J. ENTOMOL. SOC. BRIT. COLUMBIA 106, DECEMBER 2009 25

Haliplus leechi Wallis and H. salmo Wallis: a new synonymy and sexual dimorphism in the relative eye separation (Coleoptera: )

REX D. KENNER1

ABSTRACT

Examination of the holotypes, specimens in the type series and material from across their geographical ranges for Haliplus leechi W allis and H. salmo W allis shows that the two names are conspecific; H. salmo is placed as a junior subjective synonym of H. leechi. A sexual dimorphism in the relative eye separation is present in members of this complex, H. canadensis W allis and H. subguttatus Roberts. Preliminary data suggest that this dimorphism may also be present in other haliplid species. This dimorphism should be taken into account in constructing keys for the determination of haliplids.

INTRODUCTION

Haliplus leechi W allis and H. salmo Relative eye separation, the dorsal dis- W allis are very similar structurally. tance between the eyes divided by the head- Haliplus leechi is a widespread (Vondel width, is a character frequently used in keys 2005) species described from material col- for the determination of haliplids (W allis lected in Stanley Park, Vancouver, BC. 1933; Holmen 1987; Vondel 1991, 1993, Haliplus salmo was described from speci- 1995; Vondel and Spangler 2008). In his mens recovered from the stomach of a trout revision of the Nearctic species of Haliplus caught in Jasper, AB and has a more re- Latreille, W allis (1933) used this character stricted distribution (Vondel 2005). W allis in separating three species pairs: H. leechi (1933), in his description of these species, and H. salmo, H. subguttatus Roberts and admitted that —it is possible that one is but a H. salinarius W allis, and H. immaculicollis geographical race of the other“. However, Harris and H. robertsi Zimmermann. Leech he felt that these two taxa could be sepa- (1964) showed that relative eye separation rated based on differences in background was not a useful character in separating the color, maculation, punctulation and relative second pair and noted an apparent sexual eye separation. The results of an investiga- dimorphism in this character. Subsequently tion of the taxonomic status of H. leechi both the second and third pairs were syn- and H. salmo are reported here. onymized (Vondel 1991, 2005).

M ATERIALS AND M ETHODS

The minimum distance between the separation, RIO was calculated by dividing eyes, IO, and the maximum headwidth, IO by HW . HW , were measured using an ocular mi- The holotypes and allotypes of H. leechi crometer on a stereomicroscope (W ild M5, and H. salmo and the paratypes of these Leica MZ12.5). Specimens were positioned species in the Canadian National Collection such that the structure being measured was of (Ottawa, ON) were examined. parallel to the optical plane. Relative eye The relative eye separation, RIO, was meas-

1 Beaty Biodiversity Museum, University of British Columbia, 2212 Main Mall, Vancouver, BC, V6T 1Z4, CANADA, email: kenner@ zoology.ubc.ca

26 J. ENTOMOL. SOC. BRIT. COLUMBIA 106, DECEMBER 2009 ured for both these specimens and a number (San Francisco, CA), D. Kavanagh; Cana- of other specimens previously identified as dian National Collection of Insects (Ottawa, H. leechi or H. salmo. Approximately a ON), Y. Bousquet; J.B. W allis Museum third of the males in these latter series were (University of Manitoba, W innipeg, MB), dissected to allow examination of the geni- R.E. Roughley; James Entomological Col- talia. The dissected genitalia were exam- lection (W ashington State University, Pull- ined while they were floating in liquid, to man, W A), R. Zack; Michigan State Collec- prevent possible distortion due to drying tion of Insects (Michigan State University, and mounting. In addition, RIO was meas- East Lansing, MI), G. Parsons; Museum of ured for specimens identified as H. subgut- Zoology, Invertebrate Section (University tatus and H. canadensis W allis. The speci- of Calgary, Calgary, AB), J.E. Swann; mens examined are in the author‘s collec- Spencer Entomological M useum tion or were borrowed from the following (University of British Columbia, Vancou- museums: California Academy of Sciences ver, BC), K. Needham.

RESULTS AND DISCUSSION

Specific status of H. leechi and H. effect. It appears that there is a continuum salmo. W allis (1933) suggested that H. in elytral maculation, with H. salmo being leechi and H. salmo could be separated by: at one extreme and the putative subspecies i) background color, ii) maculation, iii) H. leechi carteri Leech (1949) at the other. punctulation and iv) relative eye separation. Note that the latter has since been syn- i) The color of preserved specimens is onymized with the nominate subspecies often more a function of their previous (Vondel 2005). The maculation of the head treatment than of the particular species in- and thorax are similarly variable and do not volved (e.g. Kenner 2005). W allis acknowl- provide a reliable character for separating edged this when he suggested that the color H. leechi and H. salmo. of the H. salmo type series may have iii) Examination of a large number of —undergone some change“ due to being specimens in the current complex suggests recovered from the stomach of a trout. It is that the small differences in punctulation the current author‘s experience, based on seen between the two holotypes is within the examination of large numbers of speci- the variation seen in the population as a mens belonging to the H. leechi-H. salmo whole and does not seem sufficient to jus- complex, that the apparent background tify erecting separate species. color is variable but the variation is not iv) W allis gives RIO of the H. leechi and correlated with any other morphological H. salmo holotypes as 0.46 and 0.54, re- character. spectively. The current author‘s remeasure- ii) One of the most obvious differences ment of the holotypes gives a smaller dif- in the two holotypes is in the elytral macu- ference in RIO: 0.48 and 0.51, respectively. lation, with H. leechi having elytral The mean RIOs for the two type series (H. blotches and H. salmo being immaculate. leechi: holotype, allotype and nine para- However, the maculation in H. leechi is types; H. salmo: holotype, allotype and five variable, with some of the paratypes —losing paratypes) are 0.48 (range 0.46œ0.50) and almost all traces of spots on the 0.51 (range 0.50œ0.52), respectively. W allis elytra“ (W allis 1933). On most H. leechi uses RIO < 0.50 (H. leechi) and RIO ≥ 0.50 specimens with very reduced maculation, (H. salmo) in his key; this character does one can still detect the position of at least not even correctly separate all members of some of the elytral blotches, due to infus- the two type series. cate ”halos‘ around the strial punctures in RIO was measured for 142 specimens the appropriate positions. At least one of previously identified as either H. leechi or the H. salmo paratypes shows this same H. salmo; these specimens are from a vari-

J. ENTOMOL. SOC. BRIT. COLUMBIA 106, DECEMBER 2009 27

Figure 1. The frequency distribution of the Relative Eye Separation, RIO, for specimens identi- fied as Haliplus leechi and H. salmo including the type series for each. W allis used RIO <0.50 = H. leechi and RIO ≥0.50 = H. salmo to separate the two species. RIO for the holotypes of H. leechi and H. salmo are 0.48 and 0.51 respectively. ety of localities. A histogram of RIO from nature of all other characters given by W al- these specimens plus the specimens from lis to separate these two taxa, H. salmo is the type series (Fig. 1), gives a unimodal placed as a junior subjective synonym of H. distribution with a mean of 0.49 (range leechi. Haliplus leechi was chosen as the 0.44œ0.54). It appears that there is a contin- senior synonym to maintain stability in the uum in the values of RIO, which suggests literature as it is the much more widely rec- that it is not a good character for separating ognized and cited name and to maintain the these two species. tribute to Hugh Leech intended by W allis This leaves possible differences in the (1933). male genitalia to separate these two species. Sexual dimorphism in the relative eye The apparent differences in W allis‘s draw- separation. In the data from the RIO inves- ings of the aedeagi are largely an illusion tigation discussed above, the smallest val- caused by the fact that W allis did not draw ues of RIO are from male specimens while the basal part of the aedeagus for H. salmo. the largest values are from female speci- If his two drawings are overlaid, one finds mens, although there is extensive overlap. that the differences are on the order of a This suggests that RIO may be sexually di- linewidth. Examination of the mounted morphic. However, since headwidth is pro- genitalia of the two holotypes shows that portional to size and is smallest in males the only significant difference is in the and largest in females, this could actually length of the digitus on the left paramere: be a dependence on size rather than on sex. longer in H. salmo. Due to possible distor- To test for a possible sexual dimorphism tions caused by drying and mounting, it is in the relative eye separation, the specimens not clear if this difference is real. Examina- were sorted by HW and the mean value of tion of a large number of genitalia from RIO for each size group was calculated both putative H. leechi and H. salmo speci- separately for males and females. A paired mens suggests that the difference is not t-test (W hitlock and Schluter 2009) with a constant. null hypothesis of no difference in RIO for Based on the similarity in the genitalia males and females gives a mean RIO(M-N) of the two holotypes and the apparent clinal = -0.013 (t = -3.71, df = 15, P = 0.002). The

28 J. ENTOMOL. SOC. BRIT. COLUMBIA 106, DECEMBER 2009 null hypothesis can be rejected with a high gree of confidence and a sexual dimor- degree of confidence. Since males and fe- phism in RIO is supported. Preliminary tests males of the same size were compared, one on other species suggest that this dimor- can conclude that there is a sexual dimor- phism may occur more widely in haliplids phism in RIO for H. leechi, with females (Kenner unpublished). W hile the difference averaging larger. in RIO for males and females of a given To determine if this is also true in other species are not large, one can end up with a species, similar tests were performed for H. situation, as the current author has, where canadensis and H. subguttatus. For H. males and females go to opposite sides of a canadensis: mean RIO(M-N) = -0.016 (t = - couplet using RIO as the primary character. 2.60, df = 8, P = 0.032); for H. subguttatus: Future keys should take this sexual dimor- mean RIO(M-N) = -0.016 (t = -6.23, df = 10, phism into account when the difference in P < 0.001). In both of these species the null RIO is not large for the taxa being separated. hypothesis can be rejected with a high de-

ACKNOWLEDGEM ENTS

I thank the CanaCol Foundation for a ley, Brad Hubley, Richard Zack and Dick grant that made possible my visit to the Baumann for the gift of specimens and John Canadian National Collection of Insects and Swann for sending me specimens that fi- Y. Bousquet for his generous hospitality nally pushed me into dealing with this is- during my visit; the listed Curators for the sue. loan of specimens in their care; Rob Rough-

REFERENCES

Holmen, M. 1987. The aquatic Adephaga (Coleoptera) of Fennoscandia and Denmark. I. Gyrinidae, Halipli- dae, Hygrobiidae and Noteridae. E.J. Brill/Scandinavian Science Press Ltd., Leiden, Netherlands, pp. 168. Kenner, R.D. 2005. Redescription of Haliplus dorsomaculatus (Coleoptera: Haliplidae) with a new synon- ymy and comments on habitat and distribution. Journal of the Entomological Society of British Columbia 102: 43œ56. Leech, H.B. 1949. Some Nearctic species of Hydradephagid water , new and old (Coleoptera). The Canadian Entomologist 80: 89œ96. Leech, H.B. 1964. Haliplus subguttatus Roberts from W ashington and Oregon (Coleoptera: Haliplidae). The W asmann Journal of Biology 22: 323œ329. Vondel, B.J. van 1991. Revision of the Palaearctic species of Haliplus subgenus Liaphlus Guignot (Coleoptera: Haliplidae). Tijdschrift voor Entomologie 134: 75œ144. Vondel, B.J. van 1993. Revision of the Liaphlus species of the Oriental region excluding China (Coleoptera: Haliplidae). Tijdschrift voor Entomologie 136: 289œ316. Vondel, B.J. van 1995. Haliplidae: Review of the Haliplidae of China (Coleoptera). Pp. 111œ154. In M.A. Jäch and L. Ji (eds.): W ater Beetles of China. Vol. I. W ein: Zoologisch-Botanische Gesellschaft in Österreich and W iener Coleopterologenverein, 410 pp. Vondel, B.J. van 2005. Family Haliplidae (Coleoptera: Adephaga). Pp. 20œ86, in Nilsson, A.N. and Vondel, B.J. van. 2005 Amphizoidae, Aspedytidae, Haliplidae, Noteridae and Paelobiidae (Coleoptera, Adephaga). In: W orld Catalogue of Insects 7: 1œ171. Vondel, B.J. van and P.J. Spangler. 2008. Revision of the Haliplidae of the Neotropical Region including Mexico (Coleoptera: Haliplidae). Koleopterologische Rundshau 78: 69œ194. W allis, J.B. 1933. Revision of the North American species, (north of Mexico), of the Genus Haliplus, La- treille. Transactions of the Royal Canadian Institute 19: 1œ76. W hitlock, M.C. and D. Schluter. 2009. The Analysis of Biological Data. Roberts and Co. Publishers, Green- wood Village, CO. pp. xxxiii, 700.