First Experiments. 5 



circumstance, added to a desire to investigate the 

 method of gall formation, that led me to undertake my 

 first direct experiments in breeding. These afforded me 

 the surprising result that from the eggs laid by Neuroterus 

 there appeared a totally different generation, one so wholly 

 unlike its parent that it had been described hitherto as of 

 another genus {Spathegaster). This fact was published 

 by me in 1877 \ and what Bassett had only thrown 

 out as a conjecture in 1873 was now proved and 

 demonstrated in one species at least. It is not then 

 correct to affirm that Walsh had previously discovered 

 this alternation of generations. Walsh had only 

 broached the theory that to one male form might 

 belong two entirely different female forms w^hich had 

 until then been described as distinct species : but it is 

 clear that the alternation of generations in the Cynipidae, 

 as discovered by me, had nothing whatever in common 

 with Walsh's supposed dimorphism. After I had once 

 discovered this remarkable alternation of generations in 

 Neuroterus it was interesting to investigate the propaga- 

 tion of the remaining genera and species. The fauna 

 of this locality (Schleswig) includes about forty species, 

 and this fairly represents the oak gall-flies of North 

 Germany ; to these therefore I steadily extended my 

 observations. But before I go into any account of 

 these species in detail, it may be well first to describe 

 shortly the method which I adopted in carrying out my 

 experiments. 



In order that the results which I obtained might 

 be unquestionable it was necessary to select a method 

 ^ Deutsche Entornolog. Zeitschr. 1877, Heft I. 



