Boreonectes gen. n., a new genus for the Stictotarsus griseostriatus (De Geer) group of sibling species (Coleoptera: Dytiscidae), with additional karyosystematic data on the group

A new genus, Boreonectes gen. n., is erected for the seven S. griseostriatus-group species as well as eight other American ones listed by Nilsson and Angus (1992) as comprising the Stictotarsus griseostriatusgroup as DNA work by Ribera (2003) shows this arrangement to be untenable, and no other genus-group name has been found to be available. Chromosomal investigations have shown that B. griseostriatus (De Geer, 1774) occurs as far west as the Little St Bernard Pass in the Alps, with B. alpestris (Dutton et Angus, 2007) in the Italian Gran Paradiso National Park to the south of this. B. multilineatus (Falkenström, 1922) is recorded from the Pyrenees for the fi rst time, and B. ibericus (Dutton et Angus, 2007) is shown to be present in the Moyen Atlas of Morocco.


INTRODUCTION
The Generic Placement.One of the most unsatisfactory features of the Stictotarsus griseostriatus (De Geer, 1774) species-group is its placement within the genus Stictotarsus Zimmermann, 1919. Nilsson andAngus (1992), in an attempt to sort out confusion concerning the classifi cation of the Dytiscid genera related to Deronectes Sharp, 1882, placed those species not in Deronectes, but with parameres lacking the apical hook, in Stictotarsus.They divided Stictotarsus into three species groups, the S. duodecimpustulatus (Fabricius, 1792) group (true Stictotarsus) and the S. griseostriatus and S. roffi (Clark, 1862) groups.These were left as informal groupings as no synapomorphies were found for the S. griseostriatus group.This arrangement at present stands, but has never been popular with coleopterists (e.g.Mazzoldi and Toledo, 1998).Then Ribera (2003) published a paper giving the results of mitochondrial DNA analyses which rendered Nilsson and Angus' arrangement untenable.The Stictotarsus duodecimpustulatus group was shown to come out on phylograms near Nebrioporus Regimbart, 1906 andScarodytes Gozis, 1914, while S. griseostriatus (sensu lato), along with S. striatellus (LeConte, 1852) (also in the S. griseostriatus group) and S. roffi came out between two sections of Oreodytes R.B. Angus Comp. Cytogenet., 2010 4(2) Comparative Cytogenetics Seidlitz, 1887.Trichonectes otini Guignot, 1941, included  If the present unsatisfactory classifi cation of the S. griseostriatus group is to be remedied, then at present there appear to be two alternative courses of action.One is to place the S. griseostriatus group (and probably also the S. roffi group) within an expanded Oreodytes.The other is to erect a new genus for the S. griseostriatus group of species.Neither course is without its pitfalls.Oreodytes is not regarded as one of the Deronectes group of genera (Deronectina), differing, among other ways, in having contact between the median keel of the metasternum and the mesosternal fork, and sucker-hairs on the anterior and middle tarsi of the males.Such hairs are widespread among Hydroporini but lacking in Deronectina (Nilsson and Angus, 1992).To place the S. griseostriatus group within Oreodytes would almost certainly simply result in a short-term change of name, to be altered later.
The second course of action, to be adopted here, is to erect a new genus for the S. griseostriatus group of species.The pitfalls here are fi rstly that the exact limits of the new genus must remain to be established, and secondly that, assuming the DNA phylograms are correct, Oreodytes becomes a paraphyletic genus.The limits of genera are always subject to revision as more knowledge becomes available, so that objection does not seem overriding.As to Oreodytes, if more work is needed to sort out its composition, that is a separate task awaiting further research.The new genus may now be described.
Boreonectes gen.n.Type species: Dytiscus griseostriatus De Geer, 1774.Hydroporine beetles with the Deronectes group characters of the median keel of the metasternum not touching the mesosternal fork and absence of sucker-hairs on the fore and middle tarsi of the males.Parameres more or less similar, of the Hydroporus pattern, without any apical hook or other sclerotised apical section.Ventral surface of adults densely and fi nely punctate, without visible reticulation between the punctures.Elytra densely and fi nely punctured, without reticulation between the punctures (except in Hydroporus panaminti Fall, 1923, q.v.), the primary series of larger punctures indistinct or very prominent, depending on species.Ventral surface of head behind the eyes shining, with superfi cial reticulation.Pronotum without distinct lateral grooves or depressions.Larvae with well-developed swimming-hairs on the legs, these in most cases extending on to the tarsi, including those of the anterior legs.Urogomphi of second and third instar larvae with numerous secondary setae on the basal segment, this very long, 90% + of the total length of the urogomphus in third instar larvae.
Differential diagnosis.Median keel of metasternum not touching metasternal fork; fore and mid tarsi of males without suckerhairs; parameres of the Hydroporus pattern, without apical hook or other sclerotised apical portion; ventral surface of head behind eyes shining, with only superfi cial reticulation (closely punctate in Stictotarsus and the S. roffi group); ventral surface of body fi nely and densely punctate, without visible reticulation between the punctures; urogomphi of second and third instar larvae with numerous secondary setae (absent in Deronectes).
The etymology of the name is from the Greek Boreas, the North, and Nectes, a Included species: Boreonectes may be expected to include most of the species listed by Nilsson and Angus in the S. griseostriatus group, with the proviso that as new information becomes available species may be excluded, as has happened with Trichonectes otini.Based on Nilsson and Angus (1992), this gives the following 15 species: The B. griseostriatus group -Dytiscus griseostriatus De Geer, Potamodytes multilineatus Falkenström, Potamonectes macedonicus Georgiev, 1959, Stictotarsus alpestris Dutton et Angus, 2007, S. ibericus Dutton et Angus, 2007, S. inexpectatus Dutton et Angus, 2007and S. riberae Dutton et Angus, 2007 (seven species), plus Hydroporus aequinoctialis Clark, 1862, H. coelamboides Fall, 1923, Deronectes dolerosus Leech, 1945, Hydroporus expositus Fall, 1923, H. funereus Crotch, 1873, H. panaminti Fall, 1923, Deronectes spenceri Leech, 1945 and Hydroporus striatellus LeConte, 1852.The inclusion of H. panaminti may be regarded as tentative because the elytral punctation of this species, though fi ne, is more diffuse than in the others, and the space between the punctures is fi nely reticulate.
The phylogram given by Ribera, 2003 on p 484 associates the S. roffi group with the griseostriatus group, though slightly separate from it.These beetles were shown by Nilsson and Angus to have a somewhat different morphology, with contact, apparently secondary, between the median keel of the metasternum and the mesosternal fork, and the ventral surface of the head, behind the eyes, rugosely punctate, so for the moment it seems better to exclude them from Boreonectes.The general appearance of the S. roffi group species is distinctly different from those included in Boreonectes.It should be noted that Boreonectes is an overwhelmingly Nearctic genus, with only the griseostriatus group extending into the Old World.
An ongoing programme of karyosystematic investigation of the Boreonectes griseostriatus group of sibling species (Dutton, Angus, 2007;Angus, 2008 andAngus, 2010) is clarifying the taxonomic status and distributions of the various species involved.The investigations Switzerland, Valais, Lake by the Illsee 1 France, Col du Petit Saint-Bernard, pool at 2200 m a.s.l., SW of la Baraque des Douaniers (Fig. 1).

2
Italy, Colle del Piccolo San Bernardo, pools between the French frontier and the NE edge of Lac Verney (Fig. 1).

RESULTS
Mitotic chromosomes, arranged as karyotypes, are shown in Fig. 3.

B. griseostriatus (De Geer, 1774).
Swedish reference material (fi rst used by Dutton and Angus (2007) as Fig. 2, a,  b) is shown in Fig. 3, a (plain) and Fig. 3, b (C-banded).A mid-gut nucleus from a female from the Little St Bernard Pass (French side) is shown in Fig. 3, c (plain) and Fig 3, d (C-banded).This material shows no chromosomal differences from the northern reported here extend the known distributions of B. griseostriatus (De Geer, 1774) and B. alpestris (Dutton et Angus, 2007) in the Alps, and record B. multilineatus (Falkenström, 1922) from the French Pyrenees and B. ibericus (Dutton et Angus, 2007) from the Moyen Atlas of Morocco.

MATERIAL AND METHODS
The material used for chromosome analysis is listed in Table 1, and the localities are shown on the maps in Figs 1 and 2. The methods are as described by Dutton and Angus (2007).The material is in the author's collection in the

Comp. Cytogenet., 2010 4(2)
Comparative Cytogenetics and Alpine karyotypes obtained from this species.The single male from near the Illsee (Switzerland) is not fi gured as its locality is very near the Le Louché localities from where material was fi gured by Angus (2010).

B. multilineatus (Falkenström, 1922).
Plain and C-banded karyotypes of B. multilineatus from Clatteringshaws Loch, Scotland (fi rst used by Dutton and Angus (2007) as Fig. 2, d, e) are shown in Fig. 3, e, f.Fig. 3, g, h shows a male karyotype from the Pyrenees, plain and C-banded, and Fig. 3, i, j shows a female karyotype from the Pyrenees, again plain and C-banded.These karyotypes appear identical in all respects to the Scottish material, and record B. multilineatus for the fi rst time in central Europe.
The discovery of B. multilineatus in central Europe raises the question of its relationship to B. alpestris.Dutton and Angus (2007) noted that the only obvious difference between the karyotypes of the two species is the presence of one additional pair of very small autosomes in B. multilineatus.However, they considered it wisest to regard the two as separate species in view of their very different distributions and the distinctly larger male genitalia of B. alpestris.The discovery of B. multilineatus in the Pyrenees narrows the difference between the distributions of the two species, and, along with the new material of B. alpestris (see below) allows more careful comparison of their genitalia.Parameres of Swedish and Scottish B. multilineatus are shown in Fig. 4, e, f, while a Pyrenean specimen is shown in Fig. 4 g.The Pyrenean specimen appears conspicuously small.But the beetle itself is only 4.15 mm long, as against 4.4 mm for the Scottish male and 4.3 for the Swedish one.In fact, the Scottish paramere, even allowing for its somewhat distorted appearance (due to drying), does not appear any bigger than the Pyrenean one.The parameres of the two Pyrenean males studied both appear rather more elongate than those of Scottish and Swedish specimens.Parameres of two B. alpestris from the Colle del Nivolet are shown in Fig. 4, c, d.These parameres, mounted fresh

B. alpestris (Dutton et Angus, 2007).
The karyotype of a male paratype, from Falcade, Italy, fi gured by Dutton and Angus (2007) as Fig. 2 f, is shown in Fig. 3 k.Fig. 3 l, m shows a karyotype from a male from the Colle del Nivolet, plain and C-banded.This is the fi rst time C-banded material of B. alpestris has become available.The extent and arrangement of the C-banding is very similar to that shown by B. multilineatus (Fig. 3, e-j).Comparison of the parameres of B. alpestris and B. multilineatus is given above, and confi rms the view that B. alpestris is a species distinct from B. multilineatus.

B. ibericus (Dutton et Angus, 2007).
The C-banded karyotype of a French male paratype originally fi gured by Dutton and Angus (2007) as Fig. 2 m, is shown in Fig. 3 n.Fig. 3 o shows a complete female karyotype from the Col du Zad, Morocco, while an incomplete male karyotype from the same locality is shown, plain and C-banded, in Fig 3 p, q.The suggested positions of the missing chromosomes are marked with black squares in Fig. 3 p.The reason for this incomplete karyotype is that when the fi rst batch of preparations were made, in late April and early May, room temperature was rather low (about 17 o C.) and I was unable to make a satisfactory alteration to my C-banding protocol for these beetles, even though I was successful with others (e.g.Aphodius (Scarabaeoidea)).By the time room temperature had reached the low 20s (in early June) the remaining two beetles, both male, though still alive, were in poor condition with very little cell division in the mid gut.Nevertheless, C-banding was now successful.The pattern of the C-banding is interesting, and agrees with the results obtained from Corsican material (Angus, 2010) in having prominent C-bands on autosome pairs 2, 3, 6, 7 and 13, as well as on the X chromosome.The French specimen has an additional pair of C-bands on autosome pair 8.
Two further features of the Moroccan material are of note.The material comprised eight beetles, of which seven gave defi nitive karyotypes.Of these, six are heterozygous for the fusion-fi ssion polymorphism involving

Comp. Cytogenet., 2010 4(2)
Comparative Cytogenetics autosome pairs 1 and 24, while the seventh specimen has the chromosomes homozygous unfused.The second feature of note is the conspicuously large size of the beetles -the three males range in length from 4.4 -4.5 mm, and the fi ve females from 4.5 -4.9 mm.Dutton and Angus (2007) give the size range for B. ibericus (both sexes) as 3.8 -4.7 mm.There is no feature of the Moroccan karyotypes giving any suggestion that they are not the same as those of European material.

DISCUSSION
In terms of geographical distribution, the most surprising discovery reported here is the presence of B. multilineatus in the Pyrenees.From the totality of its distribution, the expected species was B. ibericus, which may well occur in other parts of the mountain range.The presence of B. multilineatus in the Pyrenees raises the possibility of its occurring more widely in northern Spain.In view of this I checked the single male B. ibericus from the Puerto de los Señales in the Cantabrian Mountains (Dutton, Angus, 2007), and it is true B. ibericus, homozygous for the fusion of autosome pairs 1 and 24.Thus the Pyrenean population is best regarded as a glacial relict, comparable with the Alpine populations of B. griseostriatus.The presence of B. multilineatus in the Pyrenees and B. griseostriatus in the Alps is in accord with the more western distribution of B. multilineatus in Scandinavia and the northern British Isles.
The new records for B. griseostriatus and B. alpestris give westward extensions of their known ranges, this being particularly striking in the case of B. alpestris.The distributions of their known populations (Fig. 1) show both species widely distributed from east to west, with B. alpestris consistently to the south of B. griseostriatus.The main unanswered question, at least as far as distributions in the western Alps are concerned, is the distribution of B. inexpectatus (Dutton et Angus, 2007).At present this species is known only from the Lac de Lauzet Inférieur, the smaller and higher of the two Lauzet lakes.

Fig. 2 .
Fig. 2. Map of the Western Mediterranean area showing the known distributions of B. ibericus (b) and the Pyrenean locality for B. multilineatus (e).For explanation of symbols see caption to Fig. 1.
in the S. griseostriatus group by Nilsson and Angus, was shown to come out basal to the Stictotarsus (duodecimpustulatusgroup) -Nebrioporus -Scarodytes clade, thus taking it right out of the S. griseostriatus group and rendering the generic name Trichonectes Guignot, 1941 unavailable for the S. griseostriatus group.

Table 1 .
Material used for chromosomal analysis.
The larger lake has B. ibericus.It seems possible that B. inexpectatus may occur in the Massif du Pelvoux, west of Briançon.The Moroccan record of S. ibericus shows this species to have a wide and consistent distribution in SW Europe and the adjacent part of North Africa.It is unfortunate that to date no material has been available from the Italian Apennines.The species here could be B. ibericus (derived from the Alpes Maritimes), B. alpestris (from the Alps further east), B. macedonicus Georgiev, 1959 (fl own across the Adriatic) -or something completely different.