Alternating Generations in Oak Gall- Flies. 151 



the well known bean-shaped galls of this species, which 

 are often found very plentifully on Salix amygdalina. 

 In May, 1877, I reared the flies, and was able to convince 

 myself that they were all of the female sex. I continued 

 the experiment by putting them on little willow saplings 

 set in pots, and soon observed the flies begin to saw the 

 tender leaves at the end of the shoots, and deposit their 

 eggs in them. At the beginning of July the full-grown 

 larvae from the galls buried themselves in the earth, to 

 assume the pupa state. They remained in this condi- 

 tion for a very short time, and the first flies appeared on 

 July 27. Again they were all females, and began at once 

 to deposit their eggs. By the end of August the galls 

 were fully developed on the leaves which had been 

 pricked, and the larvae betook themselves to the ground 

 in October, to assume the pupa state. In this particular 

 instance, two generations appear yearly, by strictly par" 

 thenogenetic propagation ; so that while Nematus ventri- 

 cosus only occasionally reproduces itself by partheno- 

 genesis, Nematus Vallisiiierii habitually does so. At the 

 same time, the occurrence of parthenogenesis in Nematus 

 ventricosus shows that it may proceed directly from 

 a sexual generation. Probably this occurs more readily 

 amongst Hymenoptera than amongst any other class 

 of insects, and therefore I should like to add the 

 following observations on Pteromalus puparum. 



This little parasite lays its eggs in the pupae of 

 different lepidoptera, such as Vanessa lo, V. poly- 

 cJiloros, V. urticae, Pieris rapae, &c. A single pupa 

 often supplies more than a hundred of these little flies, 

 so that it is not difficult to collect a sufficiency of them. 



