Indian rodents, and their
bearing on the origin of New World hystricognaths. Natural
History Museum of Los Angeles County, Science Series 33,
WOODS,C.A.&HOWLA ND,E.B.1979.Adaptiveradiationof
capromyid rodents: anatomy of the masticatory apparatus.
Journal of Mammalogy 60,95116.
WOODS,C.A.&OTTENWALDER,J.A.1992.The Natural His-
tory of Southern Haiti. Florida Museum of Natural History,
Gainesville, FL.
mammals of the Greater Antilles: the summarised findings of
a ten weeks field survey in the Dominican Republic, Haiti and
Puerto Rico. Dodo 22,2342.
patterns and radiations of West Indian rodents. In: WOODS,
C.A. & SERGILE, F.E., Eds., Biogeography of the West In-
dies: Patterns and Perspectives.CRCPress,BocaRaton,FL,
pp. 335–353.
Associate Editor: Christoph Bleidorn
Downloaded by [The Zoological Society of London], [Samuel T. Turvey] at 05:53 19 December 2012
... The distribution and relative abundance of some taxa included in this study have been examined before, especially in a palaeogeographical framework (Rimoli, 1976;Woods, 1989;Brace et al., 2012;Hansford et al., 2012;Cooke and Crowley, 2018). However, most studies only mention the presence of certain taxa in the localities of interest or compare species lists between different regions. ...
... mm) and H. phenax (26.61e22.45 mm) considerably expand the range hitherto known and show a small overlap among the two species (Rimoli, 1976;Hansford et al., 2012;Cooke and Crowley, 2018). Furthermore, mean values of TRL and body mass available suggest that P. ipnaeum was relatively larger than P. araeum, and not smaller as Rimoli (1976) suggested (Hansford et al., 2012). ...
... mm) considerably expand the range hitherto known and show a small overlap among the two species (Rimoli, 1976;Hansford et al., 2012;Cooke and Crowley, 2018). Furthermore, mean values of TRL and body mass available suggest that P. ipnaeum was relatively larger than P. araeum, and not smaller as Rimoli (1976) suggested (Hansford et al., 2012). Measurements derived from a nearly complete mandible of Q. gravis (MNHNSD FOS 25.3421) support the suggestion that this species was the largest rodent on the island reaching~10 kg and overlapping in body mass with some of the small species of sloths classified in the genus Neocnus (Turvey and Fritz, 2011). ...
Hispaniola is the second-largest island in the West Indies bioregion and along with Cuba is considered the cradle of Greater Antillean rodent evolution and diversification. While the fossil record in Hispaniola includes ten extinct species of late Quaternary rodents, only a single species, the Hispaniola hutia (Plagiodontia aedium), is extant on the island. Changes in climate and anthropogenic effects have been suggested as drivers for these extinctions, but there are few radiometric dates associated with these fossils which limits our understanding of the timing of rodent extinctions on Hispaniola relative to shifts in Holocene climate and major events in human settlement and colonization. We report nine new AMS radiocarbon dates for six endemic rodents of Hispaniola that, when coupled with previously reported dates, provide last occurrence records for eight of the ten extinct rodent species there. Results show that Hispaniola rodent extinction occurred in a series of episodes, with more than half (63.6%) surviving into the late Holocene (4200 cal BP to present), postdating the initial arrival of Ceramic Age Indigenous groups, with extinction likely only after European colonization. This pattern in Hispaniola is consistent with what recent studies have found on the timing of extinction of other terrestrial vertebrates in the West Indies. Results from analysis of stable isotopes of fossil bone suggest that the extinction of endemic rodents from Hispaniola largely also postdate habitat changes following the onset of drier conditions in the late Holocene, suggesting that while climate may have shaped the geographic distribution of species, it probably played a minimal direct role in their extinctions on Hispaniola. Published by Elsevier Ltd.
... Of approximately 43 rodent species that have been described, only 10-13 are extant. Both climate (Dávalos and Russell, 2012) and anthropogenic factors, such as hunting and habitat destruction (Cooke et al., 2017a(Cooke et al., , 2017bHansford et al., 2012;Morgan and Woods, 1986) have been proposed as contributing to these extinctions. Last appearance dates across the Caribbean suggest that extinctions occurred in two waves, concordant with (1) initial human settlement ca. ...
... Here, we provide eight 14 C direct dates for rodents from Trouing Jérémie #5, a paleontologically-rich sinkhole in southwestern Haiti (Figure 1). This includes new dates for the two previously dated endemic species, B. voratus and I. portoricensis, three endemics that have not been dated directly previously, Plagiodontia velozi (here, considered synonymous with P. ipnaeum; Hansford et al., 2012), Hexolobodon phenax, and Rhizoplagiodontia lemkei, and invasive Rattus. ...
... Hispaniola's endemic rodents include the extant species, P. aedium, and at least nine extinct species: P. velozi, Rhizoplagiodontia lemkei, Hyperplagiodontia araeum, Isolobodon montanus, I. portoricensis, Hexolobodon phenax, Brotomys voratus, B. contractus, and Quemisia gravis (Table 1) (Hansford et al., 2012;Woods and Kilpatrick, 2005). The phylogenetic relationships of Caribbean rodents is currently under revision. ...
We present eight new radiocarbon dates for endemic and invasive rodents from Trouing Jérémie #5, a paleontologically-rich sink hole on the Tiburon Peninsula, Haiti. This includes new dates for two species that have been previously directly dated ( Isolobodon portoricensis and Brotomys voratus) as well as three endemic rodents which have no previous direct radiometric dates ( Plagiodontia velozi, Hexolobodon phenax, and Rhizoplagiodontia lemkei). The radiocarbon date for P. velozi, the largest of these species, is from the very early Holocene (10,995 ± 190 calendar years before present; Cal BP). Specimens of medium-bodied species, H. phenax and R. lemkei date to the mid-Holocene, while specimens of smaller-bodied I. portoricencis and B. voratus have dates falling in the Late-Holocene. These dates confirm that several of the extinct rodent species coexisted with the first humans, who arrived on the island ca. 6000 years ago. In contrast, murid Rattus specimens date to the last few centuries. Rats arrived with Europeans in the late 15th or early 16th Century and a radiocarbon date of ca. 500 Cal BP for one individual suggests that they likely spread quickly across the island. Collectively, these dates establish that vertebrate accumulations at Trouing Jérémie #5 span the Holocene. Remains from this site may provide a useful time sequence for future work examining ecological change across the Holocene as well as regional extirpation patterns.
... Although Caribbean mammal body masses spanned several orders of magnitude, all surviving non-volant species fall within a range of ca 0.5-3.0 kg; this pattern has prompted the 'Goldilocks Hypothesis', which suggests that intermediate-sized species were large enough to be resilient to invasive mammals yet small enough to be resilient to human offtake, and so their size was 'just right' [26]. However, fauna-wide patterns of vulnerability and survival in relation to biological parameters have not been investigated across Caribbean mammals within a rigorous statistical and phylogenetically explicit framework; it is possible that this pattern of survival is instead random with respect to body mass, as the region's late Quaternary fauna consisted of more intermediate-sized species to begin with [18]. ...
... Caribbean mammal extinction selectivity shows a significant and complex body mass effect, with both mass extremes negatively correlated with survival probability across all models. We therefore confirm the 'Goldilocks Hypothesis' proposed for the Caribbean non-volant mammal fauna [26]. Instead of survivorship representing a random subset of the pre-human fauna, or a probabilistic outcome of extinctions in a fauna containing more intermediate-sized species, we demonstrate that Caribbean medium-bodied rodents and solenodontid eulipotyphlans have been less sensitive to extinction compared to their smaller and larger non-volant counterparts. ...
... Available evidence for prehistoric hunting of larger Caribbean mammals (sloths, primates, giant rodents) is limited [30], but these species may have been particularly vulnerable to fire-driven habitat change, and even occasional harvesting could have been unsustainable for slowly reproducing populations [18]. By contrast, many smaller species might have only become vulnerable with the later introduction of invasive mammalian competitors and predators [21,26,77]. As these introductions occurred relatively recently, our model predicts their extinction risk through the square mass variable. ...
Full-text available
Preventing extinctions requires understanding macroecological patterns of vulnerability or persistence. However, correlates of risk can be nonlinear, within-species risk varies geographically, and current-day threats cannot reveal drivers of past losses. We investigated factors that regulated survival or extinction in Caribbean mammals, which have experienced the globally highest level of human-caused postglacial mammalian extinctions, and included all extinct and extant Holocene island populations of non-volant species (219 survivals or extinctions across 118 islands). Extinction selectivity shows a statistically detectable and complex body mass effect, with survival probability decreasing for both mass extremes, indicating that intermediate-sized species have been more resilient. A strong interaction between mass and age of first human arrival provides quantitative evidence of larger mammals going extinct on the earliest islands colonized, revealing an extinction filter caused by past human activities. Survival probability increases on islands with lower mean elevation (mostly small cays acting as offshore refugia) and decreases with more frequent hurricanes, highlighting the risk of extreme weather events and rising sea levels to surviving species on low-lying cays. These findings demonstrate the interplay between intrinsic biology, regional ecology and specific local threats, providing insights for understanding drivers of biodiversity loss across island systems and fragmented habitats worldwide.
... However, nearly all of the region's endemic insular mammal fauna became extinct following multiple waves of human colonisation from the mid-Holocene onwards, which led to the loss of over 100 mammal species or distinct island populations [12,14,16]. There is continuing uncertainty over the taxonomic status of many extinct Caribbean mammals, with many recently recognised species now considered dubious or invalid [17,18] and other unstudied populations potentially representing undescribed species [19,20], but molecular studies have been limited by poor DNA preservation under high thermal ages represented by the Caribbean's hot, humid tropical conditions [21][22][23]. ...
... However, these taxa instead exhibit low alpha diversity and high beta (between-landscape) diversity, with separate lineages distributed allopatrically rather than sympatrically across Hispaniola's geologically and environmentally heterogeneous landscapes. Allopatric lineages of solenodons and hutias are interpreted as representing separate subspecies rather than distinct species, because they exhibit limited (albeit statistically detectable) morphometric differentiation [18,63], in marked contrast to the substantial body size variance seen between Hispaniolan Nesophontes species that co-occur in the same landscapes. Similar patterns of allopatric differentiation and high island-wide beta diversity are also seen in several other Hispaniolan vertebrate taxa, including bats [64], birds [65] and reptiles [66]. ...
Full-text available
Background: The Caribbean offers a unique opportunity to study evolutionary dynamics in insular mammals. However, the recent extinction of most Caribbean non-volant mammals has obstructed evolutionary studies, and poor DNA preservation associated with tropical environments means that very few ancient DNA sequences are available for extinct vertebrates known from the region's Holocene subfossil record. The endemic Caribbean eulipotyphlan family Nesophontidae ("island-shrews") became extinct ~ 500 years ago, and the taxonomic validity of many Nesophontes species and their wider evolutionary dynamics remain unclear. Here we use both morphometric and palaeogenomic methods to clarify the status and evolutionary history of Nesophontes species from Hispaniola, the second-largest Caribbean island. Results: Principal component analysis of 65 Nesophontes mandibles from late Quaternary fossil sites across Hispaniola identified three non-overlapping morphometric clusters, providing statistical support for the existence of three size-differentiated Hispaniolan Nesophontes species. We were also able to extract and sequence ancient DNA from a ~ 750-year-old specimen of Nesophontes zamicrus, the smallest non-volant Caribbean mammal, including a whole-mitochondrial genome and partial nuclear genes. Nesophontes paramicrus (39-47 g) and N. zamicrus (~ 10 g) diverged recently during the Middle Pleistocene (mean estimated divergence = 0.699 Ma), comparable to the youngest species splits in Eulipotyphla and other mammal groups. Pairwise genetic distance values for N. paramicrus and N. zamicrus based on mitochondrial and nuclear genes are low, but fall within the range of comparative pairwise data for extant eulipotyphlan species-pairs. Conclusions: Our combined morphometric and palaeogenomic analyses provide evidence for multiple co-occurring species and rapid body size evolution in Hispaniolan Nesophontes, in contrast to patterns of genetic and morphometric differentiation seen in Hispaniola's extant non-volant land mammals. Different components of Hispaniola's mammal fauna have therefore exhibited drastically different rates of morphological evolution. Morphological evolution in Nesophontes is also rapid compared to patterns across the Eulipotyphla, and our study provides an important new example of rapid body size change in a small-bodied insular vertebrate lineage. The Caribbean was a hotspot for evolutionary diversification as well as preserving ancient biodiversity, and studying the surviving representatives of its mammal fauna is insufficient to reveal the evolutionary patterns and processes that generated regional diversity.
... European colonisation led to the mass extinctions of many endemic species (Crosby 2003;Turvey et al. 2007Turvey et al. , 2017 with Hispaniolan hutia (Plagiodontia aedium) and solenodon (Solenodon paradoxus) being the sole surviving endemic mammals (Mcfarlane et al. 2000;Whidden and Asher 2001). Post-AD 1500 extinctions of endemic mammals include all West Indies shrews, all but one species of hutia and all edible rats (Hansford et al. 2012;Mcfarlane et al. 2000;Macphee, Flemming, and Lunde 1999). ...
Full-text available
This study analyses zooarchaeological material recovered from the late precolumbian site of El Flaco (AD 990–1450), northern Dominican Republic. The faunal assemblage from this inland settlement demonstrates terrestrially focused modes of faunal exploitation but with some resources obtained from coastal ecosystems, such as mangrove forests, sandy-bottom, and reefs, which are located approximately 20 km to the northwest. This study establishes last occurrence dates for extinct taxa; examines the spatial distribution of fauna; explores modes of food procurement; and the effects of agricultural activities on local fauna by investigating animal remains from three excavation units. A diachronic study of animal remains from one artificial mound demonstrates changing patterns in resource exploitation, such as an increasing consumption of land crab over a roughly 100-year period. This study follows previous research that examined the isotope ecology of endemic species from El Flaco that indicates some hutias were possibly scavenging or being fed cultivated plants. Palaeoenvironmental data suggest that Indigenous landscape transformations led to the creation of mosaic environments, which may have attracted and supported certain species, implying that the inhabitants of El Flaco likely did not have to venture far to trap or hunt many of the animals upon which they relied.
... Cueva de Berna, a dry cave approximately 20 km east of Padre Nuestro (the type site for Antillothrix bernensis) has charcoal associated with faunal remains from which a radiocarbon date of 3850 ±150 yr BP was obtained (Rímoli 1977). The fauna present in Padre Nuestro and La Jeringa is similar to that found in these dated sites, though, the Haitian sites do contain a much greater diversity of rodents from the genus Plagiodontia (Hansford et al. 2012). ...
Full-text available
Parocnus dominicanus sp. nov. represents a new species of megalonychid ground sloth from the Altagracia Province of southeastern Dominican Republic. Specimens of multiple individuals, including one associated partial skeleton, were recovered from two separate underwater caves in the Parque Nacional del Este through collaborations with museums and cave divers between 2009–2013. Parocnus dominicanus sp. nov. is distinguished by its small size compared to that of P. serus, with percent differences in limb element lengths ranging from 13−24%. Numerous cranial and post-cranial elements also exhibit morphological character states that are not attributable to size variations. The recovery of multiple individuals within each locality demonstrates a size dimorphism, possibly sexual, which parallels patterns exhibited by P. serus. The two species are also geographically distinct, with no examples of co-occurrence at any localities to date. Parocnus dominicanus sp. nov. and P. serus share character states that are distinct from those of the Cuban species, P. browni, and which suggest differential usage of the forelimb. The exact age of the specimens described here is unknown, however, Parocnus has been dated to the Holocene in Haiti.
... G.T. Shev et al. practices that attracted and benefited this animal. It has been proposed that one of the reasons that I. portoricensis went extinct, whereas P, aedium did not, is due to the latter's more generalist behaviour and arboreal nature (Cooke and Crowley, 2018;Hansford et al., 2012). If I. portoricensis was a more ground-dwelling species, then the clearing of forest may have served to attract and support populations. ...
Full-text available
There are certain animal candidates for which researchers have long speculated as having been managed, or possibly domesticated, by indigenous peoples in the pre-Columbian Caribbean, the primary candidate being some members of a group of caviomorph rodents known as hutia (Capromyinae). This study comprises an isotopic and morphological investigation of the potential management of an extinct species endemic to Hispaniola, the Puerto Rican hutia (Isolobodon portoricensis). For comparisons, isotopic analysis was conducted of bone collagen samples examining carbon (∂¹³Cco) and nitrogen (∂¹⁵N) values of I. portoricensis with two other species of endemic hutia, guinea pig (Cavia porcellus), Antillean slider (Trachemys stejnegeri), rhinoceros iguana (Cyclura cornuta) and edible rat (Brotomys sp.) from four sites in the Dominican Republic: El Flaco, El Carril, El Cabo and La Entrada. This data was compared to human and dog collagen values available from El Flaco to assess similarities between these groups. Osteometric data was recorded for hutia mandibles, allowing for body mass estimations (n = 230). The findings suggest that some Isolobodon portoricensis specimens possessed carbon (δ¹³Cco) values similar to humans and dogs at El Flaco, possibly associated with the consumption of agriculturally produced maize. This research does not indicate whether domestication or management of this species was occurring, at least not in a systematic way. Concurrence with palaeoenvironmental and zooarchaeological data from the region in which El Flaco and El Carril are located indicate that indigenous agricultural practices may have affected populations of hutia, perhaps by attracting them and supporting them within anthropogenic mosaic landscapes. The data also suggests that some degree of either purposeful feeding or scavenging from human agricultural plots was occurring and supports some degree of commensalism between humans and Puerto Rican hutia at these sites.
... The insular Caribbean also experienced more mammal extinctions than any other global region during the Holocene, probably associated with hunting, landscape transformation, and invasive mammal introduction by Amerindian and subsequent European colonists from 6,000 years ago onward (MacPhee 2009;Turvey 2009;Cooke et al. 2017). From a late Quaternary fauna comprising 33 currently recognized endemic species in 19 genera (Silva Taboada et al. 2007;Turvey and Fritz 2011;Hansford et al. 2012), all Caribbean heteropsomyines and heptaxodontids are now extinct, and only 11 capromyid species in five genera probably survive, most of which are threatened with extinction . ...
Full-text available
Reconstructing the evolutionary history of island biotas is complicated by unusual morphological evolution in insular environments. However, past human-caused extinctions limit the use of molecular analyses to determine origins and affinities of enigmatic island taxa. The Caribbean formerly contained a morphologically diverse assemblage of caviomorph rodents (33 species in 19 genera), ranging from ∼0.1-200 kg and traditionally classified into three higher-order taxa (Capromyidae/Capromyinae, Heteropsomyinae, Heptaxodontidae). Few species survive today, and the evolutionary affinities of living and extinct Caribbean caviomorphs to each other and to mainland taxa are unclear: are they monophyletic, polyphyletic, or paraphyletic? We use ancient DNA techniques to present the first genetic data for extinct heteropsomyines and heptaxodontids, as well as for several extinct capromyids, and demonstrate through analysis of mitogenomic and nuclear datasets that all sampled Caribbean caviomorphs represent a well-supported monophyletic group. The remarkable morphological and ecological variation observed across living and extinct caviomorphs from Cuba, Hispaniola, Jamaica, Puerto Rico, and other islands was generated through within-archipelago evolutionary radiation following a single Early Miocene overwater colonisation. This evolutionary pattern contrasts with the origination of diversity in many other Caribbean groups. All living and extinct Caribbean caviomorphs comprise a single biologically remarkable subfamily (Capromyinae) within the morphologically conservative living Neotropical family Echimyidae. Caribbean caviomorphs represent an important new example of insular mammalian adaptive radiation, where taxa retaining "ancestral-type" characteristics coexisted alongside taxa occupying novel island niches. Diversification was associated with the greatest insular body mass increase recorded in rodents, and possibly the greatest for any mammal lineage.
... G.T. Shev et al. practices that attracted and benefited this animal. It has been proposed that one of the reasons that I. portoricensis went extinct, whereas P, aedium did not, is due to the latter's more generalist behaviour and arboreal nature (Cooke and Crowley, 2018;Hansford et al., 2012). If I. portoricensis was a more ground-dwelling species, then the clearing of forest may have served to attract and support populations. ...
Full-text available
In precolumbian insular Caribbean archaeological sites, domestic dog ( Canis familiaris ) remains have been recovered from varied contexts, such as formal burials, in refuse deposits, and as modified artifacts, indicating their complex and multifaceted role within indigenous societies. In this study, isotopic and morphometric analyses provide biochemical and morphological correlations to assess this differential treatment. We examined collagen values ( n = 21) of carbon (δ ¹³ C co ) and nitrogen (δ ¹⁵ N), and enamel values ( n = 81) of carbon (δ ¹³ C en ), oxygen (δ ¹⁸ O en ), and strontium ( ⁸⁷ Sr/ ⁸⁶ Sr) of dog remains from 16 precolumbian sites. Five comparative parameters were used to assess dietary variations between different groups: buried versus nonburied, local versus nonlocal, Greater versus Lesser Antilles, chronology, and modified versus unmodified remains. The only statistically significant difference in diets was between local and nonlocal dogs. Sufficient data were available to conduct isotopic mixing models using the FRUITS statistical program on four individuals for which depositional and morphological data were available. Results of dietary modeling indicate an unexpectedly heavy reliance on plant foods consistent with intentional feeding. This approach highlights the utility of combining isotope analysis, dietary models, morphometrics, and depositional context to provide comprehensive biographic overviews of individual animals.
Full-text available
To understand the current biodiversity crisis, it is crucial to determine how humans have affected biodiversity in the past. However, the extent of human involvement in species extinctions from the Late Pleistocene onward remains contentious. Here, we apply Bayesian models to the fossil record to estimate how mammalian extinction rates have changed over the past 126,000 years, inferring specific times of rate increases. We specifically test the hypothesis of human-caused extinctions by using posterior predictive methods. We find that human population size is able to predict past extinctions with 96% accuracy. Predictors based on past climate, in contrast, perform no better than expected by chance, suggesting that climate had a negligible impact on global mammal extinctions. Based on current trends, we predict for the near future a rate escalation of unprecedented magnitude. Our results provide a comprehensive assessment of the human impact on past and predicted future extinctions of mammals.
"Near time" -an interval that spans the last 100,000 years or so of earth history-qualifies as a remarkable period for many reasons. From an anthropocentric point of view, the out­ standing feature of near time is the fact that the evolution, cultural diversification, and glob­ al spread of Homo sapiens have all occurred within it. From a wider biological perspective, however, the hallmark of near time is better conceived of as being one of enduring, repeat­ ed loss. The point is important. Despite the sense of uniqueness implicit in phrases like "the biodiversity crisis," meant to convey the notion that the present bout of extinctions is by far the worst endured in recent times, substantial losses have occurred throughout near time. In the majority of cases, these losses occurred when, and only when, people began to ex­ pand across areas that had never before experienced their presence. Although the explana­ tion for these correlations in time and space may seem obvious, it is one thing to rhetori­ cally observe that there is a connection between humans and recent extinctions, and quite another to demonstrate it scientifically. How should this be done? Traditionally, the study of past extinctions has fallen largely to researchers steeped in such disciplines as paleontology, systematics, and paleoecology. The evaluation of future losses, by contrast, has lain almost exclusively within the domain of conservation biolo­ gists. Now, more than ever, there is opportunity for overlap and sharing of information.
The holotype (a fossil mandibular ramus) and only available specimen of Capromys (Geocapromys) megas Varona and Arredondo, 1979, is compared to the other two Cuban species of Geocapromys (also known only from fossil remains), as well as to all four species (one extant) of Capromys s.s. The oclusal design of the molar series in G. megas is described and illustrated for the first time. It is concluded that G. megas is a synonym of Capromys pilorides.
The island of Hispaniola is one of the largest land masses straddling the North American-Caribbean plate boundary and is a critical area for testing ideas about the development of the plate boundary. The aims of this volume include establishing a systematic geologic database and stratigraphic nomenclature for the island; testing recent models for the tectonic evolution of Hispaniola; and studying a variety of plate boundary zone processes including evolution of island arcs, the transition between arc and strike-slip tectonics, and terrane accretion. The 18 chapters (abstracted separately) are organized into two sections: the first section describes and interprets the composition and structure of early Cretaceous to middle Eocene rocks of the Hispaniola volcanic arc, the second section describes and interprets late Eocene to Recent sedimentary rocks which postdate arc activity and were formed in a collisional or strike-slip environment. A set of regional geologic maps of the Dominican Republic are included. -A.W.Hall