188 PROCEEDINGS OF T1IE NATIONAL MUSEUM. vol.59. 



microscope and certain ratios determined, using the width of the head 

 as a convenient firmly chitinized base. The length of the mesonotum 

 is the distance from the front margin of the mesoscutum to the hind 

 end of the scuteilum, these two remaining connected in a successful 

 dissection. The mesonotum ratio is obtained by dividing this dis- 

 tance by the width of the head. The lengths of antenna and 

 ovipositor were found by stepping dividers set for convenience at 

 5 mm. along the curves of the drawing and the ratio found by 

 dividing by the width of the head. Some preliminary study indi- 

 cates that these ratios are fairly constant regardless of the size of the 

 individual in the species. The interocular area ratio can be deter- 

 mined from balsam slide or from pinned specimen by measuring the 

 distance between the compound eyes at level of the antennae and 

 dividing by length of eye as measured by a filar micrometer with a 

 2-inch or with a two-thirds inch objective. 



In this paper galls are not regarded as a part of a species any more 

 than is the work of a leaf-mining lepidopteron or the galleries of an 

 engraver beetle, characteristic as such work may be of the species in 

 question. A gall is a part of a plant and most cecidologists now hesi- 

 tate to attach to such abnormalities alone binomial Latin names. 

 One can not predict with any certainty what genus is responsible for a 

 gall. Thus when it seems desirable to mention new unreared galls 

 they will be referred to simply by number to avoid adding useless 

 names to the bibliography. The classification of the Cynipidae will 

 progress only by a study of the adults. The more biology that can 

 be correlated the better, and the work of a species may often be the 

 means of its quickest recognition, but the taxonomy of the group 

 must rest on structural characters in the insects themselves. This 

 policy would logically lead to the exclusion of the galls from the type 

 material of a species, and yet it seems desirable that those examples 

 of the work which the author associates with a certain species should 

 be preserved and kept separate for future reference. There is no 

 way to designate them at present except to call them types; and with 

 this meaning only in mind, type labels have been attached to the galls 

 which the author associates with the species. This association is 

 absolutely certain in those rare cases where the identical gall from 

 which the type fly came has been preserved, and this may be properly 

 designated as a type gall. In most cases the association is a matter 

 of judgment. Usually a lot of galls are put in a breeding cage to- 

 gether, and it is impossible to select the one from which a given adult 

 emerged, and in some cases the type flies are cut from galls and the 

 fragments are worthless for purposes of identification and others like 

 them from the same or even different locality are the only ones at 

 hand. They are at best only illustrations of what the author con- 

 siders to be the work of the species; and with this meaning only 



