HOUSEHOLD AND STRUCTURAL INSECTS Identiﬁcation of Reticulitermes spp. (Isoptera: Reticulitermatidae) from South Central United States by PCR-RFLP ALLEN L. SZALANSKI, JAMES W. AUSTIN, AND CARRIE B. OWENS Department of Entomology, University of Arkansas, Fayetteville, AR 72701 J. Econ. Entomol. 96(5): 1514Ð1519 (2003) ABSTRACT Because of morphological ambiguity, traditional identiÞcation of Reticulitermes Holmgren termites has always been difÞcult and unreliable. A molecular diagnostic method is presented for differentiating Reticulitermes species occurring in the south central United States, which are economically important urban pests. A 379-bp region of the mtDNA COII gene and a 415-bp region of the mtDNA 16S rRNA gene were ampliÞed using polymerase chain reaction (PCR) and sequenced from Reticulitermes ﬂavipes (Kollar), Reticulitermes virginicus (Banks), Reticulitermes hageni Banks, and Reticulitermes tibialis (Banks). Applying DNA sequence data, the PCR-restriction fragment length polymorphism (PCR-RFLP) analysis of two restriction enzymes each for the COII amplicon and the 16S amplicon, were diagnostic for all of the Reticulitermes species analyzed. Based on putative mutation rates, 87% and 97% of the samples should be successfully identiÞed to species with PCR-RFLP of COII and 16S, respectively. To verify the accuracy of our predictions, we examined unclassiÞed Reticultermes populations from Arkansas, Louisiana, Missouri, Oklahoma, Texas, and Virginia using PCR-RFLP. Applying PCR-RFLP, 97 samples were correctly classiÞed to species. This technique allows the use of Þeld-collected specimens preserved in alcohol and can identify termite specimens regardless of caste. PCR-RFLP, resolved with agarose or polyacrylamide gel electrophoresis, provided an efÞcient method for identiÞcation of Reticulitermes species from the south central United States for diagnostic purposes. KEY WORDS PCR-restriction fragment-length polymorphism, molecular diagnostics, Reticuli- termes, termites, mitochondrial DNA SPECIES OF THE GENERA Reticulitermes Holmgren are the economically important species (Su and Scheffrahn major termite pests infesting wooden structures in the 1990). Similarly, in Europe and other parts of the United States and in other countries around the world. world, Reticulitermes spp. are the most expensive and Subterranean termite species, such as those found in damaging pest species. The cost of treatment against the genera Reticulitermes, live in moist soil habitats, termites in Europe is expected to top one billion euros feeding on wood buried beneath or in contact with the by 2005 (UNEP and FAO 2000). In the south central soil. It has been estimated that more than $1.5 billion region of the United States, four species of Reticuli- is spent annually for termite control in the United termes occur: R. ﬂavipes, R. virginicus, R. tibialis, and States, of which 80% is spent to control subterranean R. hageni (National Termite Survey 2002, Weesner termites (Su 1993). More recent estimates by the Na- 1965). tional Pest Management Association suggest this Þg- Correct identiÞcation is critical for pest insects, ure to be closer to $2.5 billion (NPMA 2003). A break- such as termites, that may require very different con- down by costs caused by termite species, reveal that trol methods depending on the target species. Iden- the Þve principal subterranean termite species in the tifying workers is nearly impossible and separating United States are Reticulitermes ﬂavipes (Kollar); Re- soldiers is especially difÞcult given that precise mea- ticulitermes virginicus (Banks), the dark southern sub- surements are required and overlap may occur be- terranean termite; Reticulitermes hesperus (Banks), tween species (Scheffrahn and Su 1994). DifÞculties the western subterranean termite; Reticulitermes tib- arise in species determination at individual collection ialis (Banks), the arid subterranean termite; and Cop- totermes formansanus (Shiraki). Ninety percent of the termite control business involves these Þve subterra- nean termite species (Forschler and Lewis 1997). Sub- terranean termites of the family Rhinotermitidae, ge- nus Reticulitermes, cause the most structural damage in the United States, and are considered the most species determination. Molecular genetic methods are mination of species from a given population of able to differentiate species regardless of the caste subterranean termites. This is the Þrst application of a encountered or locality obtained. molecular genetics technique for the identiÞcation of Previous research on Reticulitermes genetics has Reticulitermes termite taxa from the south central focused on genetic variation within and among pop- United States. ulations, or phylogenetic relationships within Reticu- litermes and Isoptera but not molecular diagnostics Materials and Methods (Austin et al. 2002; Miura et al. 1998; Jenkins et al. 1998, 2001; Clement et al. 2001). Research on genetic vari- ´ Termites were Þeld collected continuously from ation within Reticulitermes, using DNA sequencing of 2000 to present at various locations in Canada, The a portion of the mtDNA COII gene, has revealed Bahamas, Europe, and the United States, and pre- sufÞcient variation to distinguish many species from served in 100% ethanol (Table 1 and see Table 5). North America, Europe, and Asia (Austin et al. 2002). Morphological identiÞcation of 32 Reticulitermes was Although developments in molecular biology have performed applying the taxonomic keys of Krishna improved our ability to differentiate species, identiÞ- and Weesner (1969), Scheffrahn and Su (1994), and cation by DNA sequencing can be both costly and Hostettler et al. (1995) (Table 1). Voucher specimens, time consuming. Polymerase chain reaction-restric- preserved in 100% ethanol, are maintained at the Ar- tion fragment length polymorphism (PCR-RFLP) has thropod Museum, Department of Entomology, Uni- become established as a reliable molecular technique versity of Arkansas, Fayetteville, AR. for identifying economically important insects includ- Alcohol-preserved specimens were allowed to dry ing Rhagoletis (Diptera: Tephritidae) species (Salazar on Þlter paper, and DNA was extracted from individ- et al. 2002), boll weevil, Anthonomus grandis Boheman ual worker, soldier, or alate heads using the Puregene (Coleoptera: Curculionidae) populations (Roehrdanz DNA isolation kit D-5000A (Gentra, Minneapolis, 2001), and Muscidifurax wasps (Hymenoptera: Ptero- MN). Extracted DNA was resuspended in 50 l of malidae) (Taylor and Szalanski 1999). We chose PCR- Tris:EDTA and stored at 20 C. PCR was conducted RFLP for this study because the technique is simple using two different mtDNA primer sets. The Þrst and inexpensive, very reliable and repeatable, and primer set, tercoII-idf (5 -TCTTCTTCCACGAY- provides discrete character states for quantitative CAYACAYTAATAA-3 ) and primer tercoII-idr (5 - comparisons of two or more species. TTTATGGGTAGTACYATTCGYTT-3 ) were used to In this study, a PCR-RFLP technique was developed amplify a 379-bp portion of the 5 end of the COII to study restriction site variation of two regions of the gene. The other primer set, 16S-r (5 -CGCCTGTT- mtDNA genome. Digestion of PCR products with spe- TATCAAAAACAT-3 ciÞc restriction enzymes yielded diagnostic patterns for the identiÞcation and differentiation of Reticuli- termes termites, regardless of caste. This is of great importance because of the frequent lack of morpho- logically distinct caste members collected, namely sol- diers or alates, which are essential for accurate deter- 1516 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 96, no. 5 Table 2. Reticulitermes genetic divergence (%), COII above, Table 3. Restriction fragment length polymorphisms from the 16S below diagonal COII gene Species R. ﬂavipes R. hageni R. virginicus R. tibialis Restriction Restriction Species Fragment(s) Pattern enzyme site R. ﬂavipes 1.1/1.6 7.2 8.0 8.9 R. hageni 5.7 0.1/2.2 5.8 8.3 Hinf I R. ﬂavipes 97, 294 197, 97, 85 A R. virginicus 5.5 2.6 0.2/1.0 7.4 (G/ANTC) R. hageni 184, 294 184, 110, 85 B R. tibialis 8.3 7.5 9.2 1.3/Ð R. tibialis 97, 294 197, 97, 85 A R. virginicus 294 294, 85 C TaqI R. ﬂavipes 292 292, 87 A (T/CGA) R. hageni Ð 379 B vidual termites was puriÞed and concentrated using R. tibialis Ð 379 B Microcon-PCR Filter Units (Millipore, Bedford, MA). R. virginicus Ð 379 B Samples were sent to The University of Arkansas DNA Sequencing Facility (Fayetteville, AR) for direct se- quencing in both directions using an ABI Prism 377 eni, and 3.8 nucleotides for R. virginicus. HinfI is a 5-bp DNA sequencer (Foster City, CA). GenBank acces- recognition enzyme and TaqI is a 4-bp recognition sion numbers for the termites subjected to DNA se- enzyme. P for TaqI was thus calculated to be 0.094 for quencing in this study and DNA sequences obtained R. ﬂavipes, 0.130 for R. hageni, and 0.059 for R. virgi- from GenBank are presented in Table 1. nicus for the COII marker. Based on the 16S RNA Restriction sites were predicted from the DNA se- sequences in this study, the average intraspeciÞc sub- quence data using Web Cutter 2.0 (Heiman 1997). stitution rate of this gene is 4.6 nucleotides for R. AmpliÞed DNA was digested according to manufac- ﬂavipes, 0.4 nucleotides for R. hageni, 0.8 nucleotides turerÕs (Promega, Madison, WI) recommendations for R. virginicus, and 5.4 nucleotides for R. tibilalis. For following Cherry et al. (1997) using the restriction the 16S marker, DraI is a 6-bp recognition enzyme, and enzymes Hinf I and TaqI for the COII gene and DraI, TspR I has a 5-bp recognition site. Thus, the possibility and TspR I for the 16S rRNA gene. Fragments were of an intraspecies polymorphism, P, for TspR I is 0.022 separated by either 2% agarose or 8% polyacrylamide for R. ﬂavipes, 0.002 for R. hageni, 0.004 for R. virgi- gel electrophoresis per Taylor et al. (1996). Gels were nicus, and 0.026 for R. tibialis. Applying these rates to stained with ethidium bromide then photographed our PCR-RFLP diagnostic technique, we were able to using a UVP BioDoc-it documentation system (Up- correctly classify 97 unknown samples collected from land, CA). four south central states, Missouri, and Virginia to R. ﬂavipes, R. virginicus, R. tibialis, or R. hageni (Table 5). Results The PCR amplicon from several of the unclassiÞed populations occurring in Arkansas, Louisiana, Mis- The mtDNA COII amplicon was 379 bp long for all souri, Oklahoma, and Texas were subjected to DNA of the Reticulitermes studied, and the 16S amplicon sequencing, and conÞrmed the species identiÞcation ranged from 425 to 429 bp. PCR was conducted on 2Ð 8 achieved with PCR-RFLP. individuals from each population and one individual from each population was sequenced (Table 1). Max- imum DNA sequence variation within each species Discussion and average DNA sequence variation among the four Reticulitermes species can be readily differentiated Reticulitermes species is presented in Table 2. Analysis with HinfI and TaqI digests of the 379-bp amplicon of of the DNA sequencing data with Webcutter 2.0 soft- the mtDNA COII gene, and DraI, and TspRI digests of ware revealed that no single restriction enzyme was the 428-bp 16S amplicon. Although both markers re- able to completely separate all of the Reticulitermes quire only two restriction enzymes to differentiate all species. However, combinations of different enzyme four Reticulitermes species, the lower amount of in- digestion results could aid in species identiÞcation. traspeciÞc variation in the 16S region makes it a more Examining the COII marker Þrst, the Hinf I and TaqI reliable diagnostic marker relative to the COII region. pair was able to differentiate all of the species studied Additionally, based on our evaluation of this tech- (Table 3). No variation was observed within species for these two restriction enzymes. Viewing the 16S marker, two restriction enzymes, DraI and TspRI, Table 4. Restriction fragment length polymorphism from the were required to differentiate all four species (Table 16S gene 4). We can estimate the possibility (P) of intraspecies Restriction Restriction Species Fragment(s) Pattern enzyme site polymorphism of PCR-RFLP analysis using the for- mula: P ( /N) [(N/4n) n] n/4n, where DraI R. ﬂavipes Ð 428 A is the average intraspeciÞc diversity, N is the nucle- (TTTAAA) R. hageni Ð 425 A R. tibialis Ð 425 A otide length of the gene, and n is the number of R. virginicus 254 254, 172 B recognition sequence sites of the restriction enzyme TspRI R. ﬂavipes 375 375, 53 A (Lin et al. 2002). The average intraspeciÞc substitu- (CAGTG) R. hageni 143 283, 143 B tion of the 379-bp region of the COII gene is 6.0 R. tibialis 141, 372 231, 141, 53 C R. virginicus 142 284, 142 B nucleotides for R. ﬂavipes, 8.3 nucleotides for R. hag- October 2003 SZALANSKI ET AL.: Reticulitermes MOLECULAR DIAGNOSTICS 1517 Table 5. Identiﬁcation of unclassiﬁed Reticulitermes samples species, correct identiÞcation of the species involved can assist professional pest managers with understand- N Pattern Pattern ing of why their respective control efforts might have IdentiÞcation Collection site Colonies COIIa 16Sa failed. Numerous tactics used by professional pest R. ﬂavipes Fayetteville, AR 3 AA AA managers employ the biology and ecological demands Chauvin, LA 1 AA AA Houma, LA 1 AA AA of various termite species (e.g., baiting regimes for Beaver Co., OK 1 AA AA colony suppression or elimination), which might be Greer Co., OK 2 AA AA jeopardized because of subtle differences that can McCurtain Co., OK 2 AA AA occur between the various species and subspecies of Norman, OK 1 AA AA Oklahoma City, OK 2 AA AA Reticulitermes. Payne Co., OK 1 AA AA Molecular diagnostics will aid in the identiÞcation Spiro, OK 1 AA AA of future introductions of Reticulitermes species to the Stillwater, OK 2 AA AA United States or in ongoing genetic studies that de- Wagoner Co., OK 1 AA AA Addison, TX 1 AA AA mand correct identiÞcation to ascertain what popu- Arlington, TX 2 AA AA lation the species in question may have originated Austin, TX 1 AA AA from. Molecular diagnostics is also more cost effective Balch Springs, TX 1 AA AA than cuticular hydrocarbon examinations, which re- Beaumont, TX 1 AA AA Beeville, TX 1 AA AA quire more costly chromatographic techniques and Blanco, TX 1 AA AA materials, generally more termites per location, and Bryan, TX 1 AA AA may have chemical polymorphisms that vary among Carrollton, TX 2 AA AA castes. Additionally, based on DNA sequence data College St., TX 1 AA AA Combine, TX 2 AA AA (Austin et al. 2002), the 379-bp marker used in this Corpus Christi, TX 2 AA AA study could be used to identify Reticulitermes luciﬁgus Dallas, TX 18 AA AA and Reticulitermes grassei from Europe and Asia. Eu- DeSoto, TX 1 AA AA ropean introductions of R. ﬂavipes from the United El Paso, TX 2 AA AA Ft. Worth, TX 3 AA AA States have been documented to have occurred in Hempstead, TX 1 AA AA Germany (Becker 1970) and Austria (Hrdy 1961). ` Irving, TX 2 AA AA More recently, introductions of damaging exotic Re- MansÞeld, TX 2 AA AA ticulitermes spp. are known to occur in South America, Milano, TX 2 AA AA Odessa, TX 1 AA AA including: R. lucifugus in Uruguay (Aber and Fontes Post, TX 1 AA AA 1993) and R. ﬂavipes in Santiago and Valparaiso, Chile Red Water, TX 1 AA AA (Clement et al. 2001) where they have established ´ Rowlette, TX 2 AA AA themselves as serious structural pest of wooden tim- Stephenville, TX 2 AA AA Taylor, TX 1 AA AA bers. In Europe, discoveries of R. grassei in southwest- The Woodlands, TX 2 AA AA ern England have prompted new studies to assist in Suffolk, VA 2 AA AA understanding the risks associated with their occur- R. virginicus Fayetteville, AR 8 CB BB rence and their associated threat to structural timbers McGehee, AR 2 CB BB Columbia, MO 1 CB BB in the United Kingdom. The anthropogenic introduc- Oklahoma City, OK 1 CB BB tions of hardy Reticulitermes species (existing beyond Bryan, TX 1 CB BB what would be considered their “normal” habitation Roanoke, VA 2 CB BB range) are likely to continue. Clearly, problems with R. hageni Conway, AR 1 BB AB Fayetteville, AR 2 BB AB the classiÞcation of synonymous species are another R. tibialis Ft. Worth, TX 1 AB AC area in which having the alternative viewpoint of this El Paso, TX 1 AB AC technique is not only valid, but necessary. This has a been demonstrated when looking at the synonymy of Restriction enzyme pattern in the order HinfI, Taq I for COII and the R. lucifugus complex in Europe and the Middle Dra I, TspR I for 16S (see Tables 2 and 3). East (Austin et al. 2002), the R. hesperus complex in California and the southwest United States (Haverty nique using unclassiÞed Þeld-collected Reticulitermes, and Nelson 1997, Haverty et al. 1999), and the various the method is more reliable than predictions based on complexes of R. tibialis in the southwest United States putative mutation rates. As a diagnostic tool, this pro- (Myles 2000). Quick and accurate identiÞcation of cess could eliminate much of the ambiguity involved these species will be a daunting task for inexperienced in morphological identiÞcation of species within the professional pest managers operating in areas in which genus Reticulitermes or with other taxa within the previously undescribed species are now occurring. order Isoptera. The PCR-RFLP molecular diagnostic Molecular diagnostics is just one way that we might method presented herein can be used with workers, gain insight as to the origins of newly introduced soldiers, or alate specimens. The ability to identify species so that intervention may be directed in the specimens to species regardless of caste is a great most economically effective manner (whether iden- advantage over traditional morphological methods in tifying the source of introduction, discriminating spe- which keys are designed primarily for soldiers and cies for the application of corrective treatment mea- alates. Although the control tactic used for most Re- sures, or for cataloging species with the intent to ticulitermes infestations is unlikely to vary between correctly classify species that will likely be discovered 1518 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 96, no. 5 later). It also supports the possibility of classifying Haverty M. L., and L. Nelson. 1997. Cuticular hydrocarbons samples currently in museums that have not yet of Reticulitermes from California indicate undescribed been resolved. For example, within the insect col- species. Comp. Biochem. Physiol. 118B: 95Ð98. lection at Texas A&M University, College Sta- Haverty, M. L., L. Nelson, and B. Forschler. 1999. New tion, TX (http://entowww.tamu.edu/new/research/ cuticular hydrocarbon phenotypes of Reticulitermes from the United States. Sociobiology 34: 1Ð21. systematics/collection.html), there are presently 227 Heiman, M. 1997. Webcutter 2.0. (http://www.ccsi.com/ Reticulitermes samples, of which only 96 have been Þrstmarket/cutter/cut2.html). classiÞed to species (85% R. ﬂavipes, 7% R. virginicus, Hostettler, N. C., D. W. Hall, and R. H. Scheffrahn. 1995. 7% R. hageni, and 1% R. tibialis). Fifty-eight percent, Intracolony morphometric variation and labral shape in representing 131 vials, have not yet been classiÞed to Florida Reticulitermes (Isoptera: Rhinotermitidae) sol- species. diers: signiÞcance for identiÞcation. Flor. Entom. 78: 119 Ð The PCR-RFLP method presented herein was de- 129. veloped to provide researchers and professional pest Hrdy I. 1961. Contribution to the knowledge of European `, managers with a rapid and simple molecular method species of the genus Reticulitermes. Act faun. Entomol. for accurately differentiating four species of Reticuli- Mus. Nat. Prague 7: 97Ð107. termes encountered in the south central United States, Jenkins, T. M., C. J. Basten, R. Dean, S. E. Mitchell, S. Kresovich, and B. T. Forschler. 1998. Matriarchal ge- and could be used to identify Reticulitermes species netic structure of Reticulitermes (Isoptera: Rhinotermiti- that may be introduced to the United States. Field- dae) populations. Sociobiology 33: 239 Ð263. collected samples can be preserved in ethanol at the Jenkins, T. M., R. E. Dean, R. Verkerk, and B. T. Forschler. collection site and stored for several years until iden- 2001. Phylogenetic analysis of two mitochondrial genes tiÞcation is required. The PCR ampliÞcation is robust, and one nuclear intron region illuminate European sub- and the use of a small amplicon, 450 bp, facilitates the terranean termite (Isoptera: Rhinotermitidae) gene ßow, use of specimens that have not been optimally pre- taxonomy, and introduction dynamics. Molec. Phylog. served (Taylor et al. 1996). The restriction enzymes Evol. 20: 286 Ð293. used in this study are economical, costing less than Kambhampati, S., and P. T. Smith. 1995. PCR primers for $12.50 per 1,000 U, and yield fragments that can be the ampliÞcation of four insect mitochondrial gene frag- readily visualized under less-than-optimal PCR and ments. Ins. Molec. Biol. 4: 233Ð236. Krishna, K., and F. M. Weesner. 1969. Biology of termites, electrophoretic conditions. The entire set of proce- vol. 1. Academic, New York. dures, from DNA extraction to Þnal identiÞcation, can Lin, Y., Y. Poh, S. Lin, and C. Tzeng. 2002. Molecular tech- be completed within a single working day. niques to identify freshwater eels: RFLP analysis of PCR- ampliÞed DNA fragments and allele-speciÞc PCR from mitochondrial DNA. Zool. Stud. 41: 421Ð 430. Miura, T., K. Maekawa, O. Kitade, T. Abe, and T. Matsumoto. Acknowledgments 1998. Phylogenetic relationships among subfamilies in We thank R. Gold, B. Foster, B. Kard, M. Haßey, C. Casta- higher termites (Isoptera: Termitidae) based on mito- lano, U. Noldt, R. Houseman, T. Myles, G. Henderson, D. chondrial COII gene sequences. Ann. Entomol. Soc. Am. Miller, and R. Scheffrahn for providing samples. C. D. Steel- 91: 515Ð523. man provided helpful suggestions and critical reviews of the Myles, T. G. 2000. Evaluation of the occurrence of termite manuscript. Research was supported in part by the University species and potential termite burrowing depths in the of Arkansas, Arkansas Agricultural Experiment Station. area of the Nevada test site: present day and for the next 10, 000 years. Special Report, Neptune and Company, Inc., Los Alamos, NM, Supplemental for review of the NTS GCD PA by Sandia National Laboratory. Document References Cited No 05100-01. Aber, A., and L. R. Fontes. 1993. Reticulitermes lucifugus [NPMA] National Pest Management Association. 2003. Pro- (Isoptera, Rhinotermitidae), a pest of wooden structures, tect your largest investment: check home for termites. is introduced into the South American continent. Socio- (http://184.108.40.206/media/art_protect_your_home.asp). biology 21: 335Ð339. National Termite Survey. 2002. (http:// www.termitesurvey. Austin, J. W., A. L. Szalanski, P. Uva, A. G. Bagne `res, and A. com/). Kence. 2002. A comparative genetic analysis of the sub- Roehrdanz, R. L. 2001. Genetic differentiation of southeast- terranean termite genus Reticulitermes (Isoptera: Rhino- ern boll weevil and thurberia weevil populations of An- termitidae). Ann. Entomol. Soc. Am. 95: 753Ð760. thonomus grandis (Coleoptera: Curculioinidae) using mi- Becker, G. 1970. Reticulitermes in Mittel und West Europa. tochondrial DNA. Ann. Entomol. Soc. Am. 94: 928 Ð935. Z. angew. Entomol. 65: 268 Ð278. Salazar, M., C. Theoduloz, A. Vega, F. Poblete, E. Gonzalez, Cherry, T., A. L. Szalanski, T. C. Todd, and T. O. Powers. R. Badilla, and L. Meza-Basso. 2002. PCR-RFLP identi- 1997. The internal transcribed spacer region of Belono- Þcation of endemic Chilean species of Rhagoletis laimus (Nemata: Belonolaimidae). J. Nematol. 29: 21Ð29. (Diptera: Tephritidae) attacking Solanaceae. Bull. Ento- Cle ´ment, J. L., A. G. Bagne `res, P. Uva, L. Wilfert, A. Quin- mol. Res. 92: 337Ð341. tana, J. Reinhard, and S. Dronnet. 2001. Biosystematics Scheffrahn, R. H., and N. Su. 1994. Keys to soldier and of Reticulitermes termites in Europe: morphological, winged adult termites (Isoptera) of Florida. Flor. Entom. chemical and molecular data. Insectes Sociaux 48: 202Ð 77: 460 Ð 474. 215. Simon, C., F. Frati, A. Beckenbach, B. Crespi, H. Liu, and P. Forschler, B. T., and V. Lewis. 1997. Why termites can Flook. 1994. Evolution, weighting, and phylogenetic dodge your treatment. Pest Control 65: 42Ð53. utility of mitochondrial gene sequences and a compila- October 2003 SZALANSKI ET AL.: Reticulitermes MOLECULAR DIAGNOSTICS 1519 tion of conserved polymerase chain reaction primers. Taylor, D. B., and A. L. Szalanski. 1999. IdentiÞcation of Ann. Entomol. Soc. Am. 87: 651Ð701. Muscidifurax spp. by polymerase chain reaction-restric- Su, N. 1993. Managing subterranean termite populations, tion fragment length polymorphism. Biol. Cont. 15: 270 Ð pp. 45Ð50. In K. B. Wildey and W. H. Robinson [eds.], 273. Proceedings, 1st International Conference on Insect Taylor, D. B., A. L. Szalanski, and R. D. Peterson II. 1996. Pests in the Urban Environment. BPCC Wheatons Ltd., IdentiÞcation of screwworm species (Diptera: Callipho- Exeter, United Kingdom. ridae) by polymerase chain reaction-restriction fragment Su, N., and R. H. Scheffrahn. 1990. Comparison of eleven length polymorphisms. Med. Vet. Entomol. 10: 63Ð70. soil termiticides against Formosan subterranean termite [UNEP] United Nations Environment Program, and [FAO] and the Eastern subterranean termite. J. Econ. Entomol. Food and Agricultural Organization. 2000. Termite bi- 83: 1918 Ð1924. ology and management workshop. UNEP, Geneva, Swit- Szalanski, A. L., D. S. Sikes, R. Bischof, and M. Fritz. 2000. zerland. Population genetics and phylogenetics of the endan- Weesner. F. M. 1965. The termites of the United States. A gered American burying beetle, Nicrophorus america- handbook. National Pest Control Association, Elizabeth, NJ. nus (Coleoptera: Silphidae). Ann. Entomol. Soc. Am. 93: 589 Ð594. Received for publication 19 July 2002; accepted 19 May 2003.