GALLS CAUSED BY GALL-WASPS 31 



the genus Neuroterus, arriving at the surprising result that 

 from the eggs laid by these, wasps were produced which 

 were so unlike their parents that they had been placed in a 

 separate genus, Spatkegaster. He published this fact in 

 1877, and afterwards extended his observations to the 

 majority of the Oak gall-flies of North Germany, showing 

 that the phenomenon is observable in many species. 



Particulars concerning the methods of investigation 

 employed by Dr. Adler in the course of his researches may 

 be consulted in the first chapter of " Alternating Genera- 

 tions." Oak saplings were used, either grown in pots 

 or obtained from nurserymen ; those from four to six years 

 old proved to be the most convenient size. It was found 

 that gauze covers with a glass top were better than ordinary 

 bell-glass protectors, the absence of free ventilation causing 

 the latter quickly to become dimmed with moisture. It 

 was easy to make experiments with species which prick 

 the leaves or bark, but some difficulty was experienced 

 with those which only prick flower-buds, as the majority of 

 four to six year saplings do not produce catkins. Conse- 

 quently the experiments were made in the open air on full- 

 grown trees, using cubes of wire covered with muslin, and 

 tied round the branch. " It is an essential that a sapling 

 about to be used in an experiment should have its buds 

 well developed, as these are always preferred by the flies." 



The following tables concern the known British species 

 of gall-causing Cynipidae. If the second is examined in 

 connexion with the footnote, it will be observed (omitting 

 Cynips KoUari and C. calicis for obvious reasons) that there 

 are six species in which the agamous generation alone is 

 known, and six in which it is unknown. It is probable 

 that future research will show there are but six species, each 

 with an alternating generation. 



