BiorJiiza Group. 137 



One great difficulty in carrying this out is the fact that 

 neither the alternate generations nor the gall formations 

 of all our C3mipidae have yet been discovered. This is 

 especially the case with those species living on Quercus 

 cerris. Here therefore there is still open a wide but 

 fruitful field of observation. 



The food and manner of living among the Cynipidae 

 present the greatest uniformity, for in all that belongs 

 to the organs of nutrition, only those modifications 

 occur which the actual circumstances of life demand, 

 and hence these organs undergo little variation. 



This is especially true of the digestive tract : the 

 parts composing the mouth are all conformable ; all the 

 Cynipidae are provided with solid mandibles, in order 

 to cut through the envelope of the gall, which is often 

 very hard. The maxillary and labial palpi alone 

 offer any differences. At first Hartig assigned great 

 importance to the number of joints in the palpi, and 

 employed the variation in the number of these joints 

 as characters. Those given by Hartig have been 

 adopted in the descriptions of later writers, but there is 

 great need for their careful revision. Microscopic pre- 

 parations of these minute organs are rather difficult to 

 make and have been much neglected. As a rule the 

 two alternate generations have the same number of 

 joints in the palpi. Neuroterus and the corresponding 

 Spathegaster have each four joints to the maxillary palpi, 

 and two joints to the labial palpi ; but the Spathegaster 

 corresponding to a Dryophanta has, like it, five joints to 

 the maxillary, and three to the labial palpi. There is 

 one exception to this rule, in the generation-cycle formed 



