© Entomologica Fennica. 16 June 2006 Identification of adult males and females of Hoplothrips species (Thysanoptera: Tubulifera) known from Norway, and some deductions on their life history Sverre Kobro & Trond Rafoss Kobro, S. & Rafoss, T. 2006: Identification of adult males and females of Hoplothrips species (Thysanoptera: Tubulifera) known from Norway, and some deductions on their life history. — Entomol. Fennica 17: 184–192. The presence of multiple phenological forms has previously made adults of Hoplothrips species difficult to identify. We present a set of diagnostic metric characters by which the 8 species recorded from Norway can be recognised. Some biological interpretations are also given. Sverre Kobro and Trond Rafoss, Bioforsk Plant Health and Plant Protection Division, Høgskoleveien 7, Ås, N-1432 Norway; correspondent author’s e-mail: [email protected] Received 25 November 2005, accepted 24 January 2006 1. Introduction breeding structure. Adults of both sexes of the fungivorous thrips can be wingless for several Dead trees infected with wood-rotting fungi com- generations and will develop wings only in prep- pose habitats for a rich fauna of specialised in- aration for dispersal. Thrips species on fungus in- sects (Hamilton 1978). Many thrips species, half fected dead trees tend to live gregariously and of the suborder Tubulifera, are associated with mate within the habitat, and some of them have dead trees, their bark, dead branches or leaf litter developed subsocial behaviour and strong sexual (Palmer & Mound 1978, Ananthakrishnan 1984). dimorphism (Mound 1974, 1976, Hamilton These thrips feed on fungal spores, hyphae or 1978, Palmer & Mound 1978). The appearance of breakdown products from early stages of fungal oedymerous (major) and gynaecoid (minor) male decay (Mound & Palmer 1983, Ananthakrishnan forms reflects reproductive roles of the males 1984) and Hoplothrips species may be character- (Hamilton 1978, Crespi 1986). Mound (1974, istic for some of the habitats (Ananthakrishnan 1976, 2005) and Hamilton (1978) suggested that 1984). Schliephake and Klimt (1979) noted 11 the production of different morphs was controlled species from Europe of which two have been syn- by the amount or quality of the food ingested by onymised (Vierbergen 2004), while some 130 the larvae. The suggestion was confirmed by species are known for the genus world wide Crespi (1988) in the Nearctic species Hoplothrips (Mound & Walker 1986). karnyi (Hood). Okajima (1987) on the other hand A fungus infected dead log provides support related growth patterns to the colony size, and for numerous generations of Hoplothrips, but Hood (1940) was able to produce either macro- over evolutionary time changes in suitability of pterous or micropterous specimens at will by ma- the habitat have probably selected for differently nipulating environmental conditions. Wingless- sized individuals and specialised patterns of ness, polymorphism and gregariousness are ENTOMOL. FENNICA Vol. 17 • Identification of Norwegian Hoplothrips 185 adaptions to increase the fecundity for fungus for- Thrips were recorded from bark infested with aging thrips on dead wood (Ananthakrishnan Hymenochaete tabacina (Kobro 2003), Pseudo- 1984, Crespi 1986), and at optimal conditions al- spiropes longipilus (Kobro & Solheim 2002), most all individuals are micropterous (Hood Stereum spp. (Kobro 2003), Trichaptum abi- 1940). etinum (Kobro 2001) and T. fuscoviolaceus The extreme polymorphism of Hoplothrips (Kobro 2003). Approximately 0.2–0.3 m² of bark adults (Okajima 1987) is probably the reason containing basidiocarps was collected and depos- why identification literature (Priesner 1928, ited in Berlese funnels for about a week. Thrips 1964, Morison 1949, Stannard 1957, Mound et were stored in AGA (70% ethanol + glycerol + al. 1976, Dyadechko 1977, Schliephake & Klimt acetic acid = 10 + 1 + 1). After storage the speci- 1979, Mound & Walker 1986, Kirk 1996, Mound mens were macerated in hot lactic acid or potas- & Marullo 1996) gives no single key character to sium hydroxide, cleared with clove oil and em- identify adults in the genus. Moreover, the keys bedded in Canada balsam on glass slides. Most are often based on colour shades or general body measurements were made at 500× magnification, characters such as form of the head, body part “a and only fully extended specimens or body parts little longer than” etc. and to a lesser extent on ex- were measured. act metric characters. Some of the older descrip- When possible, 10 adults of each sex of both tions are based upon only one out of several macropterous and micropterous specimens were forms, and large and small male forms of the same selected from different populations and studied. species have been placed into separate genera Additionally, if available, a larger number of (Hood 1955). specimens from single populations of each spe- Eight species of Hoplothrips are recorded cies were examined. All specimens sampled from from Norway (Kobro 2001): Hoplothrips car- the same trunk (usually one sample) were re- pathicus Pelikán, 1961, Hoplothrips corticis (De garded as belonging to the same population. Most Geer, 1773), Hoplothrips fungi (Zetter- specimens examined were collected in Norway. stedt,1828), Hoplothrips pedicularius (Haliday, H. fungi, H. semicaecus and H. ulmi were supple- 1836), Hoplothrips polysticti (Morison, 1949), mented with 38 specimens (mainly macropterous Hoplothrips semicaecus (Uzel, 1895), Hoplo- males, which are rare) borrowed from The Plant thrips ulmi (Fabricius, 1781) and Hoplothrips Protection Service, Wageningen, The Nether- unicolor (Vuillet, 1914). Only H. semicaecus lands and The Natural History Museum, Sen- seems to be rare in Norway (Kobro 2003, Olsen ckenberg, Germany. & Solem 1982). Selection of significant identification charac- Second stage larvae of H. carpathicus, H. ters were based on sample means and standard pedicularius, H. polysticti and H. ulmi can be dis- confidence intervals (95%) calculated for all criminated by means of metric characters (Kobro measured characters. & Rafoss 2001). The initial objective of this study was to find metric diagnostic characters by which adult specimens of Norwegian species assigned 3. Results to the genus Hoplothrips could be recognised. We present and discuss a number of such characters We have measured or evaluated 53 commonly to find the most appropriate to use in an identifi- used characters on a total of 366 adults represent- cation key. We also discuss their life history as in- ing 8 species of Hoplothrips. dicated by our study. The selection of identification characters was based on comparison of confidence intervals cal- culated from the measurements of the characters 2. Material and methods measured (Fig.1). Thirteen characters were chosen according to Bark from trunks or branches of dead trees with their utility, as indicated in figure 1, and used in an visible infestation of wood-rotting fungi were in- identification key for adults of Norwegian vestigated for the presence of tubuliferous thrips. Hoplothrips species. 186 Kobro & Rafoss • ENTOMOL. FENNICA Vol. 17 Fig. 1. Illustration of the procedure for selecting of discriminating characters of adults of Hoplothrips species based on sample means and confidence intervals (95%): length (µm) of mediolaterale seta on pronotum. Other examples of the measurements are Distance between the medioposteromarginal shown in tables 1–4. A more detailed analysis of setae on tergite IX shorter than the width of the variation in the length of sense cones (multi- tubus. Anteroangular setae at the margin of porous olfactory chemoreceptors) on antennal pronotum 3 segment III and the length of mediopostero- 3. Almost always two large sense cones on marginal setae on tergite IX are shown in Figure antennal segment IV 4 2. Posterior prolongation of the fore femora of Almost always more than two sense cones on oedymerous males (Fig. 3), the apices of antennal segment IV 5 medioposteromarginal setae on tergite VIII (Fig. 4. Distance between bases of antennal segments 4), and glandular areas on sternite VIII of male H. I larger than 18 µm. Apical width of antennal carpathicus (Fig. 5) are shown. segment I larger than 40 µm. Medio- posteromarginal setae on tergite IX longer Identification key to the Hoplothrips species re- than 140 µm. Antennal segment VIII longer corded in Norway: than 75 µm Hoplothrips carpathicus Distance between bases of antennal segments 1. Mediolateral setae on pronotum shorter than I less than 18 µm. Apical width of antennal 110 µm 2 segment I less than 40 µm. Mediopostero- Mediolateral setae on pronotum longer than marginal setae on tergite IX shorter than 125 110 µm 6 µm. Antennal segment VIII shorter than 75 2. Distance between the medioposteromarginal µm Hoplothrips semicaecus setae on tergite IX the same as or larger than 5. Distance between bases of antennal segments the width of tubus. Anteroangular setae on the I larger than 16 µm. Width of antennal seg- margin of pronotum Hoplothrips unicolor ment I larger than 36 µm. Antennal segment ENTOMOL. FENNICA Vol. 17 • Identification of Norwegian Hoplothrips 187 Fig. 2. Combined use of the length of sense cones on antennal segment III and the length of mediopostero- marginale seta on abdominal tergite IX for discriminating of adults of Hoplothrips fungi from those of Hoplothrips ulmi and Hoplothrips corticis. H. fungi female macropterous (), H. fungi female micropterous (), H. fungi male macropterous (), H. fungi male micropterous