202 THE entomologist's EECORb. 



all the joints are functionally perfect. We may note that there are 

 only four tarsal joints, the last two being apparently fused together 

 in O. Corsica, whilst in (K tnt/ntepltras, though all five joints are 

 present, there is considerable variation as to whether anchylosis shall 

 involve the last two joints, or include one or two more, or even 

 apparently threaten all. (>. intermedia agrees apparently with 

 (K Corsica in structure, the tarsi having four joints, without any 

 definite indication of whether one is atrophied out of existence or has 

 united with another. My material, therefore, suggests three divisions 

 of the (K ericae group : — 



1. 0. ericae, five-jointed tarsi, antennae l"6mm. 



2. 0. intermedia, four-jointed tarsi, antennae l'65mm.; 0. comt-a, four-jointed 

 tarsi, antennae l-'Jmm. 



3. O. fripotepliras, five-jointed tarsi (anchylosing variously), antennse l-lmm. 

 The male O. anceps I have is possibly correctly named, the female 



hardly can be. It is very close indeed to i>. sjilenclida, but differs by 

 having the legs distinctly smaller, the claws rather weaker and the egg 

 also smaller, l-3mm. wide instead of l-7mm. This clearly belongs to 

 a race of (>. dttbia, distinguishable from (>. splendida, but can hardly 

 have a male like the (>. anceps I have. The habit of retaining the 

 pupal head-covers varies a good deal, I am not aware that o. f/onos- 

 tiipiia ever retains them. O. ericae is said to do so. My specimens of 

 the group show that they are frequently retained, but they are com- 

 paratively loosely attached and easily come off, so that those without 

 them, probably retained them for a time. The antennae are so well 

 developed, that one is rather surprised at this habit, the pupal antennae 

 being very well developed, so that they form projections, by which the 

 whole covering would be easily torn off, such violence is, however, 

 probably never applied within the cocoon. (>. aiirolitnbata, with much 

 more degenerate antenna', does not retain the head-covers, whilst^'. 

 ■ splendida retains them very firmly, so that they are almost a definite 

 portion of the imago, and it requires some care and trouble to remove 

 them. 



Exi'LANATIOX OF PlATIC VI. 



All the figures are enlarged 14 diameters, except fig. 8. 

 1, 2, 3. First, second and third legs of O. antiqua ? . 

 4 and 5. Fore and liindwings of <>. antiqua ? . 

 (5. Palpi and maxilla of (>. antiqua ? . 



7. Antenna of O. antiqua ? . 



8. Antenna of O. antiqua. Two segments, more enlarged. 



0. Ovipositor of O. antiqua. Segments cut through on left side and flattened 

 out. 



10. Pupal head of (). i^plendida, as removed from ? imago. 



11. Head of O. xjAendida ? showing (?) evanescent antennae. 



12. Ovipositor of <>. splendida cut through ventrally and flattened out. 



Explanation of Plate VII. 



1. First leg of 0. gorw^tifitna, ? x 14. 



2. Second leg of O. (jonoxtigvia, ? x 14. 



3. Third leg of 0. ijonustuima, ? x 14. 



4. Forewing of O. ;jono>^ti<ima, ? x 14. 



5. Hindwing of O. nonostifinia, ? x 14. 



6. Antenna of 0. [loiio.^iinnia, ? x 14. 



7. Tip of antenna of (). iioi/o-ttigma, ? x 40. 



8. Ovipositor of U. gono.^tigma, ? x 14 (divided dorsally, of other species 

 ventrally). 



9. First leg of O. corsiea, ? x 14. 

 10. Second leg of 0. Corsica, ? x 14. 



