168 BR. J. ENT. NAT. HIST., 6: 1993 



The ovipositor differs from those of the other British Opomyza in sternite 8 being 

 truncate and slightly indented (Fig. 5d). Two other European species also have a dark 

 central stripe on the thorax: nigriventris Loew has an almost black, shiny abdomen and 

 is not known from western Europe, and decora Oldenberg, known only from Italy, 

 lacks the spots on the distal section of vein R 4+5 and cell r 4+5 is somewhat darkened. 

 There are records from Kent, Sussex, Surrey (Allen, 1965), Berkshire, Somerset 

 (Drake & Godfrey, 1989), Cambridgeshire, Suffolk, Norfolk, Salop, Cheshire, 

 S. Yorkshire, Lancashire, Cumbria, Argyll, and all Welsh counties. The majority 

 of the records come from bogs, heaths and fens. Although the fly has not been reared, 

 there is strong circumstantial evidence that the host plant is Molinia caerulea (L.) 

 Moench as this was the dominant grass at many of the localities. In Czechoslovakia, 

 Martinek (1978a) has also swept it from pure stands of Molinia on wet ground in 

 partial shade but not in dry areas away from the woodland. Despite the widespread 

 occurrence of Molinia, O. lineatopunctata can be very locally distributed within 

 apparently suitable areas. Falk (1991) lists it as nationally scarce although it probably 

 does not warrant this status. The dates of capture range from late June to November 

 with the majority in mid July to early August, though the fly was frequent in October 

 in Welsh wetlands. 



Opomyza petrei Mesnil, 1934. This species is probably overlooked because of its 

 resemblance to the ubiquitous germinationis . In the field, its brighter colour helps 

 to distinguish it from that species but some specimens are dark and have only a very 

 vaguely differentiated central abdominal stripe so the other key characters need to 

 be checked. 



Opomyza petrei is a widespread species found in most parts of Britain, including 

 Orkney, though Soos (1984) does not mention its presence here. No habitat preferences 

 are apparent in the records. The larvae have been found in tillers of Holcus lanatus 

 L., Anthoxanthum odoratum L. and Agrostis tenuis Sibth., and would seem to show 

 a markedly different host plant preference to that of O. germinationis (Mesnil, 1934, 

 Nye, 1958, 1959). 



Opomyza punctata Haliday, 1833. Several authors have commented on the 

 unsatisfactory separation of this species from florum (Collin, 1945; Hackman, 

 1958; Greve, 1981). It differs from florum in having 1-6 spots on the last section 

 of vein R 4+5 and, in my sample of eight females and six males, having significantly 

 (P<0.05) smaller mean wing and thorax lengths with almost no overlap in size 

 (Table 1). I have dissected and compared the male genitalia of both species but could 

 find no differences in any external or internal components. The internal structures 

 of lineatopunctata, germinationis and petrei are, by contrast, markedly different from 

 this pair of species. The two species are provisionally kept separate in this work. 



Opomyza punctata has been recorded from Kent (Parmenter, 1960; Allen, 1965), 

 Essex, Suffolk, Norfolk, Lincolnshire, Hereford, Leicestershire, South and North 

 Yorkshire, Durham, Cumbria, Northumberland and County Down, Ireland (type 

 specimen). The only reference to the biology of the species is given by Allen (1965) 

 who recorded it in areas dominated by Dactylis glomerata L. in his garden. It has 

 twice been caught in Arrhenatherum-dominated sward on coastal dunes. The dates 

 of capture range from July to early September so it is likely to be univoltine. It is 

 classified as nationally scarce (Falk, 1991). 



Opomyza punctella Fallen, 1820. This is a small, entirely yellow species. The extra 

 crossveins and spots separate it from other British species except lineatopunctata 



