8B THE entomologist's record. 



near to 0. autumnata as to 0. dilutata. Mr. Prout pronounced all 

 these four specimens to be 0. chrutyi. Fortunately, I remembered 

 the exact spot where I had captured the three specimens— a row of 

 beeches near a lake in a private demesne near Enniskillen. The bred 

 specimen came out among larva? beaten from alders on the margin of 

 the same lake. The following autumn I tried the beeches, disturbing 

 the moths by day from the lower branches. I soon found I was in the 

 right place, for the great majority were 0. ehristyi, with only a sprink- 

 Img of 0. dilutata. Further experience extended the locality some- 

 what, but also showed its limits. Some solitary beeches on rising 

 ground adjacent to the original locality produced (J. christyi. A 

 continuation of the row of beeches produced a few 0. christyi late in 

 the season of 1904, though previously the specimens taken in this 

 place had all been 0. dilutata (I think these 0. christi/i must have been 

 stragglers, which, late in the season, had wandered away from their 

 headquarters). I have bred a few 0. christyi in different seasons from 

 supposed U. autumnata larvfe, taken near the lake above mentioned • in 

 some cases these have been detected in the larval state by their smoky 

 black markings. LarviB beaten from the same beeches have produced 

 more O. christyi than (J. dilutata: and a few larva? from sallows near 

 the lake have produced all three species, O. christyi, O. dilutata and 

 O. autumnata. A few elms growing in the same neighbourhood have 

 produced chiefly O. christyi, in both larval and imaginal states. Haw- 

 thorn and oak m the immediate vicinity of the 0. christyi ground have 

 never produced anything but O. dilutata. For several seasons, the 

 locality described above was the only one in which I could find 

 christyi with any certainty. In October, 1905, however, I tried another 

 locality— a similar row of beeches in the same demesne. I soon found 

 that it was common there, and, in the course of three or four visits I 

 beat out from these beeches about 40 specimens, of which only one was 

 (). dilutata, and all the rest O. christyi. The above facts are sufficient 

 to show the extreme local isolation of O. christyi. To summarise the 

 argument, O. christyi occurs constantly in two areas, the first not more 

 than a quarter of a mile in length, the second less. A few specimens taken 

 at a distance from these areas indicate a probable third locality Except 

 m these three places, I have never found a single O. christyi thou<^h I 

 have been working for 0. dilutata and O. autumnata all over the distnct 

 (It may here be remarked that whereas 0. christyi and O. autumnata 

 confine themselves very strictly to their special areas, O. dilutata is 

 almost ubiquitous, but seems to be partially ousted by the other species 

 from their special areas. There is always a chance of its turning up 

 among n. christyi and U. autumnata, but it is not common ) (3) 

 Fidelity to type.— This argument is so conclusive that a very brief 

 statement of it will suffice, U. christyi breeds absolutely true I have 

 reared some eight or ten broods ex ovo, both from captured females 

 and from bred females paired in captivity, and from these I have not 

 bred a single specimen of O. dilutata. I have also reared some broods 

 of O. dilutata and have never found 0. christyi among them. 



Hybridism.— 0. christyi pairs readily with 0. dilutata in confine- 

 ment, and the progeny are fertile. I have obtained both pairino-g ^ 

 christyi X ? dilutata and ^ dilutata x ? christyi. Some of 'the 

 offspring might pass for (K christyi, and some for (K dilutata, but most 

 of them partake about equally of the external characters of both 



