Biological taxonomy is unique amongst the sciences. In most sciences an obviously false hypothesis will be ignored. In contrast, a new taxon name and the biological hypothesis this name implies must be fully considered by subsequent taxonomists, so long as the name meets the minimal requirements of the relevant Code of Nomenclature. However, establishing that the author of a new taxon has published incorrect data, or drawn conclusions that are at variance with pre-existing biological knowledge, can involve more extensive research than that involved in the faulty publication. Commenting on one recent example, Mound and zur Strassen (2001) noted the absence of biological evidence to support the description by Johansen and Mojica-Guzman (1999) of a large number of new species of Scirtothrips collected mainly from two tree crops in Mexico. Subsequently, extensive field studies and DNA analyses have confirmed that only one species of Scirtothrips occurs in Mexico and Guatemala on Avocado (Persea americana), and examination of the holotypes of five of the "new species" has resulted in their formal synonymy (Hoddle et al. 2008). These extensive, and expensive, revisionary studies were required because of the quarantine significance of so many apparent pest species on avocados. The 18 Scirtothrips species described from Mexico on mango (Mangifera indica), a tree crop not native to the Americas, will require similar extensive studies.

Taxonomy requires a combination of good technical methods and critical scholarship. For thrips taxonomy this involves the examination of well-prepared microscope slides, using good optical apparatus, and the preparation of skilled illustrations and descriptions, whilst taking into consideration published observations from previous workers, and placing all this within the context of our understanding of the biology of the insects. The large number of synonyms, more than 20% of available names worldwide, produced by workers on this group of insects (Gaston and Mound 1993) has arisen through repeated failure to recognize one or more of the above disciplines and procedures. This paper records similar problems that have arisen in recent years among several taxonomic publications on Thysanoptera from Costa Rica.

This genus was erected with Anaphothrips sudanensis Trybom as type species. However, as indicated in the world list of Thysanoptera (Mound 2005), this species, in the form of one of its 10 recognized synonyms, Neophysopus medioflavus Schmutz, is the type species of the genus Neophysopus Schmutz. Thus Nakaharathrips is clearly a synonym of Neophysopus because they share the same type species. According to the generic key and the data matrix published with Nakaharathrips, the species A. sudanensis is uniform in colour. This is incorrect, because this is one of the most remarkably bicoloured species in the Thripidae, with the median abdominal segments almost white but the head and thorax and the terminal abdominal segments dark brown (see website, PADIL). This bicoloured condition has been emphasized by many previous authors, as indicated by some of the synonyms of sudanensis, such as citricinctus, bicolor, and flavicinctus as well as medioflavus. All of this information is available on the web (Mound 2005), as well as in the formal taxonomic literature. A further problem in the published data matrix is that both Nakaharathrips and Anaphothrips are scored as having ocelli reduced, whereas macropterae of both A. obscurus and A. sudanensis have well-developed ocelli (see PADIL). In Thysanoptera, development of ocelli is closely correlated with wing length; the use of these in a phylogenetic analysis as independent characters is therefore questionable.

The systematic position of Neophysopus (=Nakaharathrips) has been undisputed for many years, either as a subgenus or as a synonym of Anaphothrips (Priesner 1949, Stannard 1968). The decision (Retana-Salazar 2007) to segregate into one genus those species that have antennal segments VII and VIII fully distinct from VI, together with those that have segment VI partially divided by an incomplete oblique suture, and to place into a second genus those species in which segment VI is completely undivided, is not supported by any additional morphological or biological evidence. The conclusion seems to be derived from a mathematical paper (Retana-Salazar and Retana-Salazar, 2004) in which it is claimed that a difference such as the number of antennal segments that can be scored must indicate that different taxa are involved. If this logic were to be accepted, then it would be essential to explain why "segment VI fully divided + segment VI with a partial division" constitutes an autapomorphy distinct from "segment VI not divided". However, the available biological evidence is contrary to this phenetic claim, with the number of antennal segments varying in some species between individuals within populations, and some individuals having a different number of segments on the left and right antennae (Palmer 1992). Moreover, there are several genera of Thripidae in which different species have a different number of antennal segments (Sakimura and O’Neill 1979, zur Strassen 2003: 259, 269). Given the many character states, including biological attributes (Nakahara 1995) that are shared among the 52 species currently treated in Anaphothrips, there is little justification in transferring a few of them to a different genus. The claim that having eight segments is the plesiomorphic condition and nine segments the apomorphic condition is also questionable, because the presence of nine antennal segments is characteristic of most basal-clade Thysanoptera (Mound et al. 1980). We conclude that Neophysopus (=Nakaharathrips) should continue to be treated as a synonym of Anaphothrips.

This genus was erected for a single new species, N. garitai Retana-Salazar, although the type specimens of this species had previously been studied and identified by Mound and Marullo (1996) as Aurantothrips orchidaceus (Bagnall). The new genus and species was characterized on two structural features: antennal segments VII and VIII fused to produce a 7-segmented antenna, and a spherical structure present in the abdomen that was interpreted as a spermatheca. However, fusion of the terminal two antennal segments is not unique to the few specimens listed from Costa Rica. Similar fusion has been observed on either the left or right antenna of three females within a long series of Aurantothrips orchidaceus collected from orchids in Colombia, and one male of this species from Britain has both antennae 7-segmented (specimens in the Natural History Museum, London). Thus the first of the character states used to distinguish the new genus is merely intraspecific variation. Similarly, the spherical structure in the abdomen that is illustrated (Retana-Salazar 2007) is clearly visible in many specimens of Aurantothrips orchidaceus, including the Lectotype female. We therefore conclude that Nicolemma is a synonym of Aurantothrips Bhatti, and that N. garitai is a synonym of A. orchidaceus (Bagnall).

This genus was erected (Bhatti 1978: 90) for a single Neotropical species from orchids, Anaphothrips orchidaceus Bagnall, described originally from greenhouse specimens in Europe. Sakimura (1967) examined 24 females and nine males and concluded, on the basis of weak differences in colour and chaetotaxy, that two subspecies could be distinguished. Mound and Marullo (1996) reviewed the evidence produced by Sakimura, but reported further variation in colour and structure in specimens from Colombia and Brazil that cast doubt on the existence of two biological entities. The subsequent decision (Retana-Salazar 2007) that the two subspecies are distinct species was not supported by any evidence of having studied any specimens. Accompanying the decision were two short paragraphs copied (but translated) from Mound and Marullo (1996) that detailed the differences between the specimens available to Sakimura. However, the subsequent paragraph by the same authors, discussing the variation in and between samples from Colombia and Brazil, was neither translated nor discussed. Again, the logic seems to be derived from Retana-Salazar and Retana-Salazar (2004), in which the authors appear to claim that if differences can be observed and scored mathematically then these differences must be interpreted as indicating that different taxa are involved, regardless of any biological data. However, any such a mathematical system will be biased if only selected data is considered.

This genus was erected for two species, a new species, Q. sotoi from Costa Rica, as type species and Pseudothrips longiceps Hood based on a single female from Brazil. The generic definition was not accompanied by any comparison with other genera, being distinguished only in a key to five genera by tergite VII lacking a marginal comb of microtrichia. This state was contrasted with species that are known to exhibit a variable number of marginal microtrichia laterally on this tergite (Stannard, 1951, Palmer and Mound 1985, Mound and Marullo 1996). In view of this variation, the genus Quirosiella is here regarded as a synonym of Psectrothrips Hood, to which genus P. longiceps was transferred by Palmer and Mound (1986).

A cladogram provided by Retana-Salazar (2000) involved 11 of the 17 nominal species in the genera Psydrothrips Palmer and Mound, Psectrothrips Hood (= Lacandonithrips Johansen) and Pseudothrips Hinds, but no data matrix was provided to enable the suggested relationships to be assessed, although there was a table indicating the characters used. Moreover, no indication was given of which of the 11 species the author did, or did not, personally study; for several species the data evidently came from published literature. One species not included in the cladogram for which specimens were available was a Costa Rican species named after its collector, Pseudothrips retanae Mound & Marullo. The cladogram for the 11 species placed the two species of Quirosiella as sister to two species of Pseudothrips, and this pair of genera as sister to five species of Psectrothrips. This pattern of relationships is particularly surprising because the males of Pseudothrips species exhibit a remarkable character that occurs in only six other genera of the worldwide family Thripidae, each of these being from South America (Mound and Marullo 1996, De Borbón 2008). These genera include Psydrothrips but not Psectrothrips, and the character is the presence of a glandular opening on the antecostal region of sternite III, whereas the normal condition in this family is for a transverse porose area medially on several sternites posterior to the antecostal ridge (zur Strassen 2003, Tyagi et al. 2008).

The generic name Linneothrips Johansen and Mojica (1996) should be considered a synonym of Pseudothrips Hinds. Contrary to Retana (2000: 57), who sent the same species for description to both Mexico and Europe, the paper by Johansen and Mojica although listed as 1995 was not published until 30.viii.1996, as confirmed by the editor of the journal; the publication data of the Mound and Marullo book was 4.iii.1996.

This genus was distinguished from Karnyothrips Watson, with Hindsiana melaleuca Bagnall as type species (although quoted incorrectly as Karnyothrips melaleucus). The authors of this new generic name indicated that they had studied specimens of K. melaleucus and K. ochropezus. However, they claimed to include a total of 24 species in this new genus, although they did not provide the names of any of these species nor any evidence of having studied any of them. Vargasia was distinguished on the basis that the two species studied have only three sensoria on antennal segment IV, and this was contrasted with Karnyothrips flavipes, the type species of Karnyothrips, which the authors claimed to have four sensoria on this segment. However, specimens of both K. flavipes and K. melaleucus have been studied in the collections of the Natural History Museum, London, and the Australian National Insect Collection, Canberra, and the number of sensoria on segment IV varies from three to four in both species, within and between populations (Okajima 2006: 397). Both species are known to have a similar biology as predators of scale insects, and they are essentially similar in structure. Since there is neither morphological nor biological evidence for supporting the new genus, Vargasia is here considered a synonym of Karnyothrips. In describing this new genus, the authors also returned two species to Apterygothrips from Karnyothrips, but presented no evidence of examining either species, nor yet of examining any species of Apterygothrips for reference.

This new genus was erected for two species, with Zygothrips americanus Hood as type species (but quoted incorrectly as Karnyothrips americanus). The authors did not indicate that they had examined any specimens of either species; apparently relying on data in the key to species provided by Mound and Marullo (1996). The genus was distinguished from Karnyothrips on two character states. First, it was claimed that antennal segment IV bears only three sensoria rather than four, but as indicated above this is variable in the type species of Karnyothrips, and moreover K. americanus has been examined at the Natural History Museum, London and actually bears only two sensoria on this segment. Second, K. americanus has the anteromarginal setae on the pronotum well developed instead of reduced, although the length of this pair of setae is variable among other species of Karnyothrips. Given this variation, the known similarities in biology, and the absence of any further supporting morphological evidence, the generic name Willeia is here regarded as a further synonym of Karnyothrips.

This new genus was erected for a single species, Haplothrips (Xylaplothrips) sonorensis Stannard (but quoted incorrectly as Karnyothrips sonorensis). The authors stated that they had examined specimens of this North American species in the collection of A.P. Retana-Salazar, but did not indicate how these specimens were identified nor the locality from which they were collected. If the specimens came from Costa Rica then this would be the first published record from Central America. The genus was distinguished from Karnyothrips and also from Willeia because antennal segment IV bears only two sensoria. However, contrary to the statement given by the authors, antennal segment IV of K. americanus also bears only two sensoria, and the form of the other antennal segments in this species and in K. sonorensis is very similar, as indicated by Stannard (1968: 431). There is thus no support for placing this one species in a separate genus, and Dioclesianothrips is therefore considered a synonym of Karnyothrips. The authors failed to include their new name in their abstract, but they did include a nomen nudum "Aguilaria" as a new genus; presumably one name replaced the other during manuscript preparation.

Jironiella
Retana-Salazar & Soto-Rodriguez, 2007
]]> This new genus was erected for a single new species, J. saidi. The species was based on specimens collected from Cyperaceae in Costa Rica, and the authors claimed that it could be distinguished from Haplothrips species through the absence in the head of a maxillary bridge. With the exception of this character, the description agrees closely with specimens of Haplothrips graminis Hood, a widespread species in Costa Rica and Colombia on Cyperaceae and Poaceae. Subsequently, one paratype female of J. saidi in the collections of the Natural History Museum, London, has been examined under phase contrast illumination, and remnants of a maxillary bridge are clearly visible. In H. graminis the maxillary stylets are particularly wide apart, and the maxillary bridge is therefore unusually long and slender. This structure is thus very easily displaced during slide preparation; certainly it is not visible in all of the slides of this species in collections both in London and in Canberra. As a result, J. saidi is here placed into synonymy with the common Central American species, H. graminis, and Jironiella becomes a further synonym of the worldwide genus Haplothrips Amyot and Serville.

Gynaikothrips garitacambroneroi
Retana-Salazar, 2006

This species was described from 13 females and two males, collected in Costa Rica. The plant species from which these specimens were collected is not clearly stated in the original description, but judging from Garito-Cambronero and Lizano-Fallas (2006) it was Ficus benjamina. The thrips was distinguished from Gynaikothrips uzeli (Zimmerman) by the greater length of the pronotal posteroangular setae. However, the measurements given fall within the range of variation found in samples of G. uzeli from various parts of South East Asia including the Maldive Islands. Moreover, G. uzeli seems to be increasing in distribution in the Caribbean region and southern States of the USA on F. benjamina, presumably through the activities of the horticultural trade (Boyd and Held 2006). One female paratype of G. garitacambroneroi, deposited in the Natural History Museum, London, has been examined, and as a result the species is here considered a synonym of the Asian species G. uzeli. The illustrations of the male "genitalia" are based  on the unexpanded epiphallus, an inflatable sac  whose shape varies with the extent of inflation. The actual pseudovirga was not illustrated, and although the pseudovirga can be useful in recognizing some Haplothripini (Mound and Minaei 2007) its structure has not been demon strated to be useful for distinguishing species in the majority of leaf-feeding Phlaeothripinae. The number and length of setae on the apex of the parameres is not stable between individual males of Gynaikothrips species examined from various parts of Asia and Australia.

Nomenclatural summary

Anaphothrips Uzel, 1895: 142
    Nakaharathrips Retana-Salazar, 2007: 329. syn.n.

Aurantothrips Bhatti, 1978: 90
    Nicolemma Retana-Salazar, 2007: 324. syn.n.

Psectrothrips Hood, 1937: 262
    Quirosiella Retana-Salazar, 2000: 55. syn.n.

Psectrothrips longiceps (Hood); Hood, 1954: 211
    Quirosiella sotoi Retana-Salazar, 2000: 54. syn.n.

Pseudothrips Hinds, 1902: 146
    Linneothrips Johansen & Mojica, 1996: 59. syn.n.

Pseudothrips retanae Mound & Marullo, 1996: 181
    Linneothrips bicolor Johansen & Mojica, 1996: 62. syn.n.

Haplothrips Amyot & Serville, 1843: 640.  
    Jironiella Retana-Salazar & Soto-Rodriguez, 2007: 632. syn.n.

Haplothrips graminis Hood, 1921: 69
    Jironiella saidi Retana-Salazar & Soto-Rodriguez, 2007: 633. syn.n.

Karnyothrips Watson, 1923: 23
    Vargasia Retana-Salazar & Soto-Rodriguez, 2007: 631. syn.n.
    Willeia Retana-Salazar & Soto-Rodriguez, 2007: 632. syn.n.
    Dioclesianothrips Retana-Salazar & Soto-Rodriguez, 2007: 632. syn.n. ]]>     "Aguilaria" Retana-Salazar & Soto-Rodriguez, 2007: 635. syn.n.

Resumen

Presentamos datos para apoyar nuestro argumento de que varios artículos recientes sobre los Thysanoptera de Costa Rica se han visto afectados por procedimientos técnicos insatisfactorios, incluyendo el no reconocer la variación estructural intraespecífica. Presentamos nueve sinonimias en los tisanópteros de Costa Rica: nueve a nivel de género y cinco a nivel de especie.

Palabras clave: variación, taxonomía, identificación, relaciones sistemáticas.

References

Bhatti, J.S. 1978. Systematics of Anaphothrips Uzel 1895 sensu latu and some related genera. Senck. Biol. 59: 85-114.        [ Links ]

Boyd, D.W. & Held, D.W. 2006. Androthrips ramachandrai (Thysanoptera: Phlaeothripidae): an introduced thrips in the United States. Fla Ent. 89: 455-458.         [ Links ]

De Borbón, C.M. 2008. Desertathrips chuquiraga gen. et sp.n. (Thysanoptera, Thripidae) from Argentina. Zootaxa. 1751: 25-34.         [ Links ]

Garita-Cambronero, J. & Lizano-Fallas, J. 2006. Determinación de dimorfismo sexual de Gynaikothrips garitacombroneroi (Thysanoptera: Phlaeothripidae) inductor de agallas en Ficus benjamina. Métodos Ecología Sistemática 1: 9-13         [ Links ]

Gaston K.J. & Mound, L.A. 1993. Taxonomy, hypothesistesting and the biodiversity crisis. Proc. Roy. Soc., London B251: 139-142.         [ Links ]

Hoddle, M.S., Mound, L.A., Rugman-Jones, P.F. & Stouthamer, R. 2008. Synonomy of five Mexican Scirtothrips species (Thysanoptera: Thripidae) described from Avocados (Persea americana). Fla Ent. 91: 16-21.         [ Links ]

Johansen, R.M. & Mojica-Guzman, A. 1995. A review of the Tribe Humboldthripini Johansen (Insecta, Thysanoptera: Thripidae). Folia entomologica Mexicana 93 [1996]: 39-70. [published 30.viii.1996]         [ Links ]

Johansen, R.M. & Mojica-Guzman, A. 1999. The genus Scirtothrips Shull, 1909 (Thysanoptera: Thripidae, Sericothripini), in Mexico. Fol. Ent. Mexicana 104: 23-108.         [ Links ]

Mound, L.A., Heming, B.S. & Palmer, J.M. 1980. Phylogenetic relationships between the families of recent Thysanoptera. Zool. J. Linn. Soc. London. 69: 111-141.         [ Links ]

Mound, L.A. & Marullo, R. 1996. The Thrips of Central and South America: An Introduction. Mem. Entomol. International 6: 1-488. [published 4.iii.1996]         [ Links ]

Mound, L.A. & Minaei, K. 2007. Australian insects of the Haplothrips lineage (Thysanoptera – Phlaeothripinae). J. Nat. Hist. 41: 2919-2978.         [ Links ]

Mound, L.A. & zur Strassen, R. 2001. The genus Scirtothrips (Thysanoptera: Thripidae) in Mexico: a critique of the review by Johansen & Mojica-Guzmán (1998). Fol. Ent. Mexicana 40: 133-142.         [ Links ]

Nakahara, S. 1995. Review of the Nearctic species of Anaphothrips (Thysanoptera: Thripidae). Ins. Mundi. 9: 221-248.         [ Links ]

Okajima, S. 2006. The Suborder Tubulifera (Thysanoptera). The Insects of Japan 2: 1-720. The Entomological Society of Japan, Touka, Fukuoka, Japan.         [ Links ]

Palmer, J.M. 1992. Thrips from Pakistan to the Pacific: a review. Bull.Br. Mus. Nat. Hist. (Entom.) 61: 1-76.         [ Links ]

Palmer, J.M. & Mound, L.A. 1985. New World Thripidae (Thysanoptera) with nine-segmented antennae. Zool. J. Linn. Soc. 84: 181-194         [ Links ]

Priesner, H. 1949. Genera Thysanopterorum. Keys for the identification of the genera of the order Thysanoptera. Bull. Soc. Roy. Ent. d’Egypte. 33: 31-157.         [ Links ]

Retana-Salazar, A.P. 2000. Revision y filogenia del grupo generico Pseudothrips (Thysanoptera: Thripidae). Brenesia 54: 51-62.         [ Links ]

Retana-Salazar, A.P. 2006. Variación morfológica del complejo Gynaikothrips uzeli-ficorum (Phlaeothripidae: Tubulifera). Métodos Ecología Sistemática 1: 1-9.         [ Links ]

Retana-Salazar, A.P. 2007. Los tisanópteros del grupo genérico Anaphothrips (Thysanoptera: Thripidae), con énfasis en América Central. Rev. Biol. Trop. 55: 321-333.         [ Links ]

Retana-Salazar, A.P. & Retana-Salazar, S. 2004. Hacia una lógica simple en la determinación dc grupos biológicos: la especie y los grupos supraespecíficos. Rev. Biol. Trop. 52: 19-26.         [ Links ]

Retana-Salazar, A.P. & Soto-Rodriguez, G.A. 2007. Revisión taxonómica del grupo Haplothrips-Karnyothrips (Thysanoptera: Phlaeothripidae). Rev. Biol. Trop. 55: 627-635.         [ Links ]

Sakimura, K. 1967. Redescriptions of Anaphothrips orchidaceus and A. orchidearum (Thysanoptera: Thripdae). Fla Ent. 50: 89-97.         [ Links ]

Sakimura, K. & O’Neill, K. 1979. Frankliniella, redefinition of genus and revision of minuta group of species (Thysanoptera: Thripidae). Tech. Bul. U.S. Dept Agric. 1572: 1-49         [ Links ]

Stannard, L.J. 1951. The second species of Psectrothrips (Thysanoptera: Thripidae). Entomological News. 62: 245-249.         [ Links ]

Stannard, L.J. 1968. The Thrips, or Thysanoptera, of Illinois. Bull. Illin. Nat. Hist. Surv. 29: 213-552.         [ Links ]

Tyagi, K., Kumar, V. & Mound, L.A. 2008. Sexual dimorphism among Thysanoptera Terebrantia, with a new species from Malaysia and remarkable species from India in Aeolothripidae and Thripidae. Insect Syst. Evol. (in press).         [ Links ]

zur Strassen, R. 2003. Die Terebranten Thysanopteren Europas und des Mittelmeer-Gebietes. Die Tierwelt Deutschlands 74: 1-271.         [ Links ]

Internet references

Mound, L.A. 2005. Thysanoptera (Thrips) of the World – a checklist. http://www.ento.csiro.au/thysanoptera/worldthrips.html (downloaded 7 March 2008).         [ Links ]

PADIL – Pests and Diseases Images Library <http://www.padil.gov.au/> (downloaded 7 March 2008)         [ Links ]
]]> 1978 59 85-114 2006 89 455-458 2008 1751 25-34 2006 1 9-13 1993 B251 139-142 2008 91 16-21 1995 93 1996 1996 39-70 1999 104 23-108 1980 69 111-141 1996 6 1-488 2007 41 2919-2978 2001 40 133-142 1995 9 221-248 2006 2 1-720 1992 61 1-76 1985 84 181-194 1949 33 31-157 2000 54 51-62 2006 1 1-9 2007 55 321-333 2004 52 19-26 2007 55 627-635 1967 50 89-97 1979 1572 1-49 1951 62 245-249 1968 29 213-552 2008 2003 74 1-271 2005 PADIL