HOST RELATIONS OF COMPSILURA CONCTNNATA 29 



two things has happened ; either mella has found a host more favor- 

 able than the gipsy moth or it has suffered a marked decrease in 

 numbers. The latter view is the more logical. Compsilura is not 

 entirely responsible for this condition, however, for there must be 

 considered the prodigious waste of eggs by mella upon the gipsy 

 moth. In the studies of Forbush and Fernald (<£, p. 386) made dur- 

 ing July, 1895, it was found that, whereas about 29 per cent of the 

 gipsy-moth larvae had tachinid eggs on them, these were in nearly 

 every case moulted off before hatching. It is obvious that, if this 

 futile attack by mella continued at the same rate for several years, 

 there would necessarily be a great scarcity of the species. 



There are other instances besides that of Tachina mella in which 

 the native tachinids meet with great competition, and a certain 

 amount of duplicate parasitism results, in which the native species 

 appear to be at a disadvantage. 



The tachinids Phorocera claripennis, Frontina frenchii, and Win- 

 themia quadripustulata have a great variety of hosts, nearly all of 

 which are also acceptable to Compsilura. Like Tachina mella they 

 deposit eggs on the host larva and so are handicapped by the subcu- 

 taneous larviposition of Compsilura. Probably P. claripennis fares 

 the worst of these species, for with a single exception — Lophyrus 

 lecontei Fitch — the host species are mutual. There is evidence to 

 indicate that not only does this species overwinter as a larva within 

 its host pupa, but that it often successfully hibernates in the pu- 

 parium. This habit is of the greatest importance, for it should be 

 borne in mind that Compsilura depends upon certain lepidopterous 

 pupae for hibernation and that the number of overwintering indi- 

 viduals is to that extent limited. The hibernating quarters of 

 claripennis, on the other hand, are unlimited, and it can depend upon 

 its abundance in the spring for successful competition. 



Frontina frenchii and Winthemia quadripustulata seem to be the 

 least affected hj the presence of Compsilura. As has been previously 

 mentioned, frenchii finds an excellent overwintering host in the 

 saturniids and meets there with little competition from Compsilura. 

 There are many examples of the value of Samia cecropia as a winter 

 host for frenchii. Fiske and Thompson (3) reared it in large num- 

 bers and mention an instance in which 90 adults were reared from a 

 single cocoon by C. H. T. Townsend. No other species met in the 

 writer's studies proves itself so generous a host. 



Of these native tachinids which possess the same general hosts as 

 Compsilura, Winthemia quadripustulata seems best fitted for the 

 contest. Not only does it more nearly approach Compsilura in the 

 number of its hosts, but it also has a variety of others from which 

 Compsilura has never been reared. It has, too, a decided advantage 

 over Compsilura, inasmuch as it, like Frontina frenchii, has several 

 hosts which are capable of supporting a large number of individuals. 

 Finally, it also has the advantage of hibernation as a full-grown 

 larva or in the pupal state. 



Taken as a whole, the native larval collections reveal but few cases 

 of tachinids peculiar to a single host. Among the species attacked by 

 Compsilura, there were found but 10 where there is a single compet- 

 ing tachinid with apparently no other host. Of these, the majority 

 are of no particular host value to Compsilura. In one case only, that 

 of Euvanessa antiopa L., is there any serious competition, and here 



