Systematic studies on Pseudomyrmex acacia-ants (Hymenoptera: Formicidae: Pseudomyrmicinae).

53 Pages·1993·1.7 MB·English

by  Ward, P. S.

6
21Sf
HYM. RES.
J.
2(1), 1993 pp.117-168
Systematic studies on Pseudomyrmex acacia-ants
(Hymenoptera: Formicidae: Pseudomyrmecinae)
Ward
Philip S.
DepartmentofEntomology,UniversityofCalifornia,Davis,CA 9561
—
Abstract. Theobligateacacia-ants (Pseudomyrmexferrugineusgroup) arewellknown as defensiveinhabitants
ofswollen-thornacaciasinthenorthernNeotropics. Ataxonomicrevisionoftheseantsleadstotherecognitionof
ten species: P. ferrugineus (F. Smith), P. flavicornis (F. Smith), P. janzeni, sp. nov., P. mixtecus, sp. nov., P.
nigrocinctus (Emery),P.particeps, sp. nov., P.peperi(Forel), P. satanicus(Wheeler),P. spinicola (Emery), and
P. veneficus(Wheeler). Thefollowingnewsynonymyisproposed:P.nigrocinctus=P.alfari(Forel)=P. bicinctus
(Santschi) =P.peltatus (Menozzi); P. spinicola =P. atrox(Forel) =P. gaigei(Forel) =P. infemalis (Wheeler)=
P.scelerosus(Wheeler).Diagnosticdescriptionsandtaxonomiccommentsarealsoprovidedfortenotherunrelated
speciesofPseudomyrmexwhichhavebecomesecondarilyassociatedwithswollen-thornacaciaseitherasobligate
and, in at least one case, parasitic occupants {P. nigropilosus (Emery), P. simulans Kempfand P. subtilissimus
(Emery);P. reconditus, sp.nov.,mayalsobelonginthiscategory)orasfacultativeinhabitants P. boopis(Roger),
(
P. gracilis(Fabricius),P. hesperius, sp. nov.,P. ita(Forel), stat. nov.,P. kuenckeli(Emery) andP. opaciceps, sp.
A
nov.). cladisticanalysisoftheP.ferrugineusgroupyieldsthefollowingresultwhichappearstobefairlyrobust
insofar as there is congruence among the trees derived from worker-, queen-, and male-based character sets:
((nigrocinctus + particeps) + (peperi + ((satanicus + spinicola) +ferrugineus complex))). The "ferrugineus
complex" comprisesfivespecies whosephylogeneticrelationships arenotfullyclarified. Thecompositedataset
(47charactersfromallthreecastes) supportsthefollowingpartialresolution: (ferrugineus+janzeni+(flavicornis
+ (mixtecus + veneficus))). The cladogram ofthe P.ferrugineus group indicates that speciation in the group has
occurred primarily as a consequence of geographical isolation, and that the ants and their host acacias have
experienced diffusecoevolutionratherthan strictcospeciation.
INTRODUCTION mentshaveappearedintheecologicalliterature. In
this paper I present a taxonomic revision of the
Pseudomyrmexferrugineus (F. Smith) and re- obligate acacia-ants (Pseudomyrmexferrugineus
lated species of ants form a well-defined mono- group) and an assessment of their phylogenetic
phyletic group, the members ofwhich nest exclu- relationships. Ialsoattempttoclarifytheidentities
sivelyinthehollow,swollenthornsofseveralNew ofother,unrelatedspeciesofPseudomyrmex^\\ic\\
WorldAcac/aspecies. Becauseoftheiraggressive havebecomesecondarilyassociatedwithswollen-
behaviorandpredictableoccurrenceontheacacias, thorn acacias.
these ants have received considerable attention The earliertaxonomic literature on acacia-ants
fromtropical biologists (Belt 1874; Safford 1922; isscatteredinmorethanadozenpaperscontaining
Skwarra1934a, 1934b;Wheeler1942;Janzen1966, descriptions of various species, subspecies, and
1973). The landmark studies of Janzen (1966, "varieties". Twoofthemorecomprehensivetreat-
1967b) provided strong experimental evidence of ments are those of Emery (1890) and Wheeler
the mutualistic nature ofthe PseudomyrmexlAca- (1942). In presenting the results ofhis ecological
cia association, and the relationship between the studies Janzen (1966, 1967b, 1973) summarized
two organisms is often cited in discussions of his understanding ofacacia-ant taxonomy. Ward
coevolved mutualisms (e.g. Gilbert 1983; Beattie (1989) provided a brief diagnosis of the P.
1985; Futuyma 1986). At the same time, the ferrugineus group, together with taxonomic and
systematics ofthe acacia-ants has been neglected, nomenclatural notes on the commoner species.
withtheresultthatmisidentificationsandmisstate-
b
118 Journalof Hymenoptera Research
MATERIALS ANDMETHODS Paris, France
MZSP Museo de Zoologia da Universidade de
Collections
Sao Paulo, Brazil
NHMB
Naturhistorisches Museum, Basel, Swit-
Material forthe present study was examinedin
zerland
the following collections: NHMV
AMNHAmericanMuseumofNaturalHistory,New Naturhistorisches Museum, Vienna, Aus-
York, NY, USA tria
ANSP Academy of Natural Sciences, Philadel- PSWC P.S. Ward Collection, University ofCali-
phia, PA, USA fornia atDavis, CA, USA
BMNH The Natural History Museum, London, SEMC SnowEntomologicalMuseum,University
ofKansas, Lawrence, KS, USA
U.K.
CASC CaliforniaAcademyofSciences,SanFran- UCDC Bohart Museum of Entomology, Univer-
cisco, CA, USA sity ofCalifornia at Davis, CA, USA
UCRC UCR
CHAH Entomological Collection, Univer-
C.H.A. Hespenheide Collection, Univer-
sityofCaliforniaatLosAngeles,CA,USA USNM NsiattyioofnaClaliMfuorsneiaumat RoifveNrasitduer,alCAH,isUtoSrAy,
CISC California Insect Survey, University of
CUIC CCaolrinfeolrlniaUnaitvBeerrskietlyey,InCsAe,ctUSCoAllection, WPMCWWa.sPh.inMgatconK,ayDCC,olUlSecAtion, El Paso, TX,
USA
EBCC EItshtaaccai,oNnYd,eUBSiAologia Chamela, Jalisco, ZMHB Zoologisches Museum, Museum fiir
Naturkunde der Humboldt-Universitat,
Mexico
Berlin, Germany
FFIC Fernando Fernandez Collection, Santa Fe
ZMUC Zoologisk Museum, University of
de Bogota, Colombia
GBFM Graham B. Fairchild Museo de Copenhagen, Denmark
ZMUH
Zoologisches Institut und Zoologisches
Invertebrados, Universidad de Panama,
Museum der Universitat Hamburg, Ger-
Panama
GCWC G.C. & J. WUhSeeAler Collection, Silver ZSMC Zmoaonlyogische Staatssammlung, Munich,
Springs, FL,
Germany
INBC Instituto Nacional de Biodiversidad (col-
lectionspreviouslyheldinMNCR: Museo
Special mention should be made of the very
Nacional de Costa Rica), San Jose, Costa
large and important series ofPseudomyrmex col-
Rica
lectedbyD.H. Janzenfrom 1963 to 1974andnow
INHS IllinoisNaturalHistorySurveyInsectCol-
lection, Champaign, IL, USA housed in the Natural History Museum of Los
Angeles County (LACM). The Janzen material
JTLC J.T. Longino Collection, Evergreen State
College, Olympia, WA, USA includesalargenumberofpinnedspecimens (usu-
KSUC KansasStateUniversityInsectCollection, ally glued to the side ofthe pin rather than point-
Manhattan, KS, USA mounted)andanextensivealcoholcollection(partly
LACM Natural History Museum ofLos Angeles overlapping with the pinned series but including
County, Los Angeles, CA, USA additional accessions). Janzen's field notes per-
MCSN Museo Civico di Storia Naturale, Genoa, tainingtothecollectionoftheseantshavealsobeen
deposited in LACM. Obligate acacia-ants {P.
Italy
MCZC MuseumofComparativeZoology, Harvard ferrugineus group) constitute the bulk ofthe col-
University, Cambridge, MA, USA lectedmaterial. Theyoccuraslongnestseriesfrom
MHNGMuseum d'Histoire Naturelle, Geneva, throughoutCentralAmerica,withveryusefulqueen-
male-workerassociations, making this material of
Switzerland
MNHNMuseum National d'Histoire Naturelle, inestimable value to the currentrevision.
:
Volume2, Number 1, 1993 119
When the Janzen collection was received at The following measurements and indices are
LACM in 1984 most specimens had only code cited in this study (the first six measurements are
—
numbersassociatedwiththem amongthepinned taken with the head in a full-face, dorsal view):
HW
specimens a single individual per nest series typi- Head width: maximum width of head, in-
callycontainedacodenumber,withtheremaining cluding the eyes.
— VW
specimensbeingunlabelled andinsomeinstances Vertex width: width oftheposteriorportion
difficulties arose in retrieving full data for coded ofthe head (vertex), measured along a line
specimensfromJanzen'sfieldnotes. Inothercases drawn through the lateral ocelli.
thefieldnotescontradictedtheapparentidentityor HL Headlength: midline length ofheadproper,
composition of a nest series. This latter problem fromthe anteriorclypealmarginto themid-
applied mainly to pinned specimens; the alcohol point of a line drawn across the "occipital"
material appearedtobereliably labelled orcoded, (i.e. posterior) margin.
i.e. the contents of the vials agreed with the field EL Eye length: length of compound eye; note
notes. Thanks to the efforts of Roy Snelling and that this is measured with the head in full
JackLongino,whoincorporatedtheJanzencollec- face, dorsal view, unlike EW(below).
tionintotheLACM,manyofthesediscrepanciesor OD Ocellardistance:distancefromthemiddleof
uncertainties were resolved, but there remains a the median ocellus to the midpoint ofaline
residueof"problemmaterial"forwhichcollection drawnbetween the lateral ocelli.
dataarelackingorambiguous. Amongthepinned OOD Oculo-ocellar distance: distance from the
specimens this comprises twelve drawers in the middleofthemedianocellustothemidpoint
LACMcollectionwhichhavebeenspecificallyset ofaline drawn across the posteriormargins
asidefromthemaincollection. Noneofthisprob- ofthecompoundeyes (this distanceisnega-
lematical material has been cited in the present tive in value if the posterior margin of the
study, but I have examined it and determined that compoundeyeexceedsthemedianocellus).
no additional species are represented there. Al- MFC Minimumfrontalcarinaldistance:minimum
though omission ofthis materialmeans thepoten- distance between the frontal carinae, poste-
tiallossofsomelocalitydata,Ihaveexaminedthe riortotheirfusionwith,orapproximationto,
entirealcoholcollectionandpoint-mountedrepre- the antennal sclerites.
sentative samples so that geographic coverage re- ASD Antennal sclerite distance: maximum dis-
mainsextensive. ThemainpinnedseriesofLACM tancebetweenthelateralmarginsoftheme-
acacia-ants, i.e. thatforwhich accuratedatalabels dianlobesoftheantennalsclerites,measured
are available, comprises 30 drawers and approxi- in full-face, dorsal view ofthe head.
mately 20,000 specimens, the great majority of ASO Antennal sclerite distance, outer margins:
which were collectedby Janzen. maximumdistancebetweentheouter,lateral
margins ofthe antennal sclerites.
MetricMeasurementsandIndices CLWWidth of median clypeal lobe, measured
between the anterolateral angles (in
All measurements were made under a Wild Pseudomyrmex satanicus and P. spinicola
microscopeatSOXpower,usinganorthogonalpair only; see Figs. 10, 11).
of Nikon micrometers wired to a digital readout. MD4, MD5, MD8, MD9 A series ofmandibular
Measurement conventions follow those described measurements (see Ward 1989, figure 2).
in Ward (1985, 1989). Note that a full-face or MD4: distancealongthebasalmarginofthe
dorsal view of the head involves positioning the mandible from the base to the mesial basal
posterior margin and the anterolateral margins tooth;MD5 lengthofthebasalmargin MD8
: ;
(abovethemandibularinsertions)sothattheyliein distance along the masticatory margin from
the same plane ofview. the apex to the fourth tooth, counting from
the apex; MD9: length of the masticatory
margin.
120 Journalof Hymenoptera Research
EW Eye width: maximum width of compound PL Petiole length: length of the petiole, mea-
eye, measured along its short axis in an suredinlateral viewfromthelateralflanges
oblique dorsolateral view ofthe head. oftheanteriorpeduncletotheposteriormar-
SL Scape length: length of the first antennal gin ofthepetiole (seeWard 1985, figure4).
segment, excluding theradicle. PND Petiolar node distance: distance from the
LFl Lengthoffirstfunicularsegment: maximum lateral flanges of the anterior petiolar pe-
measurable length ofthe firstfunicular seg- duncle to the maximum height ofthe node,
ment (pedicel), including its basal articula- measured from the same view as PL and
tioninworkersandqueensbutexcludingthe along the same line of measurement (see
basal articulation in males (where it is usu- Ward 1985, figure 4).
ally hidden). PH Petioleheight: maximumheight ofthe peti-
LP2 Length of second funicular segment: maxi- ole, measured in lateral view atright angles
mum
measurable length of the second to PL, but excluding the anteroventral pro-
funicular segment. cess.
LP3 Length of third funicular segment: maxi- PPL Postpetiolelength: lengthofthepostpetiole,
mummeasurablelengthofthethirdfunicular measured in lateral view, from the anterior
segment. peduncle (ofthe postpetiole) to the point of
WF2 Width ofsecond funicular segment, contact with the fourth abdominal tergum,
PL Profemur length: length of the profemur, excluding the pretergite (see Ward 1985,
measured along its long axis in posterior figure 4).
view (see Ward 1985, figure 3). DPWDorsalpetiolarwidth:maximumwidthofthe
PW Profemurwidth maximummeasurablewidth petiole, measuredin dorsal view.
:
of the profemur, measured from the same MPWMinimumpetiolarwidth: minimumwidthof
view as PL, at right angles to the line of thepetiole,measuredindorsalview,anterior
measurement ofPL. to DPW.
DPL Diagonallengthofthepropodeum: lengthof PPW Dorsal postpetiolar width: maximum width
the propodeum, measured in lateral view ofthe postpetiole, measure in dorsal view.
along a diagonal line drawn from the LHT Length ofmetatibia: maximum measurable
"metapleural" lobe to the metanotal groove length ofmetatibia, excluding the proximal
(see Ward 1985, figure 2). partofthearticulationwhichisreceivedinto
BF Length of the basal (=dorsal) face of the the distal end of the metafemur (see Ward
propodeum, measured in lateral view from 1989, figure 5).
the metanotal groove to the point on the CI Cephalic index: HW/HL
surface of the propodeum which is maxi- OI Ocular index: EW/EL
mally distant from the diagonal propodeal REL Relative eye length: EL/HL
line. REL2Relative eye length, using HW: EL/HW
DP Length of the declivitous face of the OOI Oculo-ocellarindex: OOD/OD
VW/HW
propodeum, measured in lateral view from VI Vertex width index:
the "metapleural" lobe to the point on the PCI Frontal carinal index: MPC/HW
surface of the propodeum which is maxi- FCI2 Frontalcarinalindex,usingASD:MPC/ASD
mally distant from the diagonal propodeal ASI Antennal sclerite index: ASD/ASO
line. SI Scape index: SL/HW
MP Depthofmetanotalgroove("mesopropodeal SI2 Scape index, using EL: SL/EL
impression"), measuredin lateralviewfrom PLI Funicularlength index: (LP2 + LP3)AVF2
the bottomofthe metanotal groove to aline PI Profemurindex: FW/PL
drawn across the dorsal surface of the PDI Propodeal index: BF/DF
mesonotum and propodeum. MPI Metanotal index: MP/HW
NI Petiole node index: PND/PL
.
Volume 2, Number 1, 1993 121
PLI Petiole length index: PH/PL
PLI2 Petiole length index, using PPL: PPL/PL CladisticAnalysis
PWI Petiole width index: DPW/PL
PWI2 Petiolewidthindex,usingPPW:DPW/PPW A set of 47 characters, representing the most
PWI3 Petiole width index, using MPW: MPW/ discrete orquantifiabledifferences among species
DPW or groups of species in the P. ferrugineus group,
PWI4 Petiole width index, using LHT: DPW/ was used for phylogenetic analysis. Twenty of
LHT these characters were worker-based (11 of these
PPWI Postpetiole width index: PPW/PPL manifestedthe sameconditions inqueens), 8 were
queen-based, and 19 were taken from male mor-
Other Conventions phology, primarily male genitalia. The characters
and character states are as follows:
Other terminology follows the usage in Ward
(1989). Notethatdescriptionsofsurfacesculpture 1. Worker, median clypeal lobe (0) laterally
and integument reflectance apply to observations roundedorsubangulate,(1)laterallywithsharp
madeundersoftlight,withanopaque(Mylar)filter angles orteeth (Figs. 10, 11).
placedbetweenthespecimensandsourceofillumi- 2. Worker and queen, frontal carinae (0) rela-
nation. Palpformulareferstothenumberofmax- tively well separated, median lobes ofanten-
illary palp segments followed by the number of nalscleriteslessexposed(Figs. 12-19,32),(1)
labial palp segments; 5p4,3 indicates a condition closelyadjacentandmedianlobesofantennal
intermediatebetween5,3and4,3,i.e.partialfusion sclerites more exposed (Figs. 10, 11, 32).
ofthe fourth and fifthmaxillary palp segments. 3. Worker and queen, palp formula (0) 5,3, (1)
Listing of synonymy under each species is re- 4,3.
strictedtocitationoftheoriginaldescriptions(with 4. Worker,head(0)broader,relativetoHL,DPL
full reference given for all previously proposed andPL, (1) narrower; see regressions ofHL,
junior synonyms) and new nomenclatural combi- DPL andPL on HW (Figs. 36-38).
nations. Forecologists amoreuseful summary of 5. Worker, scape (0) short, relative to HL, (1)
name usage is offered in Table 1, which indicates longer; regression of SL on HL lying above
thecorrespondencesbetweenthenames appearing that ofother species.
in the biological literature on acacia-ants and the 6. Worker, conspicuous pit-like impression on
currently valid scientific names. The reader will midline ofhead (0) absent, (1) present.
appreciate that there has been considerable 7 Worker,petiole(0)shortrelativetopostpetiole,
misidentification ofthese ants. PLI2 0.77,(1)longerrelativetopostpetiole,
Inthelistsofmaterialexaminedofeachspecies, PLI2 < 0.77.
I have cited only locality and collector ("c.u." 8. Workerandqueen, petiole (0) withoutawell
signifies collectorunknown), with the source col- differentiated anterior peduncle, i.e. weakly
lectionslistedtogetheratthebeginning.Additional constrictedindorsalviewandwithlittleorno
localityinformationissometimesprovidedinsquare inflectionoftheanteriorfaceofthepetiolein
brackets,tofacilitatelocationofthecollectingsite. lateral profile (Figs. 22, 23), (1) with a well
Considerableeffortwasexpendedtodeterminethe differentiatedpeduncle (Figs. 20, 21, 24-29).
coordinates(latitudeandlongitude)ofeachcollect- 9. Worker and queen, petiole, dorsal view,
ingsite,andthiswasthenusedinconjunctionwith angulateposterolateralcomers (0)absent,(1)
the public domain software program Versamap moderatelydeveloped,precededbyconvexor
(version 1.20) to plot the distributions of each sinuatesides(e.g.Figs,20,27),(2)verypromi-
species (Figs. 67-72). nent, preceded by more or less straight sides
(Fig. 24).
10. Worker and queen, petiole (0) shorter and
higher, workerPLI 0.71, queenPLI 0.64,
.. ) .
122 Journalof Hymenoptera Research
(1) more slender, worker PLI < 0.72, queen HL lying above that ofother species.
PLI < 0.64. 27. Queen, metatibia (0) short, relative to HL,
11. Worker, DPWrelativetoHW(0) narrow, (1) LHT/HL < 0.62, (1) longer, LHT/HL > 0.66.
broader, (2) very broad; see regression of 28. Queen,metatibia(0)short,relativetoHW,(1)
DPW on HW (Fig. 41). longer; see regression of LHT on HW (Fig.
HW
12. Worker, plot of PWI3 by lying in (0) 31).
upperleft(1)lowerright(2)lowerleftregion, 29. MHaWle, head (0) narrower, CI 0.82-0.94HaWnd
ofFig. 39. <0.96, (1)broader,CI 0.94and/or
13 Workerandqueen,plotofPWI4byHWlying 0.96 (regressionofHLonHWlyingbelow
in (0) centerandlowerright(1) upperleft(3) thatofother species).
lowerleftregion, ofFig. 40. 30. Male,scapeindex(SI)(0)0.22-0.30,(1)0.29-
HW
14. Worker, postpetiole (0) broad, PPWI > 1.30, 0.35,andregressionofSLon lyingabove
(1) narrow, PPWIl.00-1.30. that ofother species.
15 Worker,metatibia(0)short,relativetoHL,(1 31 Male, scapelength,relativetoELandHL(0)
relativelylong; seeregressionofLHTon HL long, SI2 0.43-0.56, SL/HL 0.22, (1)
(Fig. 30). shorter, SI2 0.33-0.43, SL/HL 0.21 (and
16. Workerandqueen,headsculpture(0)densely regression of SL on HL lying below that of
punctulate, subopaquetosublucid, atleaston other species).
HW
upper third of head, (1) densely punctulate, 32. Male, compound eye length, relative to
opaque, (3) punctulate-coriarious, opaque (0) long, REL2 0.56-0.63, (1) shorter, REL2
HW
(matte). 0.49-0.58, andregressionofELon lying
17. Worker and queen, propodeum, posterolat- below that ofother species.
eral portions (0) sublucid, without overlying 33. Male, petiole (0) less slender, PLI>0.45, (1)
rugulo-punctate sculpture, (1) supopaque to more slender, PLI < 0.50, and regressions of
opaque, with rugulo-punctate sculpture. PH and DPW on LHT lying below those of
18. Worker and queen, petiolar node (0) lacking other species.
conspicuous suberect pubescence, (1) with 34. Male, stemite IX, posterior margin (0) con-
suchpubescence. vex, (I) with a moderate concavity, less than
19. Workerandqueen,standingpilosityonexter- semicircular(Fig. 54), (2) with adeep, semi-
nal faces oftibiae (0) present, (1) absent. circularconcavity (Fig. 55).
20. Worker and queen, head and gaster (0) yel- 35. Male, paramere, lateral view, posterodorsal
low- to orange-brown, (1) reddish-brown to extremity(0) rounded, (1) angulate or ex-
mediumordarkbrown,(2)verydarkbrownto panded.
black (excluding mandibles, clypeus and 36. Male, paramere, lateral view, posterodorsal
scape). extremity (0) notprojecting caudad, (1) pro-
HW
21. Queen,size(0)small, 0.85,(1)medium jecting caudad, as in Figs. 56, 57.
HW
to large, 0.85. 37. Male, paramere, lateral view, posterodorsal
22. Queen,headshape, foragivenLHT(0)elon- extremity (0) not developed as a lobe-like
gate, (1)lesselongate,(2)broad(seeFig. 52). protrusion, whosemesialfaceisasaucer-like
23. Queen, petiole (0) short, relative to HW, PL/ concavity, (1) so developed (Figs. 61-66).
HW < 0.71, (1) longer, PL/HW 0.71. 38. Male, paramere, lateral view, posterodorsal
24. Queen, petiole (0) short, relative to HL, (1) extremity(0)wellseparatedfrommesiodorsal
longer: see regression ofPL on HL (Fig.46). lobe, (1) closetomesiodorsallobe, enclosing
25. Queen, regression ofPH on HW lying in (0) anarrow spacebetween itandthelobe (Figs.
upper(1) middle (2) lowerregion, inFig. 47. 58, 61-66).
26. Queen, petiole, dorsal view (0) narrow, rela- 39. Male, paramere, digitiform mesiodorsal lobe
DPW
tivetoHL,(2)broader;regressionof on (0)absent,(1)present,slender,directedposte-
Volume 2, Number 1, 1993 123
riorlyorposterodorsally(Figs.56-60,63-65), weak ridge or carina, (2) with a central el-
(2)present, stubby,directedmoreorlessdor- evated area, delimited posteroventrally by a
sally (Figs. 61, 62). strong lamellate carina.
40. Male, paramere, mesial faceofposterodorsal 46. Male, aedeagus, external face, afore-men-
extremity (0) simple in form, not expanded tioned carina (if present) (0) well separated
mesially, (1) expanded mesially, partly ob- fromthetoothedposteriormargin,(1)running
scuringthemesialdorsoventralridgeinposte- close to and more or less parallel with the
riorview. toothed posterior margin but separated by a
41. Male, aedeagus, posterior margin (0) entire, deep groove, (2) converging posterodorsally
not medially pointed, (1) toothed, and medi- with the toothedposterior margin.
ally pointed. 47. Male, aedeagus, posteroventral extremity (0)
42. Male, aedeagus, posterior margin (0) bent broadlyrounded,(1)subangulate,(2)angulate
posterolaterally, (1) bent anterolaterally, (2) with atooth-like protrusion.
bentanterolaterallybutwiththemedialpoint
redirected posteriorly.
43. Male, aedeagus,laterallybentportionofpos- ThedatasetwasanalyzedusingFarris'Hennig86
teriormargin (0) continuous with the margin program.Characters 12, 13and16wereconsidered
oftheposterodorsalextremity,(l)discontinu- unordered. AsanoutgroupIchosePseudomyrmex
ous with margin of posterodorsal extremity fervidus (F. Smith), a Central American species
(elevated laterally), the two connected by a whichsharesanumberof(mostlyworker)features
gradual slope, (2) discontinuous with margin incommonwiththeP.ferrugineusgroup: 5,3palp
ofposterodorsalextremity,elevatedlaterally, formula, similarmandibulardentition, well devel-
and separated by a trenchant rise (posterior oped metanotal groove, abundant pilosity on
view) from the posterodorsal extremity. mesosoma dorsum, relatively small eyes, and a
44. Male, aedeagus, plate-like expansion of similarhabitus withrespectto size andcolor. In a
posterodorsal extremity (0) absent, (1) mod- few instances the outgroup species spanned the
erately developed, (2) strongly developed. phenotypic gap between two discrete states in the
45. Male, aedeagus, external face (0) without a ingroup; it was then coded as unknown for that
large, central elevatedarea, (1) with acentral character. Inadditiontoseekingthemostparsimo-
elevatedarea,delimitedposteroventrallybya nious tree for the entire data set (rooted with the
outgroup), I also compared the cladograms based
on three subsets, derived from the worker, queen,
and male characters sets, respectively. For this
second analysis the queen character set included
theeightcharactersassessedonlyinqueens(21-28)
plus those manifested identically in workers and
queens (2, 3, 8-10, 13, 16-20), for a total of 19
characters.
124 Journalof Hymenoptera Research
Figs. 1-9. Pseudomyrmex workers, lateral view ofmesonotum, propodeum and petiole, with pilosity shown in
outline;Figs. 4and5includeafrontalviewofworkerhead. 1,P. boopis(CostaRica); 2,P. ita(CostaRica); 3, P.
kuenckeli (CostaRica); 4, P. hesperius (Mexico, paratype); 5, P. opaciceps (Guatemala, paratype); 6, P. gracilis
(Mexico); 7, P. nigropilosus (CostaRica); 8, P. reconditus (Nicaragua, holotype); 9, P. simulans (Panama).
Volume2, Number 1, 1993 125
mm
1
Figs. 10-19. Pseudomyrmexferrugineus group, workers, full-face dorsal (=frontal) view ofhead with pilosity
showninoutline(exceptonmandibles);Fig. 19includesalateralviewofhead. 10,P. satanicus(Panama); 11,P.
spinicola (CostaRica); 12,P.peperi(Guatemala); 13,P. nigrocinctus(CostaRica); 14,P.particeps(CostaRica,
holotype); 15, P. mixtecus (Mexico, holotype); 16, P.flavicornis (Nicaragua); 17, P. veneficus (Mexico); 18, P.
ferrugineus (Mexico); 19, P.janzeni (Mexico, holotype).
126 Journalof Hymenoptera Research
'V' <rT/"rS^
mm
1
Figs. 20-29. Pseudomyrmexferrugineus group, workers, dorsal view of petiole paired with lateral view of
mesonotum,propodeum,petioleand, inFig. 28,postpetiole. Standingpilosityshown inoutline. 20,P. satanicus;
21, P. spinicola; 22, P. nigrocinctus; 23, P. particeps; 24, P. peperi; 25, P.flavicomis; 26, P. mixtecus; 27, P.
ferrugineus; 28, P. veneficus; 29,P.janzeni. These arethe same individualsillustratedinFigs. 10-19.