The systematics of neotropical orb-weaving spiders in the genus Metepeira (Araneae: Araneidae)

92 Pages·2001·53.6 MB·English

by  W H Piel

THE SYSTEMATICS OF NEOTROPICAL ORB-WEAVING SPIDERS IN
THE GENUS METEPEIRA (ARANEAE: ARANEIDAE)
WILLIAM H. PIEU
CONTENTS
Abstract _ ___. 1 21. Metepeira gosoga Chamberlin and Ivie
Introduction 2 59
Acknowledgments 2 22. Metepeira ohnec New Species 60
Materials and Methods 3 23. Metepeira comanche Levi 62
Collections Examined 3 24. Metepeira pimungan New Species 62
Locality Data Storage and Manipulation 4 25. Metepeira triangidaris (Franganillo) .. 63
Examination and Illustration 4 26. Metepeira arizonica Chamberlin and
Metepeira F. O. P.-Cambridge 5 Ivie 66
Key to Female Metepeira 12 27. Metepeira atascadero New Species .... 67
Key to Male Metepeira 17 28. Metepeira incrassata F. O. P.-
Metepeirafoxi Group 19 Cambridge 68
1. Metepeira datona ChamberHn and Ivie 20 Metepeira ventura Group 71
2. Metepeira desenderi Baert 21 29. Metepeira ventura Chamberlin and
3. Metepeira grandiosa grandiosa Ivie 71
Chamberlin and Ivie 23 30. Metepeira revillagigedo New Species 73
4. Metepeira grandiosa alpina 31. Metepeira celestun New Species 74
Chamberlin and Ivie 24 32. Metepeirauncata F. O. P.-Cambridge ... 76
Metepeira vigilax Group 26 33. Metepeira crassipes Chamberlin and
5. Metepeira cajahamba New Species .... 26 Ivie 77
6. Metepeira glornerabilis (Keyserling) ... 28 34. Metepeira chilapae Chamberlin and
7. Metepeira vigilax (Keyserling) 30 Ivie 78
8. Metepeira rectangula (Nicolet) 32 Metepeira ininima Group 80
Metepeira labyrinthea Group 33 35. Metepeira petatlan New Species 80
9. Metepeiraspinipes F. O. P.-Cambridge .. 34 36. Metepeira minima Gertsch 82
10. Metepeira lacandon New Species 37 37. Metepeirapacifica New Species 84
Metepeira nigriventris Group 38 38. Metepeirajamaicensis Archer 86
1112.. MMeetteeppeeiirraa tniagrraipvaecnatriNse(wTaScpzeacnioewssld) 4308 LIintdeerxature Cited 8918
13. Metepeira calamuchita New Species .. 42
14. Metepeira galatheae (Thorell) 43
Abstract. Ofthe 39 species and three subspecies
15. Metepeira karkii (Tullgren) 46
Metepeira conipsa Group 47 ofthe orb-weaver genus Metepeira in the Americas,
16. Metepeira compsa (Chamberlin) 48 36 species and two subspecies are known to occur
17. Metepeira roraima New Species 53 outside of the U.S. and Canada. Yet, despite their
18. Metepeira gressa (Keyserling) 54 conspicuous webs, diurnal foraging, and relatively
Metepeira incrassata Group 56 common presence, the taxonomy of Metepeira is
19. Metepeira maija New Species 56 poorlyunderstood, probablybecausethegenitaliaare
20. Metepeira inca New Species 58 smallanddifficultto distinguish. Infact,manynames
for species south ofthe U.S. were, at some time, in-
correctly synonymized with the name Metepeira la-
' Museum of Comparative Zoology, Harvard Uni- byrinthea. In this paper, 14 new species are named
versity, Cambridge, Massachusetts 02138. Current (Metepeira atascadero, M. cajahamba, M. calamuchi-
address: Institute ofEvolutionaiyand Ecological Sci- ta, M. celestun, M. inca, M. lacandon, M. maya, M.
ences, Leiden University, 2311 GP Leiden, The ohnec, M. pacifica, M. petatlan, M. pimungan, M. re-
Netherlands; [email protected]. villagigedo, M. roraima, M. tarapaca); 11 newjunior
Bull. Mus. Comp. ZooL, 157(1): 1-92, June, 2001 1
2 Bulletin Museum ofComparative Zoology, Vol. 157, No. 1
synonyms are reported (M. acostai, M. bani, M. dom- [1989]), but this practice can lead to trou-
inicana, M. grinnelli, M. latigyna, M. perezi, M. san- ble. In one case, the behavior of several
ntae,nsMis.);saflievie,cMas.esseodfiteirorsao,neMo.uslvyausryineoonryummi,zeMd.nvaimrgeis- different species was initially studied un-
are reversed; 22 species and two subspecies are re- der the false assumption that they all be-
described (M. arizonica, M. triangidaris, M. chilapae, longed to the same species (e.g., Uetz et
M. Comanche, M. compsa, M. crassipes, M. datona, al., 1982). Clearly, a strong taxonomic
sMo.gad,esMe.ndegrria,ndMi.osgaalaaltphienaae,,MM..ggrlaon7dneiroasbailigsr,anMd.iosgao,- foundation is important for further biolog-
M. gressa, M. incrassata, M. jamaicensis, M. karkii, ical work.
M. minima, M. nigriventris, M. rectangula, M. spi- Ultimately, the relatively small, indis-
nipes, M. uncata, M. ventura, M. vigilax); and a key tinct genitalia and the relatively homoge-
to allMetepeira speciesispresented. Inaddition,sev- neous abdominal patterns are to blame for
eral ecological and life history observations are re- the weakness in our knowledge of Mete-
ported forvarious species.
peira taxonomy. Many ofthese species are
INTRODUCTION undoubtedly hard to distinguish, and this
fact has surely intimidated arachnologists
The absence of a comprehensive revi- from taking on the painful task ofrevising
sion of Neotropical Metepeira has left the the group. In the absence of good distin-
taxonomy of this group in shambles. Over guishing characteristics, the catalogs of
the years, a fair number of species have Bonnet (1957) and Roewer (1942) synon-
been named, particularly by A. F. Archer, ymized the names of many Neotropical
R. V. Chamberlin, and W. Ivie. However, species with the name Metepeira lahyrin-
these efforts have been sporadic and, for thea. Levi's (1977) revision of Nearctic
the most part, scant. For example, the de- species observes that M. labyrinthea is ac-
scription of Metepeira dominicana (Ar- tually limited to the eastern United States.
cher, 1965) provides little information oth- One task in this revision consists of reas-
er than "form typical of Metepeira in all serting the names ofspecies that were im-
respects," a few measurements, and two properly synonymized and clarifying the
unrecognizable figures. Evenwhen species diagnostic characters that are needed to
are properly described they have far less identify them.
taxonomic value when published alone, in
ACKNOWLEDGMENTS
the absence ofa full comparative revision. i
The poor understanding of Metepeira This paper is part ofmy Ph.D. thesis for
taxonomy has persisted despite great eco- the Department of Organismic and Evo-
logical and behavioral interest in this ge- lutionary Biology, Harvard University. I ain
nus. Indeed, many species are obligate or indebted to many people for their help,
facultative social species and offer excel- assistance, and encouragement in this pro-
lent models for investigating genetic and ject. I am especially thankful for the ded-
environmental factors that influence colo- ication and support of my advisors, Her-
ny formation (e.g., Uetz and Cangialosi, bert W. Levi and Edward O. Wilson. I am
1986; Uetz et al, 1987). The monumental grateful that my colleagues in the Depart-
work carried out over manyyears by G. W. ment of Invertebrate Zoology provided
Uetz has made great strides in our under- such a pleasant place to work: Edward
standing of gregarious social behavior in Cutler, Ardis Johnston, Laura Leibensper-
spiders and in risk-sensitive foraging the- ger, Damhnait McHugh, Diana Sherry,
ory in general (e.g., Uetz, 1996). Still, in Van Wallach, and Dee Woessner, among
the absence of solid taxonomic literature, others.
behavioral ecologists have been forced to Field collecting and new specimen ac-
apply informal names to their study ani- quisitions were made possible with the
mals (e.g., Metepeira "atascadero" in Uetz help ofGita Bodner, Fundacion Capacitar,
[1989] or Metepeira "Species A" in Viera Tim Coonan (CINP), Fred Coyle, Dawn
Metepeira • Piel
Fitzpatrick, Germania Jacome, Antdnia turelles de Belgique, Brussels,
Monteiro, Tila Perez, George Putnam, Belgium; L. Baert
Linda Rayor, Grace Smith (NAWF), and JAK A. Kochalka, Ciudad Univer-
J.
George Uetz. I am particularlyindebtedto sitaria, Paraguay
George Uetz for his assistance and corre- JEC Carico, Lynchburg, Virginia,
J.
spondence. United States
I am thankful for the —comments by JMM Maes, Leon, Nicaragua
those who read this paper especially to MACN JM.useo Argentino de Ciencias
the members on my thesis committee: H. Naturales, Buenos Aires, Argen-
W. Levi, N. E. Pierce, and E. O. Wilson. tina; E. A. Maury, C. L. Scioscia
I am also indebted to Kathy Horton for MCN Museu de Ciencias Naturais,
her help in formatting and preparing the Fundagao Zoobotanica do Rio
manuscript and to the Colles Fund for de- Grande do Sul, Porto Alegre,
fraying the costs of publication. Curators Rio Grande do Sul, Brazil; E. H.
at various institutions who lent me speci- Buckup, M. A. L. Marques
mens are listed in the Materials and Meth- MCZ Museum of Coniparative Zool-
ods section. I cannot overstress the value ogy, Harvard University, Cam-
ofmuseum collections andexpertcurators, bridge, Massachusetts, United
without which research in taxonomywould States; H. W. Levi
not be possible. Museum collections are MECN Museo Ecuatoriano de Ciencias
the most important tools available for un- Naturales, Quito, Ecuador; Ger-
derstanding biodiversity. mania Estevez Jacome
MEG
MATERIALS AND METHODS M. E. Galiano, Buenos Aires,
Argentina
Collections Examined. The taxonomic MLJC Maria Luisa Jimenez, Centro de
revision was carried out on specimens bor- Investigaciones Bioldgicas del
rowed from the following collections. The Noroeste, La Paz, Mexico
abbreviations correspond to those listed MLP Museo de Universidad Nacional,
with each record after eveiy species de- La Plata, Argentina; R. F. Arro-
scription. I am grateful to the museums, zpide, C. Sutton
curators, and staff that graciously loaned MNRJ Museu Nacional, Rio de Janeiro,
the material. Brazil; A. Timotheo da Costa
ADC A&M MNSD Museo Nacional de Historia
A. Dean, Texas University,
Natural, Santo Domingo, Re-
College Station, Texas, United
publica Dominicana; Felix Del
States
AMNH AHimsetroircya,n NMeuwseuYmorko,f NUantiutreadl MUSM MMounsteeo de Historia Natural,
Universidad Nacional Mayor de
States; N. Platnick, L. Sorkin
BMNH Natural History Museum, Lon- San Marcos, Lima, Peru; D. Silva
don, England; P. Hillyard MZSP Museu de Zoologia, Universida-
CAS California Academy of Sciences, de de Sao Paulo, Sao Paulo, SP,
eSdanStFartaensc;iCs.co,GrCiaslwioflordnia, Unit- MZUF BMruazsielo; P.ZoVoalnozgoilicnoi,dJe. L".LaLeSmpee-
CV Carlos Valderrama A.; Bogota, cola" Universita di Firenze,
Colombia Florence, Italy; S. Whitman
FSCA Florida State Collection of Ar- NRMS Naturhistoriska Riksmuseet,
thropods, Gainesville, Florida, Stockholm, Sweden; T. Krones-
United States; G. B. Edwards tedt
IRSNB Institut Royal des Sciences Na- PAN Polska Akademia Nauk, Warsza-
4 Bulletin Museum ofComparative Zoology, Vol. 157, No. 1
DMAAG ONG
wa, Poland; Proszynski, A. Slo- ing contour maps, such as
J.
jewska, W. B. Jedryczkowski aeronautical maps; in most cases, elevation
REL R. E. Leech, Edmonton, Alber- was estimated using NOAA data with an
ta, Canada on-line database server (http://phylogeny.
SMF Forschungsinstitut Sencken- harvard.edu/~piel/find.htiril).
berg, Frankfurt am Main, Ger- The enhanced locality database was
many; M. Grasshoff used to reveal ecological and life history
SR Susan Riechert, Knoxville, Ten- traits. Seasonality ofspecies was expressed
nessee, United States by plotting a circular histogram showing
USNM National Museum of Natural the relative amount of collecting activity
History, Smithsonian Institution, per 5-day interval (Figs. 300-337). While
Washington, D.C., United locality dates alone cannot control for the
States; Coddington, S. F. seasonal activityofhuman collectors, these
J.
Larcher data at least provide an estimate ofspider
ZMB Zoologisches Museum der seasonal abundance, if only approximate.
Humboldt Universitat, Berlin, Some syiupatric species show incongruous
Germany; M. Moritz seasonal abundance, which is at least some
ZMUC Zoologisk Museum, Gopenha- evidence that seasonality of spider collec-
gen, Denmark; H. Enghoff, N. tors does not unduly overshadow the sea-
Scharff sonality of the spiders themselves.
ZSM Zoologische Staatssammlung, Examination and Illustration. Speci-
Munich, Germany mens were examined under 80% ethanol
in a dish with light and dark sand grains
Locality Data Storage and Manipula- for specimen support. Digital photographs
tion. Localitydata from eachcollectionvial of preserved specimens were taken
were entered into a database designed us- through a Nikon SMZ-10 photomicro-
ing Glaris FileMaker Pro®. Geographic scope using a Panasonic WV-CL320 GGD
coordinates were added to locality data video camera, chosen for its high sensitiv-
that lacked them using maps, USBGN gaz- ity to light. Video images were captured
etteers, and on-line databases (http://164. using a Quicklmage®24 digitizer and ed-
214.2.59/gns/html/ and http://mapping. ited on a Quadra 700 Macintosh® com-
usgs.gov/www/gnis/). Occasionally locality puter. The computer allows relatively in-
information was illegible or unknown or expensive pictures to be printed rapidlyon
one of several homonymous sites. In such a 1,200 dpi Xante® Accel-a-Writer 8200
cases a reasonable, educated guess was laser printer. Digital pictures were used to
inade and a "[?]" designation was append- help sort out individuals to species, to cre-
ed to the locality. In some cases the itin- ate publishable pictures of gross dorsal
erary ofa collectorwas reconstructedfrom and ventral markings, and to aid in the il-
other known records, and the ambiguous lustration ofgenitalia. As an aid in illustra-
locality was assigned a coordinate halfway tion, the digital pictures functioned as a
between the previous and followingknown camera lucida because they assured accu-
collection sites. The locality database racywhen drawing the proportions ofgen-
worked in concert with the mapping pro- ital parts and sclerites. Usually a digital
gram Atlas Pro® to generate thematic picture was laid over carbon paper and an
maps on the fly. These maps helped in the outline of the genitalia was transferred to
process of delimiting species and discov- coquille board underneath. The illustra-
ering cryptic species. tion continued on the coquille board using
Elevation (in meters) was estimated for a Staedtler OmniGhrom® pencil and a
each locality that lacked this information. drafting pen with India ink and then was
In some cases, elevation was estimated us- scanned at 600 dpi on a LaGie Silverscan-
Metepeira • Piel
ner II®. The resulting digital image was medians. Eye separations were expressed
edited in Adobe Photoshop® and reduced in terms of their own diameters, or in
in size to 1,200 dpi. The edited figures terms ofthe anterior lateral eyes when be-
were finally arranged on plates using Can- tween eyes of different types. Oval eyes
vas®. were measured as an average of the lon-
External genital structures were manip- gest and shortest lengths.
ulated with pins to reveal hidden parts. In parallel with the last revision of Me-
The terminal division on the male palp is tepeira (Levi, 1977), leg measurements
hinged, so it had to be pried open to see were made on each article distal to the tro-
the embolus and embolic apophyses prop- chanter for the first leg and on the com-
erly. In females, mating plugs had to be bined lengths ofthe patellae and tibiae for
removed from epigynal openings using all remaining legs. Variation in total body
pins. Sometimes the entire epigynum was size was provided as an average, minimum,
partly cut from the body so as to see it and maximum of the total lengths from a
from a posterior view. number ofmature specimens, usuallycho-
Internal genital structures were studied sen from a wide geographic spread.
by clearing them in clove oil and examin-
ing them using an Olympus BH-2 com- Metepeira F. O. P.-Cambridge
pound microscope. Sketches were made Metepeira F. O. P.-Cambridge, 1903: 457. Type spe-
directly on the computer in Canvas® by cies by original designation M. spinipes F. O. P.-
aiming the camera lucida at the computer Cambridge 1903. The name is feminine.
monitor. While internal genital structures Diagnostic Abstract. Web combines bar-
helped in the process of delimiting spe- rier or scaffolding structure surrounding a
cies, they did not prove to be as useful as classic araneid orb with a retreat suspend-
external genital structures in describing ed in air (Fig. 1). Like a raccoon with its
species; thus, these working sketches are facial colors reversed, the eye region is
not figured herein. lighter than any other part ofthe carapace
Measurements of the spiders were tak- (Fig. 2). The venter has a wide median
en using a Leitz stereo dissecting micro- white line set on a black background that,
scope with a calibrated reticule. Sizes of with only some exception, extends anteri-
leg articles, eyes, and carapace, were per- orly on the sternum (Fig. 3). With one ex-
formed on one specimen of each sex, for ception, the total lengths of distal leg ar-
each species. The respective localities of ticles (metatarsus and tarsus) exceed that
the candidate specimens were indicated in of the middle articles (patella and tibia).
the descriptions. This study placed little The median apophysis has two distinctive
reliance on spider leg measurements be- flagella (F in Fig. 5) and, in some species,
cause they are not usually very useful in an easily recognizable keel (K in Fig. 5).
spider taxonomy, and because Metepeira The dorsal abdominal markings (the foli-
species are notorious for theirvariabilityin um) look like an inverted fleur-de-lis, al-
size (Levi, 1977; Piel, 1996). lowing easy recognition ofthe genus in the
All eye sizes were reported as a ratio of field (Fig. 2).
the posterior median eye diameters to the Description and Diagnosis. For field
diameter of every other eye type. For ex- ecologists, the most obvious and distinctive
ample, in the case of "ratio of eye diame- feature ofMetepeira is the combination of
ters: posterior medians and anterior me- orb and barrier web (Fig. 1). The barrier
dians 2.0, anterior laterals 0.5, posterior \^eh forms scaffolding around an almost
laterals 1.0," the reader should interpret vertical orb and supports the spiders re-
the anterior medians to be halfthe size of treat, which is thus suspended away from
the posterior medians, andthe anteriorlat- any substrate.
erals to be twice the size of the posterior In contrast to most araneids, the cara-
6 Bulletin Museum ofComparative Zoology, Vol. 157, No. 1
pace ofMetepeira is lightest in the eye re- unique among araneid genera is the com-
gion. However, this distinctive feature bination of median white line on the ven-
varies within the genus: in the case of M. ter and median white line on a black or
rectangula (Nicolet, 1849), the hghter re- brown sternuin. Some Metepeira species
gion takes up ahiiost half the carapace lack a complete white line on the sternum,
(Fig. 65); in the case of Metepeira F. O. but even those, such as M. datona, that
P.-Cambridge, 1903, the lighter region is usually have an entirely black sternum
usually limited to the anterior edges ofthe nonetheless sho\v hints of white markings
carapace (Fig. 2). White, downy hairs of- in some specimens. Characteristics found
ten cover the carapace but are especially in the carapace, abdomen, and sternum of
white and conspicuous on the lighterparts Metepeira are also found in Araneus
of the carapace outside the eye region. In koepckeorurn Levi, but this last species
some species, such as M. spinipes, these lacks the white line on the venter.
hairs make the carapace look gray or sil- With the exception ofM. datona, and in
very when the spider is alive, but dark some cases, M. desenderi, all Metepeira
brown when the spider is in ethanol. species have a combined metatarsus and
The eyes of Metepeira are not particu- tarsus that is longer than the combinedpa-
larly unusual. Eye separations relative to tella and tibia. This feature is unusual
eye diameters increase with spider size: among araneids and is not found in Kaira
larger spiders tend to have relativelygreat- O. P.-Cambridge or other likely relatives
er eye separations. In either sex, the pos- to Metepeira (Levi, 1977; Piel and Nutt,
terior median eyes are between 1.1 and 1997).
1.7 times the size ofthe anterior medians, In most species the leg articles are
and the separation between posterior me- ringed, usually with brownish black on the
dian eyes is between 0.4 and 0.7 of that distal and dorsal surfaces of each article,
between anterior median eyes. The sepa- except for the patellae and tarsi which are
ration between the anterior median eyes usually entirely dark. In mainly tropical
and the anterior lateral eyes is between 1 and high-altitude species, the coxae are
and 3.7 times the size of anterior median mostly black (e.g.. Fig. 75), but in desert/
eyes in males and between one and five mesquite species they appear yellowish
times the size in females. The diameter of white (e.g.. Fig. 28). — I
the anterior median eyes exceeds the Unlike many other araneids and—per-
height of the clypeus. haps because of the small male size the
The shape ofthe female abdomen rang- coxa on leg I of male Metepeira lacks the
es from wider than long and rhomboid (M. hook and corresponding groove typically
datona. Fig. 12) to roundish (e.g., M. de- found on femur II. In addition, males lack
senderi. Fig. 20; M. rectangula. Fig. 65), a tooth on the lateral side of the endite,
to longer than wide and oval (e.g., M. inca. and they lack a basal tooth on the palpal
Fig. 169). The dorsal folium has a recog- femur. The phylogenetic analysis of
nizable white fleur-de-lis pattern, usually Scharffand Coddington (1997) incorrectly
on a dark background, its edges shaped by codes Metepeira as having a tooth on the
a wavy, zig-zag white outline (Fig. 2). The endite. However, had the authors coded
dors—um of live spiders is often more red- this character as absent, they would have
dish a pigment that rapidly dissolves in decreased the length of their preferred
alcohol. tree because the nearest relatives hypoth-
Somewhat less common among other esized for Metepeira {Kaira, Zijgiella, and
araneids is the median white line on the Singa) also lack this tooth.
venter of the abdomen (Fig. 3), which is Macrosetae usually concentrate on arti-
present (though shortened) even in the cles that contact other spiders during mat-
most darkly pignriented species. However, ing or grappling. In contrast to most gen-
Metepeira • Piel 7
BarrierWeb WhiteDorsal LightEyeRegion
Fleur-de-Lis / ^^ MedianWhite
Pattern xAW* ^Ma LineonSternum
MedianWhite
LineonBlack
OrbWeb Venter
vigilax nigriventris incrassata minima
foxi labyrinthea compsa vcntura
Figure 1. Webofimmature Metepeiragrandiosaalpinafrom Chihuahua, Mexico.
Figure2. DorsumofadultfemaleMetepeiracrassipes.
Figure3. Venterofadultfemale Metepeiratarapacanewspecies.
Figure 4. Hypothetical phylogenetic relationships among Metepeiraspeciesgroups. Shaded branchesindicatespeciesgroups
thatliveinSouthAmerica;openbranchesindicatespeciesgroupsthatliveinNorthAmerica,CentralAmerica,andtheCaribbean.
Abbreviations: DEA, distal embolic apophysis; K, keel of median apophysis; TA, terminal apophysis; (+), characterstategain;
(-), characterstate loss.
Figures 5, 6. Male palpus. 5, mesal view, Metepeira compsa. 6, ventral view ofdistal embolic division, Metepeiralabyrintliea
(Hentz).
Abbreviations:BEA, basalembolicapophysis;C,conductor; DEA,distalembolicapophysis; E,embolus; F,flagellumonmedian
apophysis; K, keel of median apophysis; MA, median apophysis;TA, terminal apophysis;TO, terminaldivision.
Figures7-13. MetepeiradatonaChamberlinand Ivie(sp. 1; 17°53'N,76°19'W).7, malepalpus, mesal.8,epigynum,posterior.
9, epigynum,ventral. 10, male, dorsal. 11, male, ventral. 12, female, dorsal. 13, female, ventral.
Scalebar:dorsum andventerfigures 1.0 mm.
—
8 Bulletin Museum ofComparative Zoology, Vol. 157, No. 1
era related to Araneus, Metepeira has con- palp is very similar in almost all species.
centrated macrosetae on femur I instead When this structure is pulled up, a basal
oftibia II (Scharffand Coddington, 1997). embolic apophysis (also known as an em-
Female Metepeira have between two and bolar lamella) can be seen in the shape of
five macrosetae on die anterior side ofthe a club or spatula (E in Figs. 5, 6). Soiue-
femur, and between zero and seven on the times a distal embolic apophysis can be
anteroventral side. Males typically have seen if it is not hidden from view by an
inore setae than their conspecific females: overhanging terminal apophysis. When the
four to nine on the anterior side and two termi—nal apophysis is large and sclero-
to nine on the anteroventral side. Variation tized which is die case in—all but the Me-
in the number of macrosetae appears to tepeira foxi species group it has a rec-
correlate with body size. In most species, ognizable toothed notch, like the mouth
the inale palpal tibia and patella each have on a wrench (Fig. 6). Virgin males have a
two strong macrosetae (Levi, 1977, fig. 8). cap on the embolus that remains in the
Compared with other araneid genera, epigynum after mating and presumably
Metepeira have rather small and similar serves as a barrier to subsequent mating
genitalia, which on the one hand makes (Levi, 1977). The shape of the embolus
the genus easy to recognize, but on the cap varies from tiny (e.g.. Fig. 178) to
other hand makes species tough to iden- short but wide (e.g., Fig. 199) to large and
tify. The small epigynum is fleshy, variable winged (Fig. 46). Fin—ally, the terminal di-
in shape, and weakly sclerotized. Unlike vision lacks a stipes a sclerite between
Araneus, Metepeiras scape never has a the radix and the embolus that is frequent-
pocket but always ends with a pointed tip ly found in other genera related to Ara-
(e.g.. Fig. 31). The cleared epigynum neus (Scharff and Coddington, 1997).
and—in many cases the uncleared epigyn- Natural History. All Metepeira species
um reveals a pair ofsclerotized spherical build a unique web that combines an orb
structures where the embolus is inserted, with a barrier web (Levi, 1977; Lopez,
aswell as ducts to pass semen to the larger, 1993). As with Cijrtophora Simon or Me-
spherical seminal receptacles. In some cynogea Simon (Levi, 1997), the retreat of
species, these spherical structures arewide Metepeira hangs in the air, away from sub-
apart (e.g.. Figs. 16, 17), in others they are strate, and is suspended by a scaffolding
tubular (e.g.. Figs. 39, 40), but in many, structure created by the barrier web (Fig.
they are closer together (e.g.. Figs. 93, 94). 1). The spider detects vibrations in the
Frequently the deeper, large seminal re- web and gains quick access to the hub us-
ceptacles can be seen through uncleared ing a signal line that runs from the retreat
tissue (e.g.. Figs. 201, 295). to the center of the orb (Fig. 1). Tan col-
The male palp is more distinctive. In ored egg sacs are strung together, usually
particular, the median apophysis (MA in above the retreat, and the most recently
Fig. 5), while not always a good character laid eggs are nearest to the spider. In some
for separating closely related species, is ex- species the egg sacs and retreat are deco-
cellent when it comes to identifying the rated with insect parts (e.g., M. spinipes);
genus. Two flagellae (F in Fig. 5) grace- in other species they are carefullywrapped
fully curve off the base of the median by leaves and woven together (e.g., M. da-
apophysis, and in some species, a toothed tona). Unlike the webs oiCijrtophora and
or smooth keel (K in Fig. 5) extends in the Mecynogea, the orb web of Metepeira is
opposite direction. This design is also seen orientedvertically, and the numberofradii
in Kaira, Aculepeira, and Amazonepeira, and sticky spirals are more typical ofother
but none of these have flagellae that ap- araneines.
pear so integral to the base structure. In some species, such as M. pimungan
The tenninal division on the Metepeira (personal observation) and, to a lesser de-
Metepeira • Piel
gree, M. incrassata (G. Uetz, personal havior, broadly spread across seven differ-
communication), juveniles and adults ent species groups, may mean that aggre-
without eggs will live on webs lacking a gation is a frequentlylost and relativelyold
suspended retreat. Instead, the spider sits trait, orit may mean that species areprone
on a white disk-shaped stabilimentum in to converge and evolve the same behavior
the center of the hub. Of 110 M. pimun- independently.
gan specimens observed on San Miguel Is- Either way, much research has focused
land, about 40% occupied webs of this on elucidating the selective forces behind
type. In two cases the disk stabilimenta colonial behavior in Metepeira. In partic-
were partly separated from the hub by ular, Uetz (1988a,b, 1996) has provided
barrier web lines and were further bent strong support for the hypothesis that Me-
over to form a partly covered protective tepeira forage using a risk-sensitive strat-
retreat for the spider. This observation egy. He suggests that spiders in abundant
makes it possible to imagine that the disk habitats seek to minimize individual vari-
stabilimentum seen in M. pimiingan re- ance in prey capture by aggregating in col-
sults from the fusion ofthe suspended re- onies, whereas spiders in poor habitats
treat with the hub. seek to maximize variance by living soli-
When food supplies are plentiful, spi- tarily—perhaps in a risky attempt to find
ders of all kinds show an increased toler- areas of local prey maxima. The diversity
ance for one another and an increasedten- of social tendencies among species is
dency to aggregate (e.g., Gillespie, 1987; therefore commensurate ofthe diversityof
Rypstra, 1986). The suspended retreats ecological habitats that they inhabit.
and barrier webs ofMetepeira, Cyrtopho- Indeed, Metepeira species thrive in a
ra, and Mecynogea mayactuallyfurtherfa- wide array of habitats, though often they
cilitate in the formation ofaggregations by are quite harsh. These include wet, mon-
easing dependency on substrate availabili- tane cloud forests in Mexico and Panama
ty and byproviding a common support sys- (M. incrassata, M. olmec); tropical andwet
tem (Burgess and Witt, 1976; Uetz, 1986). agricultural areas (M. uncata, M. vigilax,
In any case, colony formation is known to M. glomerabilis, M. roraima); high-eleva-
occur in all three genera (e.g., Rypstra, tion pine forests (M. lacandon, M. nigri-
1979), but especially in Metepeira. Small ventris, M. grandiosa alpina); Canadian
colonial aggregations oftwo to 10 individ- bogs (M. grandiosa palustris); deciduous
uals occur in M. datona (Spiller and forests in the eastern U.S. (M. labyrin-
Schoener, 1989), M. minima (personal ob- thea); Caribbean coastal shrubbery (M. da-
servation), M. glomerabilis (R. Baptista, tona, M. minima, M. triangularis, M. ja-
personal communication), and M. atascad- maicensis, M. maya, M. celestun); Mexican
ero new species (e.g., Uetz and Hodge, mesquite grasslands (M. atascadero, M.
1990). Medium-size colonies of 10 to 30 chilapae); Patagonian dunes and scrub (M.
individuals occur in M. pimungan (person- galatheae) and pampas grass (M. karkii);
al observation), M. gressa (Viera and Cos- diy Californian buckwheat and sage (M.
ta, 1988), M. nigriventris (L. Rayor, per- crassipes, M. ventura, M.foxi, M. grandio-
sonal coinmunication), M. tarapaca (V. sa grandiosa); and arid and semiarid de-
Roth, locality label), and M. spinipes (e.g., serts (M. arizonica, M. inca, M. ventura,
Uetz, 1988a). Large colonies, sometimes M. crassipes). Although some species (e.g.,
in the thousands ofindividuals, commonly M. galatheae, M. spinipes, M. cornpsa) cov-
occur in M. incrassata (e.g., Uetz and er vast geographic areas and live in many
Hodge, 1990). Near rivers and in other different habitats, many species are more
lush habitats, M. tarapaca colonies can biogeographically restricted. In fact, sev-
reach 200 individuals (M. Roy, personal eral species follow narrow ecological zones
communication). These cases ofsocial be- that decrease in elevation with distance
10 Bulletin Museum ofComparative Zoology, Vol. 157, No.
from the equator (e.g., M. rectangula, M. tatum. Colonial spiders, such as M. incras-
vigilax, M. cajahamha. Fig. 36; M. arizon- sata, are especially vulnerable to wasps,
ica. Fig. 213). other spiders, sarcophagid flies (e.g., Ar-
Close cohabitation with different inter- achnidomyia Undue, A. rayorae), and
and intrageneric species is not uncommon. hummingbirds (Hieber and Uetz, 1990;
Colonies of M. incrassata are known to Lopez, 1989; Rayor and Uetz, 1990).
contact webs ofNephila clavipes Linnaeus Species Groups. Nearctic Metepeira
(Hodge and Uetz, 1996) and Mecijnogea were divided into two species groups: the
ocosingo and Gasteracantha cancriformis M. labyrinthea group and the M. foxi
(personal observation). Often M. crassipes, group, based on the pattern on the ster-
M. Ventura, M. foxi, and M. grandiosa num and the shape of the median apoph-
grandiosa are collected together (Levi, ysis (Levi, 1977). Baert (1987) questioned
1977), as are M. minima and M. celesiun the taxonomic usefulness of the M. foxi
(personal observation). Species that have species group (M. foxi, M. grandiosa, M.
been collected from identical localities, datona) because he found that M. desen-
though not necessarily at the same time, deri has both a keel on the median apoph-
include: M. chilapae and M. spinipes; M. ysis—and a white sternal line (Figs. 15,
chilapae and M. atascadero; M. karkii and 21) a combination that is incoinpatible
M. galatheae; M. calamuchita new species, by Levi's scheme. Nonetheless, the geni-
M. gressa, and M. galatheae; M. rectan- talia of M. desenderi closely ally this spe-
gula, M. calamuchita, and M. galatheae; cies with the M. foxi group, so I am re-
M. compsa and M. gressa; M. vigilax and defining the M.foxi group based on purely
M. compsa; M. glomerahilis andM. vigilax; genitalic characters. This is likely to be a
M. compsa and M. glotnerabilis; M. comp- basal, paraphyletic group (Fig. 4) (Piel and
sa and M. nigriventris; M. compsa and M. Nutt, 1997).
mcfl; M. datona and M. jamaicensis; and Seven additional species groups are dis-
M. datona and M. triangularis. tal to the M.foxi. These remaining species
Despite the wide biogeographic ranges are united by sharing a large terminal
of M. compsa (Puerto Rico and south to apophysis that is sclerotized and usually
Argentina, Map 8) and M. datona (His- studded with teeth or denticles. The M.
paniola and north to Florida, Map 1), they vigilax group (M. vigilax, M. cajahamha,
nonetheless come geographically close to M. glomerahilis, M. rectangida) are united
one another but do not overlap. It is hard by large emboli with long scooplike basal
to imagine that the hurricanes that fre- embolic apophyses (Fig. 60). Unlike the
quently pass through the Caribbean, as remaining spe—cies, the term—inal apophysis
well as the homogeneous island environ- in this group albeit large does not ac-
ments, would not gradually cause these tually overhang or hide the embolus. In
two species distributions to overlap. Per- addition to an overhanging terminal
haps these abrupt, disjunct distributions apophysis, the remaining taxa are also
are a rare example of competitive exclu- united by a distal embolic apophysis that
sion in Metepeira, which in other species either protrudes (Fig. 76), curves off (Fig.
is not thought to be an important factor 185), or is secondarily lost (Fig. 264). The
(Wise, 1983). M. labyrinthea group (M. labyrinthea, M.
Sphecid wasps are predators on Mete- lacandon, M. spinipes) share a toothless,
peira. Locality labels indicated that M. pa- smooth keel on the median apophysis
cifica has been found in the nests of Try- (Figs. 67, 69).
pargilum nitidum, T. tenoctitlan, and T. The M. nigriventris group and the M.
hensoni. Jimenez and Tejas (1994) report compsa group together share a median
that M. crassipes is the most frequent prey apophysis with teeth on the face of the
item in the nests of Trypargilum triden- keel (Figs. 92, 149). The M. incrassata