Classification of Mosquitoes in Tribe Aedini (Diptera: Culicidae [PDF]

FORUM

Classification of Mosquitoes in Tribe Aedini (Diptera: Culicidae): Paraphylyphobia, and Classification Versus Cladistic Analysis HARRY M. SAVAGE Centers for Disease Control and Prevention, P.O. Box 2087, Fort Collins, CO 80522

KEY WORDS Aedini classiÞcation, Aedes, Ochlerotatus, paraphylyphobia, mosquito classiÞcation

MANY MOSQUITO SPECIES IN the tribe Aedini, a cosmopolitan group represented by 11 genera and ⬇1,239 species, are important vectors of human and animal diseases, and many others are of considerable economic importance as nuisance or pest species. To facilitate communication and information exchange among epidemiologists, physicians, veterinarians, virologists, parasitologists, public health workers, and medical entomologists, it is essential that a stable nomenclature or system of names be maintained. Such communication has seldom been more important worldwide, because mosquito-borne diseases draw sincere attention from those charged with the responsibility of public health at all levels. Mosquito taxonomists strive to use new methods of analysis and new data sets to address the phylogeny and classiÞcation of mosquitoes, and future change in classiÞcation and nomenclature is inevitable. However, major changes in generic concept, the elevation of 32 subgenera within Aedes (sensu Edwards 1932, 1941) to generic status, and the resultant spelling changes in hundreds of species names by Reinert et al. (2004) demand consideration by all parties interested in mosquitoborne diseases. Although there are many technical issues associated with the manuscript of Reinert et al. (2004), visit the following Web-based forum to access the discussion, http://wrbu.si.edu/forums, I believe that their entire approach to Aedini systematics was ßawed by an inordinate fear of paraphyletic taxa or paraphylyphobia, and their inability to distinguish between classiÞcation and cladistic analysis.

In this communication, I brießy review taxonomic categories in a zoological classiÞcation, the International Code of Zoological Nomenclature (the Code), the development of generic concept within the Culicidae, classiÞcation within the tribe Aedini and genus Aedes, and explain how confusion between classiÞcation and cladistic analysis and paraphylyphobia lead to the classiÞcation of Aedini proposed by Reinert et al. (2004). Taxonomic Categories and the International Code of Zoological Nomenclature Unlike the species category, which is deÞned by the biological species concept (Mayr 1969, 1970), the genus and subgenus categories are not based on a biological concept nor are they strictly deÞned. Usage and deÞnition of generic and subgeneric taxa may vary among animal groups and even between authors working on the same group of animals. Traditionally, phylogenetic taxonomists deÞned genera as a group of related species sharing a common phylogenetic origin that differ from species of other related genera by a decided gap in morphological characters (Mayr 1969). For families such as the Culicidae that are of interest to many professional groups, well-deÞned, easily identiÞable genera based on morphology are an asset. The Code (ICZN 1999) provides rules for forming names in the species, subgenus, genus, subfamily, and family categories and requires that adjectival species names correspond to the gender of the genus in which they are placed (Table 1). For example (Table 1), if

Downloaded from https://academic.oup.com/jme/article-abstract/42/6/923/886357 by guest on 08 April 2019

J. Med. Entomol. 42(6): 923Ð927 (2005)

ABSTRACT Many mosquito species are important vectors of human and animal diseases, and others are important nuisance species. To facilitate communication and information exchange among professional groups interested in vector-borne diseases, it is essential that a stable nomenclature be maintained. For the Culicidae, easily identiÞable genera based on morphology are an asset. Major changes in generic concept, the elevation of 32 subgenera within Aedes to generic status, and changes in hundreds of species names proposed in a recent article demand consideration by all parties interested in mosquito-borne diseases. The entire approach to Aedini systematics of these authors was ßawed by an inordinate fear of paraphyletic taxa or Paraphylyphobia, and their inability to distinguish between classiÞcation and cladistic analysis. Taxonomists should refrain from making taxonomic changes based on preliminary data, and they should be very selective in assigning generic names to only the most important and well-deÞned groups of species.

924 Table 1.

JOURNAL OF MEDICAL ENTOMOLOGY Classificiation of Ae. (Stegomyia) albopictus

Category Family Subfamily Tribe Genus Lineage Subgenus Species Group Species Complex Species

Culicidae Culicinae Aedini Aedes Aedes Lineage (Stegomyia) Scutellaris Group Ae. (Stegomyia) albopictus (Skuse)

the species albopictus (Skuse) is placed in the masculine genus Aedes, the correct species name is Aedes albopictus (Skuse). However, if the feminine subgenus Stegomyia was elevated to generic status, and if the species albopictus was placed in the genus Stegomyia, the species name would have to be amended to the feminine form and the correct name would be Stegomyia albopicta (Skuse). Some public health workers falsely assume that the Code requires them to accept the results of new taxonomic papers and new classiÞcations. However, this is absolutely false. The purpose of the Code is to promote stability and universality of taxonomic names and to ensure that the valid name of each species is unique within each genus, according to the rules of priority. The Code does not provide rules or guidance on assigning rank. ClassiÞcations above the species level depend on subjective interpretation of taxonomic data and the philosophy adopted by the investigator. The Code does not restrict freedom of taxonomic thought or action. The Preamble and the Introduction to the Code make this very clear: The Preamble states, “The objects of the Code are to promote stability and universality in the scientiÞc names of animals and to ensure that the name of each taxon is unique and distinct. All its provisions and recommendations are subservient to those ends and none restricts the freedom of taxonomic thought or actions”; and the Introduction further states, “Nomenclature does not determine the inclusiveness or exclusiveness of any taxon, nor the rank to be accorded to any assemblage of animals, but rather, provides the name that is to be used for a taxon whatever taxonomic limits and rank are given to it” (ICZN 1999). The only formal names available to name groups of species or lineages above species and below subfamily are subgenera and genera (see Table 1 for an example). Taxonomists should be very selective in assigning generic and subgeneric names to only the most important and well deÞned lineages or groups of species. Generic Concept with Culicidae Work on the taxonomy and classiÞcation of mosquitoes developed slowly. In the 10th edition of Sys-

tema Naturae (Linnaeus 1758), two mosquito species were placed in the single genus Culex L. In 1818, Meigen (1818) established two new genera, Anopheles Meigen and Aedes Meigen, bringing the total number of genera to three. During the remainder of the 19th century, nine new generic names were published, including four genera that remain in use today: Hemagogus Williston, Psorophora Robineau-Desvoidy, Sabethes Robineau-Desvoidy, and Uranotaenia Lynch Arribalzaga (Savage and Strickman 2004). In 1900, experimental transmission of human pathogens by mosquitoes was demonstrated. Patrick Manson experimentally infected two human volunteers with malaria by bites of Anopheles mosquitoes brought from Italy where they had fed upon a malaria patient; and Walter Reed and colleagues working in Havana demonstrated the transmission of yellow fever virus to healthy volunteers by Aedes (Stegomyia) aegypti L. infected on hospitalized patients (Philip and Rozeboom 1973). These events of 1900 spurred work on the taxonomy and biology of mosquitoes. Just 10 yr later, Theobald (1910) recognized 1,050 species in 149 genera (Edwards 1932). In 1906, Dyar and Knab (1906) developed a classiÞcation of the mosquitoes based on larval morphology and deÞned genera, including Aedes, broadly. Howard et al. (1915, 1917), in their classiÞcation of the mosquitoes of North and Central America and the West Indies, furthered developed the concept that genera should be broadly deÞned and easily identiÞable. This broad usage of genera within the mosquitoes, including Aedes, was accepted by Edwards (1917) and by Barraud (1928), and the name Aedes was universally used in this inclusive sense during the interval 1906 Ð2000 (Savage and Strickman 2004). Dyar (1928), in his comprehensive classiÞcation of the mosquitoes of the Americas, deÞned genera broadly and generally as they are used today, with well deÞned internal groups placed as subgenera. In 1932, Edwards extended the generic and subgeneric concepts employed by Dyar (Dyar and Knab 1906, Dyar 1928) and placed 1,400 species in 30 genera and 89 subgenera and offered a classiÞcation of the Culicidae of the World that forms the basis for the present classiÞcation. Edwards (1932) pointed out that the use of more inclusive generic concepts made the relationships among species clearer and the generic limits more easily deÞned. Edwards (1932) also pointed out that smaller groups of species can be usefully designated by subgeneric names. The broad deÞnitions of genera developed by Edwards (1932) allowed adult female mosquitoes to be identiÞed to genus in the laboratory with a good-quality dissecting microscope and allowed mosquito genera to be recognized in the Þeld as morphological and ecological units. Tribe Aedini The traditional classiÞcation of the tribe Aedini is presented in Table 2. Eleven genera and ⬇1,239 species are included. All genera except Aedes are small to moderate in species number. All genera were established based on distinct morphological characters between 1818 and 1908, with the exception of two mor-

Downloaded from https://academic.oup.com/jme/article-abstract/42/6/923/886357 by guest on 08 April 2019

The formation of names in the family, subfamily, genus, subgenus, and species categories is treated by Articles in the International Code of Zoological Nomenclature (the Code); above these categories are in boldface. The category Tribe is addressed by Recommendations in the Code. Other hierarchical levels sometimes used by biologist such as Lineage, Species Group, and Species Complex are not addressed by the Code.

Vol. 42, no. 6

November 2005

SAVAGE: CLASSIFICATION OF MOSQUITOES IN TRIBE AEDINI

Table 2. Traditional classification of genera within the Tribe Aedini (after Edwards 1932, Belkin 1962, Knight and Stone 1977, Reinert 1999) Approximate no. of species

Aedes Meigen 1818 Armigeres Theobald 1901 Ayurakitia Thurman 1954 Eretmapodites Theobald 1901 Haemagogus Williston 1896 Heizmannia Ludlow 1905 Opifex Hutton 1902 Psorophora Robineau-Desvoidy 1827 Udaya Thurman 1954 Verrallina Theobald 1903 Zeugnomyia Leicester 1908 Total

928 51 2 48 28 34 1 47 3 93 4 1,239

The number of species in each genus was obtained by multiple searches of the Water Reed Biosystematics Unit Web site in April, 2005 (http://wrbu.si.edu).

phologically unusual taxa with very few species established by Thurman in 1954 (Thurman 1954a, b; Mattingly 1958): Ayurakitia Thurman and Udaya Thurman. Verrallina Theobald was established in 1903, but redeÞned by Reinert (1999). The genus Aedes includes 44 subgenera (Table 3) that can be divided into two lineages, the Aedes Lineage and the Ochlerotatus Lineage, based primarily on characters of the male and female genitalia discovered by Dyar (1918), Edwards (1921), and Belkin (1962) and discussed in detail by Reinert (2000) and Savage and Strickman (2004).

Table 3. Classification of the subgenera of Aedes Meigen into two lineages after the arguments of Dyar (1918), Edwards (1921, 1941), Belkin (1962), Reinert (2000), and Savage and Strickman (2004) Aedes lineage

Ochlerotatus lineage

(Aedes) Meigen (Aedimorphus) Theobald (Alanstonea) Mattingly (Albuginosus) Reinert (Belkinius) Reinert (Bothaella) Reinert (Cancraedes) Edwards (Christophersiomyia) Barraud (Diceromyia) Theobald (Edwardsaedes) Belkin (Fredwardsius) Reinert (Huaedes) Huang (Indusius) Edwards (Isoaedes) Reinert (Leptosomatomyia) Theobald (Lorrainea) Belkin (Neomelaniconion) Newstead (Paraedes) Edwards (Pseudarmigeres) Stone & Knight (Scutomyia) Theobald (Skusea) Theobald

(Abraedes) Zavortink (Aztecaedes) Zavortink (Bruceharrisonius) Reinert (Chaetocruiomyia) Theobald (Finlaya) Theobald (Geoskusea) Edwards (Gymnometopa) Coquillett (Halaedes) Belkin (Howardina) Theobald (Kenknightia) Reinert (Kompia) Aitken (Levua) Stone and Bohart (Macleaya) Theobald (Molpemyia) Theobald (Mucidus) Theobald (Nothoskusea) Dumbleton (Ochlerotatus) Lynch Arribalzaga (Protomacleaya) Theobald (Pseudoskusea) Theobald

(Stegomyia) Theobald

(Rhinoskusea) Edwards (Rusticoidus) Shevchenko and Prudkina (Zavortinkius) Reinert

Note subgeneric names are enclosed in parentheses.

Cladistics Analysis and Paraphyly With the advent of cladistics (Hennig 1966), taxonomists began to emphasize that genera should be monophyletic assemblages of species and that gap criteria, or morphological distinction, were inappropriate (Wiley 1981). A monophyletic group is a group that includes an ancestor and all of its descendants: for example, Psorophora or Hemagogus in Fig. 1. The acceptance of cladistic methods of phylogenetic analysis lead to a search for monophyletic groups or taxa by a generation of taxonomists. A polyphyletic group is a group that includes descendants from different ancestors. Nearly all taxonomists agree that polyphyletic taxa are to be avoided (Diggs and Lipscomb 2002). However, cladists failed to realize that the removal of a monophyletic group from a phylogenetic tree resulted in the formation of a paraphyletic group (Brummitt 2002, 2003). A paraphyletic group is a group of species that share a common ancestor but does not include all the descendants of the common ancestor: for example, the group containing the remaining species in Fig. 1 after the monophyletic taxa Psorophora, Hemagogus and Eretmapodites are removed. Although the search for monophyletic groups deÞned by synapomorphies, or shared derived character states, received a conceptual basis from cladistic theory, it is important to note that 10 of the 11 Aedini genera in Table 2 were established before 1908 based on unusual morphological characters and that these 10 genera are monophyletic in the analysis of Reinert et al. (2004). These genera, or lineages, were cut out of the Aedini evolutionary tree, and the less morphologically welldeÞned lineages representing the 44 subgenera of Aedes (Table 3) were left in the paraphyletic genus Aedes by early taxonomists. Although cladistic terminology was not in use, early workers recognized that the distinct genera of Aedini, for example Psorophora and Hemagogus (Dyar 1928, Hendrickson and Sokal 1968), evolved from within Aedes and that Aedes was therefore paraphyletic. To avoid use of the resultant paraphyletic taxon Aedes, Reinert et al. (2004) felt obligated to dismember Aedes by elevating all monophyletic groups at the corresponding level within the tree to genera (Kristensen 1982). For example, all taxa represented by the unnamed lines in Fig. 1, which have common ancestors denoted by squares at approximately the same level as the common ancestors of Psorophora, Hemagogus, and Eretmapodites, also would be elevated to generic status. If all of the monophyletic groups in Fig. 1 can be deÞned by morphological characters that allow them to be identiÞed and recognized, this leads to the generation of numerous new well-deÞned genera and would be considered beneÞcial by many taxonomists. However, if the new monophyletic groups are poorly deÞned and cannot be recognized as morphological or biological groups, the destruction of well-established and deÞned paraphyletic taxa into many poorly deÞned monophyletic taxa results in poor taxonomy and a useless classiÞcation (Michener 2000, Brummitt 2002, Diggs and Lipscomb 2002, Nordal and Stedje 2005). Because most of these new genera are

Downloaded from https://academic.oup.com/jme/article-abstract/42/6/923/886357 by guest on 08 April 2019

Genus

925

926

JOURNAL OF MEDICAL ENTOMOLOGY

Vol. 42, no. 6

poorly deÞned and difÞcult or impossible to identify, medical entomologists and other Þeldworkers would be forced to identify a specimen Þrst to species by using a broad Aedini key then look in an appendix or table to determine what genus it was assigned to. ReinertÕs fear of paraphyletic taxa (Reinert et al. 2004), admitted on page 327, “we stress that our study removes the evidently paraphyletic ÔlargeÕ genera, with a necessary fragmentation of traditional taxa and a subjective decision that the fragments be treated as genera rather than subgenera,” results in the treatment of numerous poorly deÞned “fragments” as genera. A conservative and rational approach would have been to place these “fragments” as subgenera within the paraphyletic genus Aedes as they were in the traditional classiÞcation of Edwards 1932, 1941 following the argument of Savage and Strickman (2004); see Table 3. The treatment of these fragments as subgenera maintains all the information associated with the subgeneric name, for example subgenus Ae. (Stegomyia) and the genus Stegomyia, convey exactly the same information, but use of the former subgeneric name avoids the need to amend the species names of included species. The article by Reinert et al. (2004) was the Þrst major cladistic analysis of the Aedini, and future studies using different sets of morphological characters or/and molecular data will certainly produce different fragments, and different relationships among the fragments and well-deÞned monophyletic groups. If these fragments are continually treated as genera, the names of common vector and nuisance species also will continually change. This will lead to disruption of communication among those interested in mosquito-borne diseases, the destruction of the information system associated with a stable classiÞcation, and difÞculties in literature search and database management. Phylogenetic analysis is desperately needed within the family Culicidae. However, I am not hopeful that the phylogeny, classiÞcation, and nomenclature of the Aedini will stabilize in the near future. The use of

subgeneric names for fragments will signiÞcantly reduce confusion. If the monophyletic nature of subgenera of Aedes becomes Þrmly established, then analysis of the evolutionary relationships among the subgenera of Aedes and genera of Aedini will eventually lead to a reliable phylogeny and stable classiÞcation. As subgenera become well deÞned, taxonomists may choose to elevate selected subgenera to genera based on Þrm phylogenetic evidence and morphological characters that allow these new genera to be reliably identiÞed. It is unfortunate that Reinert et al. (2004) felt compelled to change the classiÞcation of Aedini to reßect the topology of their preferred tree based on study of less that 10% (119/1,239) of the species in the tribe. If these authors would have presented their preliminary results as the Þrst major attempt to understand the phylogeny of the Aedini, their article would have been hailed as an important contribution to mosquito systematics. These authors amassed a large morphological data set and presented a framework for future analyses. However, because these authors tried to translate the results of their cladistic analysis directly into a reclassiÞcation of genera, their contribution should be rejected. Other cladists, recognizing that more names are needed for clades within a tree than are available in the Code, are developing an alternative code of names, PhyloCode, for naming the many clades and recognizing hierarchy within phylogenetic trees (Cantino and de Queiroz 2004). These codes could be used simultaneously to discuss the evolution and biology of lineages or groups of species. However, only the most well-deÞned and hence stable lineages should be assigned the rank of genera. Acknowledgments I thank Richard K. Brummitt (The Herbarium, Royal Botanic Gardens, Kew, Richmond, United Kingdom), Tom

Downloaded from https://academic.oup.com/jme/article-abstract/42/6/923/886357 by guest on 08 April 2019

Fig. 1. Hypothetical phylogenetic tree. Black circles represent extant species; open circles and open squares represent postulated ancestor species. Tree is ⬇3% of the size of the Aedini tree. Number of slash marks on basal lineages is indicative of degree of evolutionary change, either unusual or distinct morphological changes as noted by early taxonomists, or potential synapomorphies in the sense of cladistics. Heavy lines above extant species indicate limits of taxa, either the monophyletic genera Psorophora, Hemogogus, and Eretmapodites, or selected subgenera of Aedes from Table 3.

November 2005

SAVAGE: CLASSIFICATION OF MOSQUITOES IN TRIBE AEDINI

Zavortink (Bohart Museum, Department of Entomology, University of California, Davis, CA), and Roy Campbell and Theresa Smith (Centers for Disease Control and Prevention, Fort Collins, CO) for valuable comments on a draft of this manuscript, although the author retains sole responsibility for the opinions expressed herein.

References Cited

menclature, 4th ed. International Trust of Zoological Nomenclature, Padova, Italy. Knight, K. L., and A. Stone. 1977. A catalog of the mosquitoes of the world (Diptera: Culicidae), vol. VI. The Thomas Say Foundation, Entomological Society of America, Lanham, MD. Kristensen, N. P. 1982. Splitting or widening: remarks on the taxonomic treatment of paraphyletic taxa. Ann. Zool. Fennici 19: 201Ð202. Linnaeus, C. 1758. Systema naturae, 10th ed., vol. 1. Holmiae. Mattingly, P. F. 1958. The culicine mosquitoes of the Indomalayan Area. Part Spaceiiiqq. Genus Aedes Meigen, subgenera Paraedes Edwards, Rhinoskusea Edwards and Cancraedes Edwards. British Museum of Natural History, London, United Kingdom. Mayr, E. 1969. Principles of systematic zoology. McGrawHill, New York. Mayr, E. 1970. Populations, species, and evolution. Belknap Press, Cambridge, MA. Meigen, J. W. 1818. Systematische Beschreibung der bekannten europaischen zweißugligen Insekten. Vol. 1. Forstmann, Aachen, Germany. Michener, C. D. 2000. The bees of the world. Johns Hopkins University Press, Baltimore, MD. Nordal, I., and B. Stedje. 2005. Paraphyletic taxa should be accepted. Taxon 54: 5Ð 6. Philip, C. B., and L. E. Rozeboom. 1973. Medico-veterinary entomology: a generation of progress, pp. 333Ð360. In R. F. Smith, T. E. Mittler, and C. N. Smith [eds.], History of entomology. Annual Review of Entomology, Palo Alto, CA. Reinert, J. F. 1999. Restoration of Verrallina to generic rank in tribe Aedini (Diptera: Culicidae) and descriptions of the genus and three included subgenera. Contr. Am. Entomol. Inst. (Gainesville) 31: 1Ð105. Reinert, J. F. 2000. New classiÞcation for the composite genus Aedes (Diptera: Culicidae: Aedini), elevation of subgenus Ochlerotatus to generic rank, reclassiÞcation of the other subgenera, and notes on certain subgenera and species. J. Am. Mosq. Control Assoc. 16: 175Ð188. Reinert, J. F., R. E. Harbach, and I. J. Kitching. 2004. Phylogeny and classiÞcation of Aedini (Diptera: Culicidae), based on morphological characters of all life stages. Zool. J. Linn. Soc. 142: 289 Ð368. Savage, H. M., and D. Strickman. 2004. The genus and subgenus categories within Culicidae and placement of Ochlerotatus as a subgenus of Aedes. J. Am. Mosq. Control Assoc. 20: 208 Ð214. Theobald, F. V. 1910. A monograph of the Culicidae or mosquitoes, vol. 5. British Museum, London, United Kingdom. Thurman, D. C., Jr. 1954a. Ayurakitia, a new genus of mosquito from northern Thailand (Diptera: Culicidae). J. Wash. Acad. Sci. 44: 197Ð200. Thurman, D. C., Jr. 1954b. The discovery of Paraedes (Udaya) argyrurus Edwards, 1934, in Thailand (Diptera: Culicidae). Mosq. News 14: 83Ð 86. Wiley, E. O. 1981. Phylogenetics: the theory and practice of phylogenetic systematics. Wiley, New York. Received 3 June 2005; accepted 9 August 2005.

Downloaded from https://academic.oup.com/jme/article-abstract/42/6/923/886357 by guest on 08 April 2019

Barraud, P. J. 1928. A revision of the culicine mosquitoes of India. Part XXIII. The genus Aedes (sens. lat.) and the classiÞcation of the subgenera. Descriptions of the Indian species of Aedes (Aedimorphus), Aedes (Ochlerotatuts) and Aedes (Banksinella), with notes on Aedes (Stegomyia) variegates. Indian J. Med. Res. 15: 653Ð 699. Belkin, J. N. 1962. The mosquitoes of the South PaciÞc (Diptera: Culicidae), vol. 1. University of California Press, Berkeley, CA. Brummitt, R. K. 2002. How to chop up a tree. Taxon 51: 31Ð41. Brummitt, R. K. 2003. Further dogged defense of paraphyletic taxa. Taxon 52: 803Ð 804. Cantino, P. D., and K. de Queiroz. 2004. PhyloCode: a phylogenetic code of biological nomenclature, version 2b. http://www.ohiou.edu/phylocode. Diggs, G. M., Jr., and B. L. Lipscomb. 2002. What is the writer of a ßora to do? Evolutionary taxonomy or phylogenetic systematics? Sida 20: 647Ð 674. Dyar, H. G. 1918. The male genitalia of Aedes as indicative of natural afÞnities. Insecutor Inscitiae Menstruus 6: 71Ð86. Dyar, H. G. 1928. The Mosquitoes of the Americas. Carnigie Inst. Wash. Publ. No. 387: 1Ð 616. Dyar, H. G., and F. Knab. 1906. The larvae of Culicidae classiÞed as independent organisms. J. N.Y. Entomol. Soc. 14: 169 Ð230. Edwards, F. W. 1917. Notes on Culicidae, with descriptions of new species. Bull. Entomol. Res. 6: 357Ð364. Edwards, F. W. 1921. A revision of the mosquitoes of the Palaearctic region. Bull. Entomol. Res. 12: 263Ð351. Edwards, F. W. 1932. Diptera Fam. Culicidae, 194: 1Ð258. In P. Wytsman [ed.], Genera Insectorum. Desmet-Verteneuil, Brussels, Belgium. Edwards, F. W. 1941. Mosquitoes of the Ethiopian Region. III. ÐCulicine adults and pupae. British Museum (Natural History), London, United Kingdom. Hendrickson, J. A., Jr., and R. R. Sokal. 1968. A numerical taxonomic study of the genus Psorphora (Diptera: Culicidae). Ann. Entomol. Soc. Am. 61: 385Ð392. Hennig, W. 1966. Phylogenetic systematics. University of Illinois Press, Urbana, IL. Howard, L. O., H. G. Dyar, and F. Knab. 1915. The mosquitoes of North and Central America and the West Indies. Systematic description, part I. Carnegie Inst. Wash. Publ. No. 159: 1Ð523. Howard, L. O., H. G. Dyar, and F. Knab. 1917. The mosquitoes of North and Central America and the West Indies. Systematic description, part II. Carnegie Inst. Wash. Publ. No. 159: 525Ð1064. [ICZN]International Commission on Zoological Nomenclature. 1999. International Code of Zoological No-

927