338 LMay 



ter of the genus is not very clearly stated. Pediaspis has, like Cynips 

 Hartig, (in the restricted sense.) a pubescent thorax and seems in 

 general rather closely allied to it. 



In my former publications on Cynipidse, 1 have more than once com- 

 plained of the incompleteness of Hartig's definitions of the genera in- 

 troduced by him. The reader may judge now for himself. For my 

 own part, even with the aid of typical specimens, which I owe to the 

 kindness of l>r. Reinhard, I am still unable to recognize the genera 

 Andricus, X> urotents, Spathegaster and Triognaspis with any degree 

 of scientific accuracy. 



Without speaking of the difficulty of counting the joints of the palpi, 

 it is contrary to all analogy, that their number should be so variable 

 in closely allied genera. And that the European entomologists them- 

 selves do not value this character very highly is proved by the fact 

 that Giraud united Teras and Andricus into one genus, from want of 

 sufficient characters to distinguish them, although Teras, according to 

 Hartig. has 4-jointed, Andricus 5-jointed maxillary palpi. The palpi 

 beina left out. what is the difference between Andricus and Neurote- 

 run f Hartig's phrase : " thorax bare, usually coriaceous" for the former, 

 and -'thorax bare, usually smooth." is the only, but not a sufficient, defi- 

 nition. In the above quoted passage, Haiti" informs us that Spathe- 

 aaster has the % abdomen pedunculated ; but the same is the case with 

 Trigonaspis ! (A specimen of this genus was sent to me by Dr. Rein- 

 hard.) In what. then, does the difference between these genera con- 

 sist? As if to increase the difficulty, Mr. Hartig tells us (compare 

 above) that there is tht must remarkable resemblance between the males 

 of Andricus ami Trigonaspis, and. in the enumeration of the differences 

 which follow, he makes no mention of the pedunculated £ abdomen. 

 Have we to conclude from this statement that the % abdomen of An- 

 dricus is also sometimes pedunculated? 



It is the place here to bring to notice, that Hartig generally counted 

 one antenna! joint more than necessary, as he evidently considered as 

 the first joint, the socket of the antenna, which is often visible below the 

 first joint. Noticing the frequent disagreement between Hartig's state- 

 ments in this respect and my own observations. I always suspected that 

 such was the case, until recently my suspicions were confirmed, when 

 I happened to notice Hartig's remark about Trigonaspis (1. c. II. p. 

 195): " articulus 4 antennarum £ curvatus," which evidently refers 

 to the third joint. 



The difficulty experienced by me in arranging the North American 





