302 THE CANADIAN ENTOMOLOGIST. 



winter in this manner the parasite occupies that portion of the year, during 



which the cHmatic conditions would prohibit its activities in the way of 



attacking fresh cocoons, were it able 

 to produce more broods. 



The Larva. 



The full-grown larva (fig. 22), 



measures 2.35 mm. in length. It is 



FiG. 22.— Fuii-o:rown larva of c. white, and the body, which con- 



nemaiicida, x 12. 



sists of thirteen segments, has the 

 dorsal side more strongly convex The larva feeds externally, usually with 

 its head buried in the side of the host. 



Nu77iber of Broods. 

 As it is not found possible in these experiments to follow the develop- 

 ment of a single line of the parasites through the year, exact statements as 

 to the number of broods during the season cannot be made at present. I 

 believe, however, that it is safe to make inferences from the observations 

 which were made on material collected in Massachusetts and in Canada. 

 In one batch of cocoons, adults emerged on October 9th from cocoons in 

 which eggs had been deposited on September 13th to i6th, from which it 

 would appear that the time of development of a summer brood was about 23 

 days. It has been found that the females oviposit shortly after emergence, 

 so that no lengthy period necessarily intervenes between the development 

 of the broods if the parasite can find healthy cocoons of the sawfly. 



The prevalence of healthy cocoons of the host would determine the 

 efficiency of the parasite. It has been found in studying the life-history 

 of the host L. erichsoiiii that the sawfly larvte may become full grown 

 and form the cocoons as early as June 12th to June 17th, which would 

 mean that in any year cocoons of that year's sawfly larvae could be found 

 from the middle of June. Further, it has also been found that sawflies 

 will continue to emerge from the cocoons of the larvse of the previous 

 year until the end of June, which indicates that there is a supply of the 

 previous year's larvae in their cocoons until the first or second week in 

 June. In short, it has been found that cocoons containing larviie of L. 

 ericlisonii may be found throughout the whole year, the time of least 

 abundance being in June. It may be assumed, therefore, that if the 

 Chalcid can find heaUhy cocoons, and this has been shown to be possible 

 during the whole of the year, the production of broods may continue 

 throughout that portion of the year during which the climatic conditions 



