Molecular phylogeny of enigmatic Caribbean
spider crabs from the Mithrax?Mithraculus
species complex (Brachyura: Majidae:
Mithracinae): ecological diversity and a
formal test of genera monophyly
j. antonio baeza1,2, juan a. bolan~os3, soledad fuentes4, jesu? s e. hernandez3, carlos lira3
and re?gulo lo? pez3
1Smithsonian Tropical Research Institute, Apartado Postal 0843-03092, Balboa, Anco?n, Republic of Panama, 2Smithsonian Marine
Station at Fort Pierce, 701 Seaway Drive, Fort Pierce, FL 34949, USA, 3Grupo de Investigacio?n en Carcinolog??a, Escuela de Ciencias
Aplicadas del Mar, Nu?cleo Nueva Esparta, Universidad de Oriente, Isla Margarita, Venezuela, 4NOAA Fisheries, Milford Laboratory,
212 Rogers Avenue, Milford CT 06460, USA
Crabs from the Mithrax?Mithraculus species complex are known for their diversity of lifestyles, habitats, and coloration. This
group includes small, colourful, symbiotic species and much larger, reef-dwelling crabs targeted by ?shermen. The evolution-
ary relationships between the species within this complex are not well-de?ned. Previous studies based upon morphological
characters have proposed the separation of this complex into two genera (Mithrax and Mithraculus), but cladistic analyses
based upon larval characters do not support this division. A molecular phylogeny of the group may help to resolve this long-
standing taxonomic question and shed light on the ecological conditions driving the diversity of these crabs. Using a 550-bp
alignment of the 16S rRNA mitochondrial DNA segment we examined the phylogenetic relationships between 8 species within
the Mithrax?Mithraculus complex native to the Caribbean. The resulting phylogeny indicates that this complex is paraphy-
letic, as it includes the genus Microphrys. The analyses revealed a well-supported, monophyletic group containing four species
of Mithraculus (M. cinctimanus, M. coryphe, M. sculptus and M. forceps) and supported one pair of sister species from the
genus Mithrax (M. caribbaeus and M. spinosissimus). No complete segregation of species, according to genera, was evident,
however, from tree topologies. Bayesian-factor analyses revealed strong support for the unconstrained tree instead of alterna-
tive trees in which monophyly of the two genera was forced. Thus, the present molecular phylogeny does not support the sep-
aration of the species within this complex into the genera Mithrax and Mithraculus. A review of the literature demonstrated
considerable phenotypic variation within monophyletic clades in this group.
Keywords: Mithrax, Mithraculus, Mithracinae, emerald crab, spiny crab, spider crab, Venezuela
Submitted 19 May 2009; accepted 2 August 2009; ?rst published online 23 November 2009
I NTRODUCT ION
Among the Decapoda, crabs from the infraorder Brachyura
are recognized for their astonishing anatomical, ecological,
and behavioural diversity (Martin & Davis, 2001; Ng et al.,
2008). During the last decade, various phylogenetic studies
have supported monophyly of the Brachyura and have
begun to reveal internal relationships and clarify the position
of this clade relative to other major, decapod lineages (Scholtz
& Richter, 1995; Sturmbauer et al., 1996; Ahyong & O?Meally,
2004; Porter et al., 2005; Schubart et al., 2006). Recent studies
also have uncovered an evolutionary history much more
complex than originally recognized (Ahyong & O?Meally,
2004; Porter et al., 2005; Schubart et al., 2006; Ahyong et al.,
2007; Hultgren & Stachowicz, 2008). Furthermore, some sys-
tematic studies, combined with behavioural and ecological
observations, have exposed the evolutionary basis for most
peculiar behaviours and the conditions favouring them
(land colonization and subsequent evolution of cooperative
behaviour in bromeliad dwelling crabs?Schubart et al.,
1998; independent evolutionary origins of tree climbing beha-
viours?Fratini et al., 2005; increasing complexity of courtship
through time?Sturmbauer et al., 1996; origins and evolution-
ary trade-offs associated with camou?age?Hultgren &
Stachowicz, 2009). Albeit, our knowledge of the evolutionary
history of true crabs has increased substantially, the internal
relationships between many genera and families still remain
unknown. This remaining ignorance is mostly a consequence
of high morphological diversity, convergence of adult features
at high taxonomic levels, dif?culties sampling extreme
environments colonized by some groups (e.g. hydrothermal
vents), and the burden of taxonomic expertise and costs of
molecular techniques (see Ahyong et al., 2007).
Corresponding author:
J.A. Baeza
Email: baezaa@si.edu
851
Journal of the Marine Biological Association of the United Kingdom, 2010, 90(4), 851?858. # Marine Biological Association of the United Kingdom, 2009
doi:10.1017/S0025315409991044
Among spider crabs of the superfamily Majoidea (sensu Ng
et al., 2008), one of the most species-rich clades of brachyur-
ans, the Mithrax?Mithraculus species complex is of particular
interest. Crabs from these two genera demonstrate a consider-
able diversity of lifestyles, body sizes, habitats, and coloration.
The 25 recognized species (Ng et al., 2008; but see Boschi,
2000) are restricted to the Americas and inhabit shallow or
relatively deep, warm-temperate, subtropical, and tropical,
rocky and coral reefs. Some species live in groups (small
aggregations); whereas, others remain solitarily within shelters
(M. spinosissimus?Rathbun, 1925; personal observations).
Several species with cryptic coloration dwell under rocks or
in crevices, but other, more-colourful species inhabit sea ane-
mones in shallow reefs (M. commensalis?Manning, 1970).
These colourful species, and other less-conspicuous, congene-
ric crabs, are capable of controlling aquarium pests, and thus,
are heavily traded in the marine aquarium industry (Rhyne
et al., 2004). Lastly, most species are small (e.g. M. forceps,
M. sculptus, M. cinctimanus, M. coryphe and M. ruber) but
some grow up to 20 cm carapace length (e.g. M. spinosissi-
mus?Rathbun, 1925). These large species are of economic
importance as targets of local ?sheries in developing countries.
The ecological diversity of crabs from the genera Mithrax and
Mithraculus has already attracted the attention of systematists,
?sheries biologists and aquaculturists. The same diversity
suggests that these crabs are ideal model systems to explore
the role of environmental conditions in explaining evolution-
ary innovations in the marine environment. Phylogenetic
studies in this Mithrax?Mithraculus species complex are war-
ranted because of the implications for conservation biology
and biodiversity.
The genus Mithrax (Desmarest, 1823) was originally desig-
nated to contain various species of majids (currently compris-
ing the subfamily Mithracinae McLeay 1838) characterized,
among other traits, by an ovate or oblong carapace (either
broader than long or slightly longer than broad) with antero-
lateral margins having four or three spines or lobes behind the
orbital front, two small horns at the rostrum, and robust legs
(Rathbun, 1925). Later, White (1847) proposed the separation
of this genus in two subgenera, Mithrax and Mithraculus.
Mithraculus contains species, including the type M. sculptus,
which differ from members of Mithrax by having a smooth
carapace with oblique branchial sulci, very short rostral
horns, and minor teeth at the orbits that are inconspicuous
and tubercle-like (Rathbun, 1925). All species but one
(M. cinctimanus) in this subgenus are characterized by
having a carapace broader than long. White?s suggestion
was followed by scientists throughout the 20th Century,
including Rathbun (1925), and these two subgenera were elev-
ated to the genus level by Wagner (1990), a view supported by
Ng et al. (2008). Most recently, detailed morphological exam-
ination of larval characters (e.g. setae distribution) revealed
only minor differences among species from Mithrax (e.g.
M. caribbaeus, M. hispidus and M. pleuracanthus) and
Mithraculus (e.g. M. sculptus, M. coryphe, and M. forceps)
(Provenzano & Brownell, 1977; Scotto & Gore, 1980;
Bolan?os & Scelzo, 1981; Goy et al., 1981; Fransozo &
Hebling, 1982; Bolan?os et al., 1990, 2000; Santana et al.,
2003; Rhyne et al., 2004). The strong similarities in larval
characters do not support the separation of this species
complex into two genera, as proposed by studies based upon
adult morphology (Wagner, 1990). Furthermore, additional
cladistic analyses with larval characters have been unable to
support the monophyly of the Mithracinae and its position
among other majoid lineages (Pohle & Marques, 2000).
Comprehensive taxonomic studies, including adult mor-
phology, larval characters, and molecular markers are
needed to resolve outstanding systematic problems in the
Mithrax?Mithraculus species complex and the Mithracinae
(Hultgren & Stachowicz, 2008).
This study represents the ?rst contribution to the phylogeny
of crabs from the genera Mithrax and Mithraculus that uses
molecular characters. We have focused speci?cally on addres-
sing the hypothesis of monophyly of these two genera. We pre-
dicted that, a molecular phylogeny of the species included
within the two genera should segregate the species into two
well-supported monophyletic clades. We present a molecular
phylogeny of this species complex based upon the large-
subunit, 16S mitochondrial rRNA gene applied to eight avail-
able species from the two genera plus outgroups.
MATER IALS AND METHODS
A total of 18 specimens from eight species of crabs from
the genera Mithrax and Mithraculus were included in the
present study (Table 1). Two specimens of Microphrys bicor-
nutus, one specimen of Pitho lherminieri (Mithracinae) and
another of Herbstia parvifrons (Pisinae) were also included
as outgroups during the phylogenetic analyses. Most crab
species were collected between 2005 and 2008 from different
localities at Isla Margarita, Venezuela, and Carrie Bow Caye,
Belize. Immediately after collection, specimens were preserved
in 95?99% ethanol. In the laboratory, the different species
were identi?ed using descriptions of Rathbun (1925),
Rodr??guez (1980), Williams (1984), Abele & Kim (1986),
Wagner (1990), Melo (1996) and Bolan?os et al. (2000).
Total genomic DNA was extracted from leg-muscle tissue
using the QIAGENw DNeasyw Blood and Tissue Kit follow-
ing the manufacturer?s protocol. The polymerase chain reac-
tion (PCR) was used to amplify an approximately 550
base-pair (bp) region (excluding primers) of the 16S rRNA,
using the primers 16ar (50?CGCCTGTTTAACAAAAAC
AT?30) and 16br (50?CCGGTTTGAACTCAGATCACGT?
30) (Palumbi et al., 1991), or 16L2 (50?TGCCTGTTTA
TCAAAAACAT?30) and 161472 (50?AGATAGAAACCAA
CCTGG?30) (Schubart et al., 2000; Baeza et al., 2009).
Standard PCR 25-ml reactions (2.5 ml of 10  Taq buffer,
2 ml of 50 mM MgCl2, 2.5 ml of 10 mM dNTPs, 2.5 ml each
of the two primers (10 mM), 0.625 U Taq, 1.25 ml of
20 mM BSI and 8.625 ml double distilled water) were per-
formed on a Peltier Thermal Cycler (DYADw) under the fol-
lowing conditions: initial denaturation at 968C for 4 minutes
followed by 40 cycles of 948C for 45 seconds, 52?578C
(depending on the species) for 1 minute, and 728C for 1
minute, followed by chain extension at 728C for 10 minutes.
PCR products were puri?ed with ExoSapIT (a mixture of exo-
nuclease and shrimp alkali phosphatase, Amersham
Pharmacia) and then sent for sequencing with the ABI Big
Dye Terminator Mix (Applied Biosystems) to the
Laboratory of Analytical Biology of the National Museum of
Natural History (LAB?NMNH, Maryland), which is
equipped with an ABI Prism 3730xl Genetic Analyser
(Applied Biosystems automated sequencer). All sequences
were con?rmed by sequencing both strands, and a consensus
sequence for the two strands was obtained using the software
852 j. antonio baeza et al.
Sequencher 4.5 (Gene Codes Corp). The sequences for
another six specimens from four species were retrieved from
Genbank (Hultgren & Stachowicz, 2008; Table 1).
The ?nal set of consensus sequences was aligned with the
integrated ClustalW, corrected manually with MEGA4
(Tamura et al., 2007), and then exported to PAUP
(Swofford, 2002) and MrBayes 3.1.2 (Huelsenbeck, 2000;
Ronquist & Huelsenbeck, 2003). This dataset was analysed
with the software Modeltest v. 3.7 (Posada & Crandall,
1998) in PAUP, which compares different models of DNA
substitution in a hierarchical, hypothesis-testing framework
to select a base-substitution model that best ?t the data. The
optimal model found by Modeltest (selected by both the
hierarchical likelihood ratio test and the Akaike information
criteria) was a TVM? G evolutionary model (? lnL ?
22207.9871). The calculated parameters were as follows:
assumed nucleotide frequencies A ? 0.3614, G ? 0.1795,
T ? 0.3350, C ? 0.1242; substitution rate matrix with A?C
substitution ? 0.6795, A?G ? 6.7878, A?T ? 2.5647,
C?G ? 0.2506, C?T ? 6.7878, G?T ? 1.0; rates for variable
sites assumed to follow a gamma distribution (G) with shape
parameter ? 0.2056.
Phylogenetic analyses conducted herein were maximum
parsimony (MP) and maximum likelihood (ML, in PAUP)
and Bayesian inference (BI, in MrBayes). We applied the
TVM? G model of substitution to our ML and Bayesian ana-
lyses. Maximum parsimony analysis was performed as a heur-
istic search, with a starting tree obtained by stepwise addition,
random addition of sequences, random replicates, and TBR
(tree?bisection?reconnection) branch swapping. For ML,
the speci?cations were the same as in MP. Branch swapping,
however, was performed in the starting tree and all other par-
ameters used were those of the default option in PAUP.
For BI, we used unique, random-starting trees in the
Metropolis?coupled Markov Monte Carlo Chain (MCMC)
(Huelsenbeck, 2000). The analysis was performed for
6,000,000 generations. Every 100th tree was sampled from
the MCMC analysis, resulting in a total of 60,000 trees. We
determined a burn-in period of less than 10,000 generations
after examining a plot of the log likelihood values against
the number of generations. Next, we calculated a consensus
tree with the 50% majority rule for the last 59,900 sampled
trees. We assessed the robustness of the MP and ML tree
topologies by bootstrap reiterations of the observed data
2000 and 1000 times, respectively, and reconstructing trees
using each resampled data set. Support for nodes in the BI
tree topology was obtained by posterior probability values
that represent the frequency that each clade occurred within
the collection of trees provided by the analysis.
We tested if the different species of the genera Mithrax and
Mithraculus segregated and formed different genus-speci?c
monophyletic clades. For this purpose, constrained trees (in
which the monophyly of a particular genus was enforced)
were obtained in MrBayes with the command constraint.
We ran MCMC searches and obtained the harmonic mean
of tree-likelihood values by sampling the post burn-in, pos-
terior distribution as above. Next, Bayes factors were used to
evaluate whether or not there was evidence against mono-
phyly (constrained versus unconstrained trees) according to
the criteria of Kass & Raftery (1995). Bayes factors compare
the total harmonic mean of the marginal likelihood of uncon-
strained versus monophyly-constrained models. The higher
the value of the Bayes factor statistic implies stronger
support against the monophyly of a particular group (Kass
& Raftery, 1995). Speci?cally, a value for the test statistic
2 loge(B10) between 0 and 2 indicates no evidence against
H0; values from 2 to 6 indicate positive evidence against H0;
values from 6 to 10 indicate strong evidence against H0; and
values .10 indicate very strong evidence against H0 (Kass
& Raftery, 1995; Nylander et al., 2004).
Table 1. Mithrax?Mithraculus species and other majoid crabs used for the phylogeny reconstruction. The sites of collection, museum catalogue number
(CN; MOBR: Museo Oceanolo?gico Hermano Benigno Roman, Estacio?n de Investigaciones Marinas de Margarita, Fundacio?n La Salle de Ciencias
Naturales, Venezuela; UMML, University of Miami Marine Laboratories, Rosenstiel School of Marine Science, University of Miami) and the
Genbank accession numbers (GenBank) are shown in bold for each species. NA, not available.
Species Collection site CN/GenBank
Mithraculus cinctimanus Carrie Bow Key, Belize GQ438762/UMML32.9623
Mithraculus coryphe 01 Isla Margarita, Venezuela GQ438771/MOBR-C-1535
Mithraculus coryphe 02 Isla Margarita, Venezuela GQ438778/MOBR-C-1535
Mithraculus forceps Bocas del Toro, Panama EU682782/NA
Mithraculus forceps 01 Isla Margarita, Venezuela GQ438761/MOBR-C-1533
Mithraculus forceps 02 Isla Margarita, Venezuela GQ438764/MOBR-C-1537
Mithraculus sculptus Florida, USA EU682783/NA
Mithraculus sculptus Bocas del Toro, Panama EU682784/NA
Mithraculus sculptus Bocas del Toro, Panama EU682785/NA
Mithraculus ruber 01 Isla Margarita, Venezuela GQ438769/MOBR-C-1531
Mithraculus ruber 02 Isla Margarita, Venezuela GQ438770/MOBR-C-1532
Mithrax caribbaeus Large 01 Isla Margarita, Venezuela GQ438773/MOBR-C-1528
Mithrax caribbaeus Large 02 Isla Margarita, Venezuela GQ438794/MOBR-C-1528
Mithrax caribbaeus Small 01 Isla Margarita, Venezuela GQ438765/MOBR-C-1529
Mithrax caribbaeus Small 02 Isla Margarita, Venezuela GQ438766/MOBR-C-1529
Mithrax verrucosus Isla Margarita, Venezuela GQ438767/MOBR-C-1534
Mithrax verrucosus Isla Margarita, Venezuela GQ438768/MOBR-C-1536
Mithrax spinosissimus Carrie Bow Key, Belize GQ438763/UMML
Herbstia parvifrons Catalina Island, CA, USA EU682792/NA
Microphrys bicornutus Isla Margarita, Venezuela GQ438760/MOBR-C-1530
Microphrys bicornutus Bocas del Toro, Panama EU682781/NA
Pitho lherminieri Bocas del Toro, Panama EU682789/NA
molecular phylogeny of mithrax mithraculus crabs 853
RESULTS
A total of 550 homologous alignment positions were used
during the present phylogenetic analysis; 140 of these were
found to be parsimony-informative positions. All phyloge-
netic trees obtained with the different inference methods
(BI, ML and MP), resulted in the same general topology
(Figure 1). Considering our limited number of outgroup
species, belonging to three different genera within the
Majoidea, the Mithrax?Mithraculus species complex is not
a monophyletic group (Figure 1).
The overall topology of the trees indicated one well-
supported, monophyletic clade that includes four of ?ve
species of Mithraculus (M. cinctimanus, M. coryphe,
M. forceps and M. sculptus). The basal position of
M. coryphe within this clade was well supported. The position
of the fourth species of Mithraculus (M. ruber) was not
resolved in the trees. The tree topology also indicated a pair
of sister species (well supported) that included two of the
three species of Mithrax (M. spinosissimus and M. caribbaeus).
Reciprocal monophyly, that would have indicated putatively-
cryptic species, was not observed between large and small
specimens of M. caribbaeus collected from different localities
in Venezuela. The position of M. spinosissimus and M. carib-
baeus among the other representatives of the complex was not
resolved. Similarly, the position of the third species of Mithrax
(M. verrucosus) that did not group together with the other two
congeners was not resolved.
The tree topologies demonstrated that species within this
complex did not segregate according to genus and formed
well-supported, monophyletic clades, as should be expected
according to adult morphology. Similarly, Bayes factor
analyses revealed no support for the separation of species
within the complex into two different genera (Mithrax and
Mithraculus). Comparisons of the unconstrained tree (harmo-
nic mean ? ?3258.48) versus the trees wherein Mithrax or
Mithraculus were imposed as monophyletic clades, indicated
strong support for the unconstrained tree (Mithrax: harmonic
mean ? ?3216.16, 2ln(B10) ? 7.49; Mithraculus: harmonic
mean ? ?3117.80, 2ln(B10) ? 9.89).
The literature review of the 8 species included in our ana-
lyses revealed that the coloration, lifestyle (habit), habitat,
sizes, and bathymetric distributions of these species are rela-
tively well known. Caribbean crabs from the Mithrax?
Mithraculus species complex can be found among rocks or
fossilized-coral terraces, live corals, under rocks or stones in
muddy and/shell bottoms, and associated with seaweeds and
sea anemones in the intertidal or subtidal (up to 60 m
depth). Most species are colourful, having plane greenish,
brownish or reddish coloration. Only one species
(Mithraculus cinctimanus) has a striking colour pattern
(white yellowish with reddish spots in the carapace). This
crab is one of two symbiotic species in the Caribbean
(if M. commensalis is not considered synonymous?see Ng
et al., 2008). Interestingly, M. sculptus and M. forceps appear
to form occasional facultative associations, respectively, with
Fig. 1. Phylogenetic tree obtained from BI analysis of the partial 16S rRNA gene for crabs from the Mithrax?Mithraculus species complex, and other selected taxa
from the family Majidae. Numbers above or below the branches represent the posterior probabilities from the BI analysis and bootstrap values obtained from ML
and MP analyses in PAUP (BI/ML/MP). The general topology of the trees obtained from MP and ML analyses was the same.
854 j. antonio baeza et al.
the coralline algae Neogoniolithon strictum and the coral
Oculina arbuscula (Stachowicz & Hay, 1996, 1999).
Unfortunately little is known about the costs and bene?ts
experienced by M. cinctimanus as consequences of association
with sea anemones. On the other hand, M. forceps obtains
food (lipid-rich mucus), shelter and protection against ?sh
predators from its coral host (Stachowicz & Hay, 1999).
Similarly, M. sculptus obtain protection against predators
from its algae host and, at the same time, this species
appears to clean its algal host of fouling seaweeds
(Stachowicz & Hay, 1996). Information regarding the
natural diet of most of the species is lacking, although all
species are considered to be herbivorous or omnivorous
(e.g. M. sculptus?Stachowicz & Hay, 1996) (Figure 2).
Information on the lifestyle (distribution) is available for
most species. Three species live solitarily (M. cinctimanus in
sea anemones, and M. verrucossus and M. caribbaeus on
holes or crevices), and the remaining species live in aggrega-
tions (Rathbun, 1925; personal observations). Interestingly,
M. spinosissimus have been seen living both solitarily and in
aggregations at different localities in the Caribbean (Carrie
Bow Caye, Belize and Los Roques, Venezuela, respectively).
This revision con?rms that the lifestyle of crabs from this
species complex is diverse and demonstrates that this high
diversity might be found within small, monophyletic clades
(see below).
D ISCUSS ION
This study presents for the ?rst time a molecular phylogeny of
crabs from the genera Mithrax and Mithraculus, based upon a
segment of the 16s rRNA mitochondrial gene. Considering our
limited pool of species (only eight out of the 25 valid species
within the complex?Ng et al., 2008), these analyses with
three different phylogenetic reconstruction methods recognized
only one monophyletic group consisting of four species of
Mithraculus (M. cinctimanus, M. coryphe, M. forceps and
M. sculptus) and also supported two members from the genus
Mithrax as sister species (M. spinosissimus and M. caribbaeus).
The positions of the remaining species from the complex
(Mithraculus ruber and Mithrax verrucosus) were not resolved.
In disagreement with the hypothesis based solely upon adult
morphology, a well-resolved grouping of all of the species
from a particular genus was not revealed by our analyses.
Also, Bayesian factors analyses strongly supported uncon-
strained trees over trees in which monophyly (either of
Mithrax or Mithraculus) was imposed. The present results do
not support the separation of these species into two different
genera as proposed by Wagner (1990) that was based upon
adult morphology alone. Instead, the present study agrees
with cladistic analyses based upon larval characters, indicating
that the Mithrax?Mithraculus divide within the complex is not
natural (Bolan?os & Scelzo, 1981; Bolan?os et al., 1990; Santana
Fig. 2. Ecology and lifestyle in crabs from the Mithrax?Mithraculus species complex. Colour: G, green; HV. highly variable, including reddish, reddish-brown,
green, bluish-green; LG, light green; R, reddish; R-Y, reddish-yellow; Y, yellow; WR, wine red. Habit: G, gregarious, S, solitary. Habitat: A, symbiotic with sea
anemones and sponges, C, among clumps of pearl oysters or other bivalves (Arca zebra), F, facultatively symbiotic with coralline algae or branching corals, H/
C, holes/crevices on reefs or brain corals, R, rubble, RR, rock and rubble, U, under stones; S, sand bottom with red algae. Size: bars represent the range in size
reported for the species. Depth: bars represent the range of depth from where specimens have been collected. Asterisk indicates node of well supported
monophyletic clades or sister species. Information is based on Williams (1984), Wagner (1990), Stachowicz & Hay (1996, 1999), Bolan?os et al. (2000) and
personal observations (J.A.Bs., J.E.H., R.L.).
molecular phylogeny of mithrax mithraculus crabs 855
et al., 2003). We argue in favour of future phylogenetic studies
using various types of evidence (molecular, adult morphology
and larval anatomy) to improve our knowledge of the natural
relationships within this species complex and its position in
the Mithracinae.
Mithrax verrucosus and Mithraculus rubber did not group
together with the other members of their respective genera.
Mithrax verrucosus differs from most of its congeners by
having a carapace paved with closely set granules (Wagner,
1990). Mithrax hemphlli Rathbun, 1892 and M. pilosus
Rathbun 1892 are the only other species from the genus that
also have the carapace covered with granules or tubercles
(Wagner, 1990). Given the basal position of M. verrucosus,
as indicated by our phylogenetic analyses, the presence
of granules/tubercles in the carapace might represent an
ancestral condition lost in some members from the studied
species complex (e.g. Mithrax spinosissimus and M. carib-
baeus). Future phylogenetic studies including M. hemphlli
and M. pilosus will help in determining whether or not the
presence/absence of tubercles/granules on the carapace is a
character with phylogenetic value in Mithrax?Mithraculus
crabs. Mithrax verrucosus features the entire set of characters
that de?ne the genus Mithrax (sensu Desmarest, 1823); ovate
or oblong carapace (either broader than long or slightly longer
than broad) with antero-lateral margins having four or three
spines or lobes behind the orbital front, two small horns at
the rostrum, and robust legs (Rathbun, 1925). That this
species did not group together with the other congeners also
suggests that carapace shape and the presence/absence of
lateral spines in the carapace might not be phylogenetically
informative characters within the studied species complex.
With regards to Mithraculus rubber, this species is similar
to M. forceps. Importantly, adults of Mithraculus rubber differ
from all of its congeners by lacking a large tooth on the pro-
podus of the cheliped (Wagner, 1990). Also, Mithraculus
rubber is the only species from the genus having the last tuber-
cle on the lateral margin of the carapace situated at a poster-
olateral angle and not at a position that separates the
posterolateral from the anterolateral carapace margins as
occurs in the remaining species from the genus (Wagner,
1990). The presence/absence and position of this tooth/tuber-
cle might be a character of phylogenetic relevance. New
studies combining molecular and morphological characters
will help in determining whether or not carapace shape, the
position of the lateral spines or tubercles on the carapace as
well as other morphological traits are phylogenetically infor-
mative characters within the studied species complex.
The results in the present study suggest that the Mithrax?
Mithraculus complex is paraphyletic; Microphrys bicornutus
clusters together with Mithraculus ruber and has a derived pos-
ition compared to Mithrax verrucosus (although not well sup-
ported). This observation agrees with previous phylogenetic
analyses based upon larval characters, revealing that the mono-
phyletic status of the Mithracinae is still uncertain, as well as its
position within the family Majidae (Marques et al., 2003).
The set of species considered in the present study does not
allow us to address other relevant systematic questions for the
group. For instance, an unresolved question is whether or not
the two sea anemone associates M. cinctimanus, recorded
from several localities in the Caribbean, and M. commensalis,
from Dominica (West Indies) are separate species or consti-
tute a single, pan-Caribbean species (Rathbun, 1925;
Manning, 1970; Bolan?os et al., 2000; Ng et al., 2008).
Similarly, Mithrax hispidus and Mithrax caribbaeus, once
considered different species, have been recently proposed as
synonyms based upon studies of larval and adult morphologi-
cal traits (Santana et al., 2003; Ng et al., 2008). It must be high-
lighted that, in contrast to other subfamilies of the Majidae
where larvae from different species are relatively easy to differ-
entiate (e.g. Pisinae?see Santana et al., 2004), larval mor-
phology is highly consistent within the Mithracinae
(especially in zoeal stages?Santana et al., 2003). Future phy-
logenies combining molecular, adult, and larval characters
might reveal an evolutionary history much more complex
than previously thought in this group (Hultgren &
Stachowicz, 2008).
The ecology of the Mithrax?Mithraculus species complex
is diverse, as con?rmed by a thorough literature review.
Most strongly, our study demonstrates that small, monophy-
letic groups within this complex can have considerable diver-
sity in terms of coloration, body size, habitat, and lifestyle. For
instance, the small and solitary, yellow and reddish-coloured,
symbiotic crab, M. cintimanus, was included in the same
group in which M. coryphe was placed, a larger greenish-blue
species that lives in small aggregations and inhabits deeper
waters (Rathbun, 1925; personal observations). The solitary
habit of M. cintimanus might be a consequence of a symbiotic
lifestyle, as suggested for other crustaceans associated with sea
anemones (Baeza et al., 2001; Baeza & Thiel, 2007). Similarly,
the large, reef-dwelling crab, M. spinosissimus, is the sister
species to M. caribbaeus, with populations composed of
small (or large) individuals that inhabit shallow waters. The
absence of reciprocal monophyly between large and small
specimens of M. caribbaeus also suggests that body size is a
highly plastic phenotypic trait in this species. Mithrax carib-
baeus might be used to test the role of environmental con-
ditions (e.g. food, habitat constraints and predation
pressure) in explaining striking size differences in the ?eld
(including dwar?sm) and possible relationships to life
history schedules.
In general, this study has shown that the separation of the
Mithrax?Mithraculus species complex into two genera is not
supported by molecular characters. Crabs from these two
genera demonstrate a considerable diversity of lifestyles, body
sizes, microhabitats, and coloration. Studies describing life
history and ecology within a phylogenetic framework are under-
way. This approach is expected to prove most useful in under-
standing the role of environmental conditions in driving the
evolution of morphological, ecological, and behavioural traits
in the marine environment. The present study included only
species from the Caribbean Sea; nevertheless, the
amphi-American nature of the Mithrax?Mithraculus species
complex (see Rathbun, 1925) suggests that this group might
also be a model to study speciation mechanisms, as in other
trans-isthmian clades of shrimps (Williams et al., 2001), ?sh
(Bermingham et al., 1997), and sea urchins (Lessios, 2008).
ACKNOWLEDGEMENTS
We thank Raphael Ritson-Williams and Alexis Zabala for
their help with collection of spider crabs from the turbid
and harsh waters of Belize and Venezuela, respectively. Gary
Wikfords critically reviewed the English as well as the
content and provided helpful comments. The ?rst author
thanks support from the Smithsonian Tropical Research
856 j. antonio baeza et al.
Institute (STRI, Panama) Marine Fellowship and Smithsonian
Marine Station (Fort Pierce, Florida) Fellowship. Support for
research from the Consejo de Investigacio?n, Universidad de
Oriente, Venezuela is deeply appreciated. The ?rst author
thanks the National Geographic Society, USA for funding a
research trip to Papua New Guinea where part of this manu-
script was written. Especial thanks to Jeff Hunt and Lea Weigt
at the Laboratory of Analytical Biology, NMNH, Washington,
DC for their logistical support. We appreciate the comments
by two anonymous referees that substantially improved our
manuscript. This is contribution number 799 of the SMSFP.
REFERENCES
Abele L. and Kim W. (1986) An illustrated guide to the marine decapod
crustaceans of Florida. Tallahassee: Department of Environmental
Regulation.
Ahyong S.T. and O?Meally D. (2004) Phylogeny of the Decapoda
Reptantia: resolution using three molecular loci and morphology.
Raf?es Bulletin of Zoology 52, 673?693.
Ahyong S.T., Lai J.C.Y., Sharkey D., Colgan D.J. and Ng P.K.L. (2007)
Phylogenetics of the brachyuran crabs (Crustacea: Decapoda): the
status of Podotremata based on small subunit nuclear ribosomal
RNA. Molecular Phylogenetics and Evolution 45, 576?586.
Baeza J.A., Thiel M. and Stotz W. (2001) The life history of
Allopetrolisthes spinifrons, a crab associate of the sea anemone
Phymactis clematis. Journal of the Marine Biological Association of
the United Kingdom 81, 69?76.
Baeza J.A. and Thiel M. (2007) The mating system of symbiotic crus-
taceans. A conceptual model based on optimality and ecological con-
straints. In Duffy J.E. and Thiel M. (eds) Reproductive and social
behaviour: crustaceans as model systems. Oxford: Oxford University
Press, pp. 245?255.
Baeza J.A., Schubart C.D., Zillner P., Fuentes S. and Bauer R.T. (2009)
Molecular phylogeny of shrimps from the genus Lysmata (Caridea:
Hippolytidae): the evolutionary origins of protandric simultaneous
hermaphroditism and pair-living. Biological Journal of the Linnean
Society 96, 415?424.
Bermingham E., McCafferty S.S. and Martin A.P. (1997) Fish biogeo-
graphy and molecular clocks: perspectives from the Panamanian
isthmus. In Kocher T.D. and Stepien C.A. (eds) Molecular systematics
of ?shes. San Diego: Academic Press, pp. 113?128.
Bolan?os J.A. and Scelzo M.A. (1981) Larval development of the spider
crab Mithrax verrucosus Milne-Edwards, reared in the laboratory
(Decapoda: Brachyura: Majidae). The American Zoologist 21, 989.
Bolan?os J.A., Lares L.B. and Herna?ndez J.E. (1990) Desarrollo larval de
Mithrax caribbaeus Rathbun, 1920 (Crustacea: Decapoda: Majidae)
realizado en condiciones de laboratorio. Bolet??n del Instituto
Oceanogra??co de Venezuela 29, 67?89.
Bolan?os J.A., Herna?ndez G. and Lira C. (2000) Mithraculus cinctimanus
Stimpson, 1860 y Speloeophorus pontifer (Stimpson, 1871) (Crustacea:
Decapoda: Brachyura), dos nuevas adiciones a la carcinofauna venezo-
lana. Bolet??n del Instituto Oceanogra??co de Venezuela 39, 25?31.
Boschi E. (2000) Species of decapod crustaceans and their distribution in
the American marine zoogeographic province. Revista de Investigacio?n
y Desarrollo Pesquero 13, 7?136.
Fransozo A. and Hebling N.J. (1982) Desenvolvimento potsembrionario
de Mithrax hispidus (Herbst, 1790) (Decapoda, Majidae) em labora-
torio. Cie?ncia e Cultura 34, 385?395.
Fratini S., Vannini M., Cannicci S. and Schubart C.D. (2005)
Tree-climbing mangrove crabs: a case of convergent evolution.
Evolutionary Ecology Research 7, 219?233.
Goy J.W., Bookhout C.G. and Costlow Jr J.D. (1981) Larval development
of the spider crab Mithrax pleuracanthus Stimpson reared in labora-
tory (Decapoda: Brachyura: Majidae). Journal of Crustacean Biology
1, 51?62.
Huelsenbeck J.P. (2000) MrBayes: Bayesian Inferences of Phylogeny (soft-
ware). New York: University of Rochester Press.
Hultgren K. and Stachowicz J.J. (2008) Molecular phylogeny of the bra-
chyuran crab superfamily Majoidea indicates close congruence with
larval morphology-based trees. Molecular Phylogenetics and
Evolution 48, 986?996.
Hultgren K. and Stachowicz J.J. (2009) Evolution of decoration in majoid
crabs: a comparative phylogenetic analysis of the role of body size
and alternative defensive strategies. The American Naturalist 173,
566?578.
Kass R.E. and Raftery A.E. (1995) Bayes factors. Journal of the American
Statistical Association 90, 773?795.
Lessios H.A. (2008) The great American schism: divergence of marine
organisms after the rise of the Central American isthmus. Annual
Review of Ecology and Systematics 39, 63?91.
Manning R.B. (1970) Mithrax (Mithraculus) commensalis, a new West
Indian spider crab (Decapoda, Majidae) commensal with a sea
anemone. Crustaceana 19, 157?161.
Marques F.P.L., Pohle G.W. and Vrbova L. (2003) On the larval stages of
Macrocoeloma diplacanthum (Decapoda: Brachyura: Majidae), and a
review of Mithracine phylogenetic aspects. Journal of Crustacean
Biology 23, 187?200.
Martin J.W. and Davis G.E. (2001) An updated classi?cation of the
Recent Crustacea. Natural History Museum of L.A. County, Science
Series 39, 1?124.
Melo G.A.S. (1996) Manual de identi?cac?a?o dos Brachyura (Caranguejos e
Siris) do litoral brasileiro, Sa?o Paulo, Brasil. Sa?o Paulo: Editora Ple?iade.
Ng P.K.L., Guinot D. and Davie P.J.F. (2008) Systema Brachyurorum:
Part I. An annotated checklist of extant brachyuran crabs of the
world. Raf?es Bulletin of Zoology 17, 1?286.
Nylander J.A.A., Ronquist F., Huelsenbeck J.P. and Nieves-Aldrey J.L.
(2004) Bayesian phylogenetic analysis of combined data. Systematic
Biology 53, 47?67.
Palumbi S.R., Martin A., Romano S., Mcmillan W.O., Stice L. and
Grabowski G. (1991) The simple fool?s guide to PCR. A collection of
PCR protocols. Honolulu: University of Hawaii.
Pohle G.W. and Marques F.P.L. (2000) Larval stages of Paradasygius
depressus (Bell, 1835) (Brachyura: Majidae) and a phylogenetic
hypothesis for 21 genera of Majidae. Proceedings of the Biological
Society of Washington 113, 739?760.
Porter M.L., Perez-Losada M. and Crandall K.A. (2005) Model-based
multi-locus estimation of decapod phylogeny and divergence times.
Molecular Phylogenetics and Evolution 37, 355?369.
Posada D. and Crandall K.A. (1998) Modeltest: testing the model of
DNA substitution. Bioinformatics 14, 817?818.
Provenzano A.J. and Brownell W.N. (1977) Larval and early post-larval
stages of the west Indian spider crab, Mithrax spinosissimus (Lamarck)
(Decapoda: Majidae). Proceedings of the Biological Society of
Washington 90, 735?752.
Rathbun M.J. (1925) The spider crabs of America. Bulletin of the United
States National Museum 129, 1?613.
molecular phylogeny of mithrax mithraculus crabs 857
Rhyne A.L., Fujita Y. and Calado R. (2004) Larval development and 1st
crab of Mithraculus sculptus (Decapoda: Brachyura: Majoidea:
Mithracidae) described from laboratory-reared material. Journal of
the Marine Biological Association of the United Kingdom 86, 1133?
1147.
Rodr??guez G. (1980) Crusta?ceos deca?podos de Venezuela. Caracas: IVIC.
Ronquist F. and Huelsenbeck J.P. (2003). MRBAYES 3: Bayesian phylo-
genetic inference under mixed models. Bioinformatics 19, 1572?1574.
Santana W., Pohle G. and Marques F. (2003) Zoeal stages and megalopa
of Mithrax hispidus (Herbst, 1790) (Decapoda: Brachyura: Majoidea:
Mithracidae): a reappraisal of larval characters from laboratory cul-
tured material and a review of larvae of the Mithrax?Mithraculus
species complex. Invertebrate Reproduction and Development 44,
17?32.
Santana W., Marques F. and Pohle G. (2004) Larval stages of
Stenocionops furcatus (Olivier, 1791) (Decapoda: Brachyura:
Majoidea) and a reappraisal of larval morphological characters for
Mithracidae. Journal of Plancton Research 26, 859?874.
Scholtz G. and Richter S. (1995) Phylogenetic systematics of the reptan-
tian Decapoda (Crustacea, Malacostraca). Zoological Journal of the
Linnean Society 113, 289?328.
Scotto L.E. and Gore R.H. (1980) Larval development under laboratory
conditions of the tropical spider crab Mithrax (Mithraculus) coryphe
(Herbst, 1801) (Brachyura: Majidae). Proceedings of the Biological
Society of Washington 93, 551?562.
Schubart C.D., Diesel R. and Hedges S.B. (1998) Rapid evolution to ter-
restrial life in Jamaican crabs. Nature 393, 363?365.
Schubart C.D., Neigel J.E. and Felder D.L. (2000) Use of the mitochon-
drial 16S rRNA gene for phylogenetic and population studies of
Crustacea. Crustacean Issues 12, 817?830.
Schubart C.D., Cannicci S., Vannini M. and Fratini S. (2006) Molecular
phylogeny of grapsoid crabs (Decapoda, Brachyura) and allies based
on two mitochondrial genes and a proposal for refraining from
current superfamily classi?cation. Journal of Zoological Systematics
and Evolutionary Research 44, 193?199.
Stachowicz J.J. and Hay M.E. (1996) Facultative mutualism between an
herbivorous crab and a coralline alga: advantages of eating noxious
seaweeds. Oecologia 105, 377?387.
Stachowicz J.J. and Hay M.E. (1999) Mutualism and coral persistence:
the role of herbivore resistance to algal chemical defense. Ecology 80,
2085?2101.
Sturmbauer C., Levinton J.S. and Christy J. (1996) Molecular phylogeny
analysis of ?ddler crabs: test of the hypothesis of increasing behavioral
complexity in evolution. Proceedings of the National Academy of
Sciences 93, 10855?10857.
Swofford D.L. (2002) PAUP: Phylogenetic Analysis Using Parsimony
(and other methods), version 4.010b. Sunderland, MA: Sinauer
Associates.
Tamura K., Dudley J., Nei M. and Kumar S. (2007) MEGA4: Molecular
Evolutionary Genetics Analysis (MEGA) software version 4.0.
Molecular Biology and Evolution 24, 1596?1599.
Wagner H.P. (1990) The genera Mithrax Latreille, 1818 and Mithraculus
White, 1847 (Crustacea: Brachyura: Majidae) in the western Atlantic
Ocean. Zoologische Verhandelingen 264, 1?65.
White A. (1847) List of the specimens of Crustacea in the collection of the
British Museum. London: British Museum.
Williams A.B. (1984) Shrimps, lobsters and crabs of the Atlantic coast of
the eastern United States, Maine to Florida. Washington:
Smithsonian Institution Press.
and
Williams S.T., Knowlton N., Weigt L.A. and Jara J.A. (2001) Evidence
for three major clades within the snapping shrimp genus Alpheus
inferred from nuclear and mitochondrial gene sequence data.
Molecular Phylogenetics and Evolution 20, 375?389.
Correspondence should be addressed to:
J.A. Baeza
Smithsonian Tropical Research Institute
Apartado Postal 0843-03092, Balboa
Anco?n, Republic of Panama
email: baezaa@si.edu
858 j. antonio baeza et al.