lepidopterous larvce and pupce. 167 



four more became free, and the two last were also 

 detached upon the 23rd, upon which date the larva was 

 dead and collapsed, having thus lasted through twelve 

 days of attack from the parasites. Newport describes 

 the change in appearance and detachment of the larvae 

 of I\ ti'staceus, but states that they then cease to feed. 

 This was far from being the case with P. cephalotes, for 

 the seven larvae vigorously attacked the dead body, 

 forcing their heads into it to such an extent as to cause 

 pits in the integument. In this way they increased 

 largely in size, and completely finished the larva, except 

 for the dried and empty integument. I cannot find an 

 account of this feeding after detachment in previous 

 descriptions. It would be interesting in the future to 

 supply such larvae with another host, and thus to see if 

 their size is modifiable to any great extent by renewing 

 the supply of food, as one might expect from Newport's 

 observation of their small size when too many developed. 

 The parasites grew very rapidly between the time when 

 they became free and August 2Sth, when they began to 

 spin. Thus the larvae fed during a period of from four- 

 teen to seventeen days. In the natural condition within 

 the cocoon of D. vinula the parasites spin oval black 

 cocoons so closely pressed together as to render the sides 

 polyhedral. My larvas were in a large chip-box, and it 

 seems likely that all their silk was exhausted in making 

 a scaffolding upon which to construct their cocoons. As 

 it was, when they became inactive and could spin no 

 more, the space was very far from being filled up, and 

 all seven larvae were left lying close together (in contact) 

 at the bottom of the concave upper surface of the silk 

 which thickly lined the lower side of the box. The top of 

 the box and the sides above the thick concave pad, were 

 only covered with a thin layer of silk (especially thin on 

 the sides). The seven larvae still remain alive, but 

 without further indications of change, crowded together 

 at the bottom of a common nest (February 20th, 1886). 

 The silk when first spun was white, but it subsequently 

 darkened, becoming finally black. DeGeer also describes 

 the scaffolding, and points out that it encloses the oval 

 cocoons in a common web. Newport says that he found 

 that his parasitic larvae had recently pupated in April, 

 while the perfect insects emerged in May, a week later 

 than the moths of D. vinula, which had been subject to 

 the same conditions of temperature. This accords with 



