BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, researchlibraries, and research funders in the common goal of maximizing access to critical research.

Species Richness of the Campopleginae and Ichneumoninae (Hymenoptera:Ichneumonidae) Along a Latitudinal Gradient in Eastern North American Old-Growth ForestsAuthor(s): Elizabeth Lockard Skillen, John Pickering, and Michael Joseph SharkeySource: Environmental Entomology, 29(3):460-466. 2000.Published By: Entomological Society of AmericaDOI: http://dx.doi.org/10.1603/0046-225X-29.3.460URL: http://www.bioone.org/doi/full/10.1603/0046-225X-29.3.460

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, andenvironmental sciences. BioOne provides a sustainable online platform for over 170 journals and books publishedby nonprofit societies, associations, museums, institutions, and presses.

Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance ofBioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use.

Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiriesor rights and permissions requests should be directed to the individual publisher as copyright holder.

COMMUNITY AND ECOSYSTEM ECOLOGY

Species Richness of the Campopleginae and Ichneumoninae(Hymenoptera: Ichneumonidae) Along a Latitudinal Gradient in

Eastern North American Old-Growth Forests

ELIZABETH LOCKARD SKILLEN, JOHN PICKERING, AND MICHAEL JOSEPH SHARKEY1

Institute of Ecology, University of Georgia, 711 BioSciences Building, Athens, GA 30602

Environ. Entomol. 29(3): 460Ð466 (2000)

ABSTRACT We ran 10 Malaise traps along a latitudinal gradient in Þve old-growth forests: SanFelasco Hammock, FL (308 N, 828 W); Tall Timbers Research Station, FL (318 N, 848 W); HitchitiExperimental Forest, GA (338 N, 848 W); Patuxent Wildlife Research Center, MD (398 N, 778 W),and Shaw Woods, Ontario (468 N, 778 W). Of 18,791 ichneumonid wasps collected, 1,487 camp-oplegines and 4,494 ichneumonines were sorted to 241 species (95 species of Campopleginae and146 of Ichneumoninae). Campoplegine species richness was highest at the Georgia site: San Felasco(20 species), Tall Timbers (32), Hitchiti (49), Patuxent (35), and Shaw Woods (12). Ichneumoninespecies richness was highest at the Maryland site: San Felasco (47), Tall Timbers (30), Hitchiti (44),Patuxent (68), andShawWoods (37). The results conÞrmreported trends that peakNorthAmericanichneumonid species richness lies at midlatitudes. However, they document greater biodiversity inthe southern states and a broader peak than previously reported.

KEY WORDS Ichneumonidae, biodiversity, parasitoids, latitude, old-growth forests

NUMEROUS STUDIES DOCUMENT and attempt to explainwhy ßoral and faunal diversity generally increase asone approaches the equator (Dobzhansky 1950,Stevens 1989, Sime and Brower 1998). Certain taxa ofparasitic Hymenoptera are notable exceptions to thispattern. Owen and Owen (1974) Þrst proposed thatthe trend for the Ichneumonidae runs contrary to thatfor most plant and insect families; they reported morespecies in samples from temperate Europe than intropical Africa. Janzen (1981) analyzed published dis-tribution maps of 1,717 North American ichneumonidspecies and concluded that their peak species richnesslies at midlatitudes between 38 and 428 N. On the eastcoast, for instance, his analysis suggested that speciesrichness from central Virginia to central New York is1.5 times higher than in central Georgia and 2.5 timeshigher than in central Florida. He further predictedfewer species of ichneumonids per host species as oneapproaches the equator, with proportionately fewerspecies of tropical specialists.

More recently,Gauld (1986, 1987, 1991) andGastonand Gauld (1993) examined latitudinal trends for anumber of subfamilies of the Ichneumonidae. Theirresults are equivocal. Some subfamilies appear morediverse in the tropics, others, in temperate areas.Gauld (1986)presentedevidence suggesting thatAus-tralian ichneumonid diversity decreases slightlynearer the equator. In part, this is because of theabsence in northern Australia of taxa such as theCtenopelmatinae, which parasitize sawßies and are

relatively rare in the tropics. Gauld (1987) reportedhigher diversity for the Cryptini, Pimplini, and Ophio-ninae toward the equator and higher diversity of saw-ßy parasitoids away from it. In GauldÕs (1991) mono-graph of the lower pimpliform subfamilies of CostaRica and later analysis by Gaston and Gauld (1993),both local and regional richness of these wasps isgenerally higher in the tropics than elsewhere. Incontrast to the Ichneumonidae, other parasitic fami-lies such as the Chalcididae (Hespenheide 1979) andthe Encyrtidae (Noyes 1989) appear to be more spe-ciose in the tropics.

It is difÞcult to weigh the relative importance offactors that governparasitoid community compositionas a function of latitude. It requires comparable di-versity estimates between locations. Few studies haveemployed the same sampling methods across largegeographic areas to obtain comparable site or regionaldiversity estimates. With speciose taxa, estimates ofspecies richness generally increase with sampling ef-fort (Colwell and Coddington 1994). Well-collectedareas can, therefore, be expected to yield higher es-timates of species richness thanpoorly collected areas.JanzenÕs (1981) Þnding that the peak of North Amer-ican ichneumonid species richness lies at midlatitudesand a similar Þnding by Quicke and Kruft (1995) forthe Braconidae are based on published records overlarge geographic areas.We suspected that their resultsmay have been in part an artifact of the proximity ofmajor collections and may not reßect underlying bi-ological processes. The current study was designed toovercome this potential artifact. We examine speciesdiversity of two ichneumonid subfamilies along a lat-

1 Department of Entomology, University of Kentucky, S-2227 Ag-ricultural Sciences Building-North, Lexington, KY 40546Ð0091.

0046-225X/00/0460Ð0466$02.00/0 q 2000 Entomological Society of America

itudinal gradient in eastern North America using Mal-aise traps as a standard sampling technique at Þve sitesfrom Ontario to Florida.

Materials and Methods

Study Organisms. The Ichneumonidae, consistingof 35 subfamilies (Wahl 1993) and an estimated 60,000species, is thought tobe themost speciose family in theworld, with the possible exception of the Curculion-idae (Townes 1969) or the Staphylinidae. We exam-ined two subfamilies that represent the two major lifehistory strategieswithin the Ichneumonidae, idiobiontand koinobiont parasitism (Askew and Shaw 1986).The Campopleginae are an abundant group of koino-biont endoparasitoids; hosts are mostly Lepidopteraand Symphyta (Wahl 1993). The Ichneumoninae at-tack Lepidoptera exclusively (Gauld 1988). Most areidiobiont endoparasitoids that attack pupal stages,whereas others are koinobiont endoparasitoids thatattack Þnal-instar larvae (Wahl 1993). The Ichneu-moninae are the most speciose group of idiobiontendoparasitoids within the Ichneumonidae.

Sites and Sampling. We collected samples in 1993with Þne mesh (0.33 mm) Townes-style Malaise traps(Townes 1972) purchased from Sante Traps (Lexing-ton, KY). We sampled Ontario and Maryland in 1992as well. We ran two traps at each site and used 70%ethanol as a preservative. The traps were aligned on anorthÐsouth axis, with the collecting head toward thesouthernend.Collaborators servicedeach trapweeklythroughout the period of ßight activity, except late orearly in the season when they collected some samplesbiweekly or monthly.

We sampled in Þve old-growth forests from 458 420N to 298 440 N. We collected samples at Shaw Woods,Renfrew County, Ontario (458 420 N, 778 40 W) from6 May to 22 October 1992 and 6 May to 21 October1993. This siteÕs trees are dominated by Fagus grandi-folia Ehrhart and Acer saccharum Marshall (Dugaul1980, Lockard 1995). We collected samples at Patux-entWildlifeResearchCenter,PrinceGeorgesCounty,MD (398 030 N, 768 470 W) from 6 April to 9 November1992 and 12 April to 25 October 1993. Hotchkiss andStewart (1947) described this site,which is dominatedby F. grandifolia. We collected samples at HitchitiExperimental Forest, Jones County, in GeorgiaÕs Pied-mont (338 030 N, 838 430 W) from 23 March to 21December 1993. Lockard, (1995, Appendix 2) lists thissiteÕs tree species, dominated by Acer rubra L., Quer-qus albaL., andLirodendron tulipiferaL.Wecollectedsamples in the Woodyard Hammock at Tall TimbersResearch Station, Leon County, in FloridaÕs coastalplain (308 390 N, 848 150 W) from 30 March to 14December 1993. Because we missed the Þrst part ofthis ßight season, we also include a sample from 4 to28March 1994.Hirsch (1981)described this site; dom-inant tree species include F. grandifolia and Magnoliagrandiflora L. We collected samples at San FelascoHammock State Preserve, Alachua County, FL (298440 N, 828 270 W) from15March to 18December 1993.

Dunn (1982) provides details of this climax mesichammock in the Coastal Plain.

Specimen Handling and Tabulation. All ichneu-monids were sorted to subfamily and all ichneumoni-nes and campoplegines were sorted to morphospecieswith the assistance of D. B. Wahl. We identiÞed theIchneumoninae and the Campopleginae genera Du-sona andCasinaria to specieswheneverpossible. Iden-tiÞcations were based on Heinrich (1960a, 1960b,1961a, 1961b, 1961c, 1962a, 1962b, 1977),Walley (1940and 1947), and the resources of the American Ento-mological Institute (AEI) in Gainesville, FL. Sexualdimorphism in the Ichneumoninaemade it impracticalto associate males and females in some genera. Thus,within these sexually dimorphic genera where bothmales and females were present at a site, we countedonly the number of species in the most speciose sex.We deposited specimens at the University of Georgiaand the AEI. See Wayman (1994) for details of sortingtechniques.

Statistical Estimators. Malaise traps are generallyrecognized as efÞcient collecting devices of ichneu-monoids (Matthews and Matthews 1970, Owen et al.1981,Darling andPacker 1988, andNoyes 1989).Com-bining both sites in Florida, for example, we collected53 ichneumonines in one season, nearly three-fourthsof the known species in Florida. Nevertheless, it isunlikely that we collected all the species at any site,and hence, we must consider possible site differencesin the proportion of fauna collected. For example, wemay have sampled a lower proportion of the speciesfrom sites with high diversity. Thus, in comparingspecies richness among sites, we might underestimaterich taxa and sites relative to depauperate ones. Toexamine such potential bias across sites, we used sta-tistical estimates of species richness and data fromadditional years.

To estimate species richness, we used theEstimateSstatistical package version 5 (Colwell 1997) and used100 randomizations for the following nine estimators:abundance-based coverage (SACE), incidence-basedcoverage (SICE), Chao 1 (SC1), Chao 2 (SC2), Þrst-order Jackknife (SJ1), second-order Jackknife (SJ2),bootstrap (SB), MichaelisÐMenten averaged over ran-domizations (SMR), and MichaelisÐMenten computedonce for mean species accumulation curve (SMM). Toour knowledge, no one has rated estimator perfor-mance using Malaise trap samples. Thus, we used allnineestimators available in theEstimateS softwarebutonly present results for the maximum, minimum, andmedian estimator. For references and review of thesestatistical estimators, see Colwell (1997), Colwell andCoddington (1994), and Chazdon et al. (1998).

As with all collecting methods, there are inherentbiases in using onemethodover another. For instance,two Malaise traps in the same 1-ha plot may collectsigniÞcantly different material, making between sitecomparisons difÞcult. The observed data are pre-sented by trap and by total to show the between trapdifferences. To address the difference in sample sizebetween sites, we generated Coleman richness expec-tation curves (Coleman 1981, Coleman et al. 1982)

June 2000 SKILLEN ET AL.: SPECIES RICHNESS OF THE CAMPOPLEGINAE AND ICHNEUMONINAE 461

with 95% conÞdence interval in EstimateS. Thesecurves (Figs. 1Ð3) plot the expected richness for ran-domized subsamples of the entire data set providing amethod to evaluate the degree of sample heteroge-neity (Colwell and Coddington 1994).

Results and Discussion

Observed Species Richness. We sorted a total of18,791 Ichneumonidae from the Þve study sites, in-cluding 4,494 ichneumonines in 37 genera and 1,487campoplegines in 18 genera. Across the Þve sites, wetabulated a minimum of 146 ichneumonine and 95campoplegine species. We caught more species at themidlatitude sites (Table 1). The Georgia site had thehighest observed campoplegine species richness (49species), and the Maryland site had the highest ob-served ichneumonine species richness (68 species).The Maryland site had the highest observed speciesrichness for the two subfamilies combined (103 spe-cies), with the Georgia site second (93 species).

Estimates of Species Richness. To estimate speciesrichness at the sites,wemodeled species accumulationas a function of trapping effort (Table 2). Some ofthese species accumulation curves did not reach anasymptote(not shown), indicating thatmore sampling

is needed to estimate accurately total species richness.The species accumulation curves did not lie below theColeman (or rarefaction) curves, indicating low sam-pleheterogeneity. For campoplegine species richness,theobserveddata (Table 1;Georgia highest,Marylandsecond, Tall Timbers [Florida] third, San Felasco[Florida] fourth, and Ontario lowest) generally cor-respond with the median rank orders of the estimates(Table 2). The inversion of San Felasco (Florida) andOntario is the exception, possibly because of the highpercent (53%)of singletons atOntario, a consequenceof few individuals being caught. Regarding the Ich-neumoninae, the rank order of the median estimatorand the observed data are the same. Maryland isranked Þrst, San Felasco (Florida) second, Georgiathird, Ontario fourth, and Tall Timers (Florida) last.Thus, the observed data and the estimators generallysupport the conclusion that species richness in thesetwo subfamilies is greater at midlatitudes. However,our results suggest relatively higher ichneumonidrichness at southern sites, particularly inGeorgia, thanreported by Janzen (1981).

The strength of this conclusion assumes that we cantrust the estimators. Some evidence suggests other-wise. Because some estimators did not reach an as-ymptote, suggesting that two trap-years may be insuf-

Table 1. Campopleginae and Ichneumoninae species (and individuals) collected

Site(latitude)

YearCampopleginae Ichneumoninae

Trap 1 Trap 2 Total Trap 1 Trap 2 Total

Shaw Woods, Ontario 1992 7 (11) 5 (9) 9 (20) 24 (46) 21 (50) 29 (96)(458 420 N) 1993 6 (9) 8 (14) 12 (23) 24 (64) 22 (53) 37 (117)

199211993 10 (20) 11 (23) 16 (43) 38 (110) 34 (103) 50 (213)Patuxent, Maryland 1992 21 (73) 14 (83) 23 (156) 44 (506) 45 (294) 59 (800)

(398 030 N) 1993 31 (168) 20 (144) 35 (312) 45 (422) 60 (308) 68 (730)199211993 37 (241) 23 (227) 40 (468) 64 (928) 73 (602) 89 (1530)

Hitchiti, Georgia(338 030 N) 1993 35 (252) 41 (320) 49 (572) 30 (320) 38 (373) 44 (693)

Tall Timbers, Florida(308 390 N) 1993a 24 (76) 25 (137) 32 (213) 21 (282) 24 (400) 30 (682)

San Felasco, Florida(298 440 N) 1993 15 (47) 16 (33) 20 (84)b 25 (243) 37 (407) 47 (663)b

a To standardize trapping duration, data from Tall Timbers include 3 wk in 1994.b Includes 4 Cratichneumon, 2 Campoplex, 1 Casinaria, 1 Hyposoter, 1 Limonthe, and 8 Barichneumon specimens for which trap numbers are

unknown.

Table 2. Median (SMed), maximum (SMax) and minimum (SMin) estimates of campoplegine and ichneumonine species richness andpercent singleton’s (PS) collected

Site YearCampopleginae Ichneumoninae

SMed SMax SMin SMed SMax SMin PS, %

Shaw Woods, Ontario 1992 16 24MR 11B 41 54MR 34B 451993 40 58ICE 16B 57 135MR 46B 53

199211993 43 77C1,2 20B 71 467MR 59B 47Patuxent, Maryland 1992 36 40J2 28B 89 97J2 67MM 36

1993 45 52J2 40MR 106 114J2 79MM 37199211993 47 53J2 43MM 131 148J2 94MM 34

Hitchiti, Georgia 1993 67 77J2 57B 60 87C2 48MM 37Tall Timbers, Florida 1993 42 52J2 36MM 45 66C2 31MM 34San Felasco, Florida 1993 27 30J2 23B 86 119C2 51MM 45

Estimates are shown as some value (Sx), where S is the estimate of species richness and x is the method used to calculate species richnessby the incidence-basedcoverage (SICE),Chao1(SC1),Chao2(SC2), Second-order jackknife (SJ2), bootstrap (SB),MichaelisÐMentenaveragedover randomizations (SMR) and MichaelisÐMenten computed once for mean species accumulation curve (SMM).

462 ENVIRONMENTAL ENTOMOLOGY Vol. 29, no. 3

Þcient toestimate accurately total species richness,wepresent Coleman curves as an additional means ofcomparing diversity across sites (Figs. 1Ð3). For theIchneumoninae, Fig. 1 shows that the Maryland sitehas higher species richness than the three southernsites. This ÞgureÕs Coleman curve for Ontario shows asmall sample size and steep slope, indicating that On-tariomayhavehigher ichneumonine richness than thecurrent studyÕs observed value and estimators suggest.For the Campopleginae, the Coleman curves in Fig. 2strongly suggest that species richness is highest inGeorgia. The data in Fig. 3 strengthen this Þnding. Itshows that two trap-years of effort in Georgia (1993)yielded both a steeper Coleman curve and higherabsolute species richness than four trap-yearsof effort,at the second richest site, Maryland (1992 and 1993).

Furthermore, Gaasch (1996) found 117 (N 5 3395)campoplegine and 100 (N 5 3106) ichneumonine spe-cies in 1993 and 1994 from a variety of habitats adja-cent to our old-growth site in Georgia. These resultsshow that the diversity in the neighboring landscapeof ourGeorgia site exceedsmaximumestimates for theCampopleginae (SJ2 5 77) and Ichneumoninae(SC2 5 87) (Table 2). Thus, unless the additionalspecies found by Gaasch never enter old-growth, it islikely that all the statistical methods that we usedunderestimate true species richness.Despite such lim-itations, until furtherwork tests their validityorprovesotherwise, we feel that the estimators as a groupstrengthen our conclusions.

Peak Ichneumonid Diversity. Table 3 shows thecomparison of our results to those in JanzenÕs (1981)report on eight ichneumonid subfamilies. Our analysisfor the two subfamilies generally supports his Þndingof peak parasitoid richness at midlatitudes. However,Janzen reported a 41% drop in species richness fromhis maximum, between latitudes 38 and 428 N, whichincludes our Maryland site, to the latitudes 32.5 and34.98N,which includesourGeorgia site.Henoted thatthe sharpest decline in species richness occurred at'37.48 N, with virtually no change of land mass in this

Fig. 1. Coleman richness expectation curves with 95%conÞdence intervals of Ichneumoninae at Shaw Woods, On-tario; Patuxent, Maryland; Hitchiti, Georgia; Tall Timbers,Florida; and San Felasco, Florida, in 1993. Sampling effort isthe number of samples that contain at least one ichneumo-nine.

Fig. 2. Coleman richness expectation curves with 95%conÞdence intervals of Campopleginae at Shaw Woods, On-tario; Patuxent, Maryland; Hitchiti, Georgia; Tall Timbers,Florida; and San Felasco, Florida, in 1993. Sampling effort isthe number of samples that contain at least one camp-oplegine.

Fig. 3. Coleman richness expectation curves with 95%conÞdence intervals of Campopleginae at Hitchiti, Georgia,in 1993 and Patuxent, Maryland, in 1992 and 1993. Samplingeffort is the number of samples that contain at least onecampoplegine.

Table 3. Latitudinal trends in ichneumonid species richness

Location LatitudeCurrent study Janzen (1981)

SOBS POBS SJAN PJAN

Ontario 458 420 N 49 0.48 992 0.95Maryland 398 030 N 103 1 1039 1Georgia 338 030 N 93 0.90 612 0.59Florida 308 390 N 62 0.60 340 0.33Florida 298 440 N 67 0.65 197 0.19

SOBS is the number of species collected in the current study in 1993.POBS expresses site species richness as a proportion of that at Mary-land, the site with the highest observed species richness, and isSOBS/103. Correspondingly, SJAN is the total number of species re-ported by Janzen (1981) for each latitudinal band that contains oneof the current study sites, and PJAN is SJAN/1039.

June 2000 SKILLEN ET AL.: SPECIES RICHNESS OF THE CAMPOPLEGINAE AND ICHNEUMONINAE 463

latitudinal band. In contrast, our data show only an11% decline between Maryland and Georgia. Our re-sults show a wider peak than reported by Janzen,extending south to include the Georgia Piedmont. Wepresume that insufÞcient collecting in southern states,particularly in the northern regions of Georgia, con-tributed to Janzenunderestimating species richness insouthern latitudes relative to better collected north-ern ones. Heinrich (1977) stated that “. . . no attempthas been made to explore the [Ichneumoninae] faunaof these areas [Georgia, Alabama, Tennessee, and Ar-kansas] exhaustively.”

The difference between JanzenÕs (1981) Þndingsand the current study may be biological, stemmingfrom differences in the subfamilies used in the twostudies or because our eastern old-growth forest sitesdo not adequately represent the species richness inJanzenÕs corresponding latitudinal bands. Alterna-tively, the differencesmay result frommethodologicalproblems. The latitudinal bands of highest ichneu-moniddiversity reportedby Janzen include theSmith-sonian Institution and a disproportionate number ofother major North American museums. We collectedless than half the number of species in Ontario than inMaryland; Janzen (1981), for the equivalent latitudi-nal bands, reported only a 5% drop in species richness(Table 3). One explanation for the difference in theseÞndings is that collecting around the Canadian Na-tional Museum, which is located in Ontario, has un-covered a higher proportion of a relatively depauper-ate fauna than has occurred elsewhere.

We caution that geographic analyses based on col-lections and published reports that do not rely onstandard methods or equal sampling effort will under-estimate biodiversity in under-collected areas. FunkandMorin(2000)document thedisparity incollectingeffort for plants by surveying the size and activity ofsoutheastern herbaria. For example, in terms of plantcollections per square mile, they report that Georgiaherbaria contain only 5.66 specimens, under half theaverage for the United States (minus Alaska) of 13.21specimens, and considerably under the 33.15 speci-mens held within the North Carolina herbaria. Simi-larly, depauperate ichneumonid collections fromsouthern states may explain why Janzen reported analmost twofold drop in species richness between thelatitudinal bands corresponding to our Maryland andGeorgia sites.

Why Peak in Midlatitudes? We report a peak ofichneumonid species richness at midlatitudes, withlowerdiversity inFloridaandCanada.Whythisoccursremains unclear. The processes that determine therelationship between parasitoid species richness andlatitude are poorly understood (Hawkins et al. 1992).It has been suggested that resource fragmentation(Janzen and Pond 1975, Janzen 1981, Askew and Shaw1986, Hawkins 1990), plant-host-parasitoid interac-tions (Hawkins and Lawton 1987, Gauld et al. 1992,Sime and Brower 1998), competition (Hawkins 1990),predation (Rathcke andPrice 1976), andenvironmen-tal factors (Townes 1972, Hawkins 1990, Gauld 1991,Hawkins et al. 1992) inßuence parasitoid diversity.

Alternatively, this peak at midlatitudes may also resultfrom the mid-domain effect, a geometric theory usedto explain species diversity gradients (Colwell andLees 2000). This model predicts maximum speciesdiversity at the midpoint of large bounded geographicareas relative to boundary constraints and range cen-ters. In thecurrent study,under themid-domaineffectone would expect species ranges to overlap more inMaryland and Georgia and less in Florida and Ontariowhere species ranges are constrained by oceans to thesouth and Arctic climate to the north.

The lower species richness found at our Floridiansites may be explained by climate. Shapiro and Pick-ering (2000) report higher ichneumoninae activity ina tropical wet than in a tropical moist forest and sug-gest that ichneumonines may be restricted by hot, dryconditions. Lockard (1995) reports that parasitoid ac-tivity at each site peakedbefore summer temperaturesaverage above 278C; although such peaks may havebeen driven by host availability (Gaasch et al. 1998).Heinrich (1977, p. 3) suggests that “the Ichneumoni-nae have proliferated in speciation only in moderateand cool climates.”

Similarly, the lower diversity observed in Ontariomay be explained by climate. CanadaÕs harsh wintersand short ßight season may be adequate enough toexplain the low diversity measured at Shaw Woods.Furthermore,Ontariomay still be recovering from thelast glaciation and still be in the process of accumu-lating species.

Our data are generally consistent with JanzenÕs(1981) Þnding that maximum ichneumonid speciesrichness occurs at midlatitudes in North America.However, we conclude that peak diversity covers awider area than Janzen reported. Rather than a rela-tively narrow peak centered around 38Ð428 N, ourdata suggest that high ichneumonid species richnessoccurs in a broad swath from at least Maryland toGeorgiaÕs Piedmont, dropping in the south near theCoastal Plain and in the north near Canada. We feelthat the almost twofold drop in species richness from37.5 to 32.58 N (Janzen 1981) is possibly the result ofunder-collecting in the southeast relative to extensivecollecting in proximity to large, northern museums.Alternatively, it may be because Janzen consideredeight different subfamilies than the two considered inthe current work. Nevertheless, we caution againstusing museum data to examine geographic trends inbiodiversity without correcting for differences in col-lecting effort.

Acknowledgments

We thank C. Gaasch, L. Marchman, and D. Wahl for theirassistance in specimen identiÞcation; S. Rathbun and T.Fukasawa for statistical consultation; and N. Ayoub, R. Bar-tlett, M. Hunter, F. Messina, B. Shapiro, D. Wahl, I. Gauld,and an anonymous referee for review of the manuscript. Weare grateful to undergraduate laboratory assistants especiallyJ. BufÞngton, S. Councelman, P. Crumrine, J. Imperatori, N.Pournaras, and P. Smotherman. We thank H. Shaw for use ofTheShawWoods.For logistical support atÞeld siteswe thank

464 ENVIRONMENTAL ENTOMOLOGY Vol. 29, no. 3

M. Perry and L. Atkinson at Patuxent Wildlife ResearchCenter; The U.S. Forest Service at Hitchiti ExperimentalForest; D. Wahl and V. Doig at San Felasco Hammock; T.Engstromm, L. Brennan, W. Flowers, I. Askevold, and M.Lara at Tall Timbers. This research was supported by theUSDA-Forest Service (supplement 183 to contract 12Ð11-008Ð876 to J.P, D. Bramlett, J. Hanula, and K. Franzreb), theNational Science Foundation (DEB-9522681 andDEB-9642121 to J.P.) and Agriculture Canada (Agricultureexperiment No. 99Ð08-103).

References Cited

Askew,R.R., andM.R. Shaw. 1986. Parasitoid communities:their size, structure and development, pp. 225Ð259. In J.Waage and D. Greathead [eds.], Insect parasitoids. Ac-ademic, London.

Chazdon, R. L., R. K. Colwell, J. S. Denslow, and M. R.Guariguata. 1998. Statisticalmethods for estimating spe-cies richness of woody regeneration in primary and sec-ondary rain forests of NE Costa Rica, pp. 285Ð309. In F.Dallmeier and J. A. Comiskey [eds.], Forest biodiversityresearch, monitoring and modeling: Conceptual back-ground and Old World case studies. Parthenon Publish-ing, Paris.

Coleman, B. D. 1981. On random placement and species-area relations. Math. Biosci. 54: 191Ð215.

Coleman, B. D., M. A. Mares, M. R. Willig, and Y. H. Hsieh.1982. Randomness, area, and species richness. Ecology63: 1121Ð1133.

Colwell, R. K. 1997. EstimateS: statistical estimation of spe-cies richness and shared species from samples, version 5.UserÕs guide and application (http://viceroy.eeb.u-conn.edu/estimates).

Colwell, R. K., and J. A. Coddington. 1994. Estimating ter-restrial biodiversity through extrapolation. Philos. Trans.R. Soc. Lond. 345: 101Ð118.

Colwell, R. K., and D. C. Lees. 2000. The mid-domain ef-fect: geometric constraints on the geography of speciesrichness. TREE 15: 70Ð76.

Darling,D.C., andL.Packer. 1988. Effectiveness ofMalaisetraps in collection of Hymenoptera: the inßuence of trapdesign, mesh size and location. Canadian Entomol. 120:787Ð796.

Dobzhansky, T. 1950. Evolution in the tropics. Am. Sci. 38:209Ð221.

Dugaul, A. 1980. Shaw Woods Nature Preserve. TrailLandsc. 14: 46Ð56.

Dunn, J. W. 1982. Plant Communities and vascular ßora ofSan Felasco Hammock, Alachua County, Florida. M.S.thesis, University of Florida, Gainesville.

Funk, V. A., and N. Morin. 2000. A survey of the herbaria ofthe south eastern United States. SIDA (in press).

Gaasch, C. M. 1996. Flight phenology and species distribu-tion of parasitic wasps in a heterogeneous landscape inGeorgiaÕs Piedmont, with special reference to the Ich-neumoninae and Campopleginae (Hymenoptera: Ich-neumonidae). M.S. thesis, University of Georgia, Athens.

Gaasch, C. M., J. Pickering, and C. T. Moore. 1998. Flightphenology of parasitic wasps (Hymenoptera: Ichneu-monidae) in GeorgiaÕs Piedmont. Environ. Entomol. 27:606Ð614.

Gauld, I.D. 1986. Latitudinal gradients in ichneumonid spe-cies-richness in Australia. Ecol. Entomol. 11: 155Ð161.

Gauld, I. D. 1987. Some factors affecting the composition oftropical ichneumonid faunas. Biol. J. Linn. Soc. 30: 299Ð312.

Gauld, I. D. 1988. Evolutionary patterns of host utilizationby ichneumonid parasitoids. Biol. J. Linn. Soc. 35: 351Ð377.

Gauld, I. D. 1991. The Ichneumonidae of Costa Rica, 1.Mem. Am. Entomol. Inst. 47: 1Ð589.

Gaston, K. J., and I. D. Gauld. 1993. How many species ofpimplines (Hymenoptera: Ichneumonidae) are there inCosta Rica? J. Trop. Ecol 9: 491Ð499.

Gauld, I. D., K. J. Gaston, and D. H. Janzen. 1992. Plantallelochemicals, tritrophic interactions, and the anoma-lous diversity of tropical parasitoids: the ÔnastyÕ host hy-pothesis. Oikos 65: 353Ð357.

Hawkins, B. A. 1990. Global patterns of parasitoid assem-blage size. J. Anim. Ecol. 59: 57Ð72.

Hawkins, B. A., and J. H. Lawton. 1987. Species richness forparasitoids of British phytophagous insects. Nature(Lond.) 326: 788Ð790.

Hawkins,B.A.,M.R.Shaw, andR.R.Askew. 1992. Relationsamong assemblage size, host specialization, and climaticvariability in North American parasitoid communities.Am. Nat. 139: 58Ð79.

Heinrich,G.H. 1960a. SynopsisofNearctic IchneumoninaeStenopneusticae with particular reference to the north-eastern region (Hymenoptera), part I. Can. Entomol. 15:3Ð87 (suppl.).

Heinrich, G. H. 1960b. Synopsis of Nearctic Ichneumoni-nae Stenopneusticae with particular reference to thenortheastern region (Hymenoptera), part II. Can. Ento-mol. 18: 91Ð205.

Heinrich,G.H. 1961a. SynopsisofNearctic IchneumoninaeStenopneusticae with particular reference to the north-eastern region (Hymenoptera), part III. Can. Entomol.21: 209Ð368.

Heinrich, G. H. 1961b. Synopsis of Nearctic Ichneumoni-nae Stenopneusticae with particular reference to thenortheastern region (Hymenoptera), part IV. Can. En-tomol. 23: 371Ð505.

Heinrich,G.H. 1961c. SynopsisofNearctic IchneumoninaeStenopneusticae with particular reference to the north-eastern region (Hymenoptera), part V. Can. Entomol. 26:509Ð671.

Heinrich,G.H. 1962a. SynopsisofNearctic IchneumoninaeStenopneusticae with particular reference to the north-eastern region (Hymenoptera), part VI. Can. Entomol.27: 677Ð802.

Heinrich, G. H. 1962b. Synopsis of Nearctic Ichneumoni-nae Stenopneusticae with particular reference to thenortheastern region (Hymenoptera), part VII. Can. En-tomol. 29: 805Ð886.

Heinrich, G. H. 1977. Ichneumoninae of Florida and neigh-boring states (Hymenoptera: Ichneumonidae, subfamilyIchneumoninae). Arthropods Fla. Neighb. Land Areas 9:1Ð350.

Hespenheide, H. A. 1979. Are there fewer parasitoids in thetropics? Am. Nat. 113: 766Ð769.

Hirsch, D. W. 1981. Physiognomy and spatial patterns of abeech-magnolia hammock in North Central Florida. M.S.thesis, Florida State University, Tallahassee.

Hotchkiss, N., and R. E. Stewart. 1947. Vegetation of thePatuxent Research Refuge, Maryland. Am. Midl. Nat. 38:1Ð75.

Janzen, D. H. 1981. The peak in North American ichneu-monid species richness lies between 388 and 428 N. Ecol-ogy 62: 532Ð537.

Janzen, D. H., and C. M. Pond. 1975. A comparison, bysweep sampling of the arthropod fauna of secondaryvegetation inMichigan,EnglandandCostaRica.Trans.R.Entomol. Soc. Lond. 127: 33Ð50.

June 2000 SKILLEN ET AL.: SPECIES RICHNESS OF THE CAMPOPLEGINAE AND ICHNEUMONINAE 465

Lockard, E. I. 1995. Biodiversity and geographic distribu-tions of parasitic Hymenoptera (Ichneumonidae: Camp-opleginae and Ichneumoninae) along a latitudinal gradi-ent in eastern North America. M.S. thesis, University ofGeorgia, Athens.

Matthews, R. W., and J. R. Matthews. 1970. Malaise trapstudies of ßying insects in a New York mesic forest. I.Ordinal composition and seasonal abundance. J. N.Y. En-tomol. Soc. 78: 52Ð59.

Noyes, J. S. 1989. A study of Þve methods of sampling Hy-menoptera (Insecta) in a tropical rainforest, with specialreference to the Parasitica. J. Nat. Hist. 23: 285Ð298.

Owen, D. F., and J. Owen. 1974. Species diversity in tem-perate and tropical Ichneumonidae. Nature (Lond.) 249:583Ð584.

Owen, J., H. Townes, and M. Townes. 1981. Species diver-sity of Ichneumonidae and Serphidae (Hymenoptera) inan English suburban garden. Biol. J. Linn. Soc. 16: 315Ð336.

Quicke, D.L.J. and R. A. Kruft. 1995. Latitudinal gradientsin North American Braconid wasp species richness andbiology. J. Hymen. Res. 4: 194Ð203.

Rathcke, B. J., andP.W.Price. 1976. Anomalous diversity oftropical ichneumonid parasitoids: a predationhypothesis.Am. Nat. 110: 889Ð902.

Shapiro, B. A. and J. Pickering. 2000. Rainfall and parasiticwasp (Hymenoptera: Ichneumonoidea) activity in suc-cessional forest stages at Barro Colorado Nature Monu-

ment, Panama, and La Selva Biological Station, CostaRica. Agric. For. Entomol. 2: 1Ð9.

Sime, K. R., and A.V.Z. Brower. 1998. Explaining the lati-tudinal gradient anomaly in ichneumonid species rich-ness: evidence frombutterßies. J. Anim.Ecol. 67: 387Ð399.

Stevens,G.C. 1989. The latitudinal gradient in geographicalrange: how so many species coexist in the tropics. Am.Nat. 133: 240Ð256.

Townes, H. 1969. The genera of Ichneumonidae, part 1.Mem. Am. Entomol. Inst. 11: 1Ð300.

Townes, H. 1972. A lightweight Malaise trap. Entomol.News 18: 239Ð247.

Wahl, D. B. 1993. Family Ichneumonidae, pp. 395Ð448. InH. Goulet and J. T. Huber [eds.], Hymenoptera of theWorld: an identiÞcation guide to families. Canada Com-munications Group, Ottawa.

Walley, G. S. 1940. Ichneumon-ßies of the genus Camp-oplegidea. Sci. Agric. 20: 654Ð661.

Walley, G. S. 1947. Ichneumon-ßies of the genus Casinaria.Sci. Agric. 27: 368Ð370.

Wayman, L. D. 1994. Spatial distribution and sex ratios ofparasitic Hymenoptera (Ichneumonidae: Campoplegi-nae and Ichneumoninae: Braconidae: Aphidius ervi) in adisturbed Georgia Piedmont landscape. M.S. thesis, Uni-versity of Georgia, Athens.

Received for publication 30 June 1999; accepted 8 February2000.

466 ENVIRONMENTAL ENTOMOLOGY Vol. 29, no. 3