A. D. Imms 217 



fir trees. He suggests the gall-forming Trypetid Urophora solstitialis as 

 its host, "unless of course its presence there was purely accidental and 

 its true association were with the overhanging conifers, in which case it 

 would surely have shown itself in the jar in the course of the preceding 

 month." In my opinion it is more probable that one or other of the 

 several species of Lepidopterous larvae which inhabit the flower heads 

 of Centaurea is the host in question. Pimpla examinator Fab. has been 

 bred from Retinia buoliana, Liparis chrysorrhoea, and many other Lepi- 

 doptera. Pimpla pomorum is exceptional in that it has only been bred 

 with certainty, from Anthonomus pomorum. From analogy with other 

 species of its genus it is probable that it utilises Lepidopterous hosts also, 

 although it is remarkable that it has not so far been reared therefrom. 

 Morley states that he has invariably beaten it from Pinus sylvestris and 

 Picea excelsa in fir woods about Bentley in Suffolk. This suggests that 

 Retinia buoliana or other pine infesting Tortricid may prove to be one of 

 its hosts. In the absence of pine trees (as in the Chatteris locality, Cam- 

 bridgeshire) certain of the commoner Lepidopterous larvae feeding on 

 apple, hawthorn and other plants, may not unlikely prove to be para- 

 sitised by this species. 



As a general rule the parasites of exposed hosts — in other words host 

 insects not sheltered by a cocoon, or in a bud, leaf or stem — live in- 

 ternally. Howard has remarked (1913, p. 161) that it is very rare to find 

 an external Hymenopterous parasite on an unprotected host insect, and 

 it occasionally happens that the same species of parasite, in its larval 

 stage, will feed exteriorly upon a protected host and interiorly upon an 

 unprotected host. So little is known concerning the biology of the genus 

 Pimpla that it is impossible to assert how far the ecto- and endoparasitic 

 habits prevail, as both protected and exposed hosts may be utilised by 

 one and the same species. A thorough investigation of the economy of 

 members of this genus is likely to yield results of considerable biological 

 interest and importance. 



5. THE YOUNG LARVA. 



The smallest larva met with measured 1-1 mm. long (Plate XII, fig. 1) 

 and, although I have had no opportunity of examining the newly hatched 

 individual, I have little doubt that it does not differ from the latter in any 

 essential details of its morphology. It is pearly white in colour, and differs 

 from the fully grown larva in the following among other features. 

 (1) The relatively large size of the head, which is completely exserted, 

 and incapable of any withdrawal into the first trunk segment. It 



Ann. Biol, iv 15 



