Kinsey, Ayres: Varieties of a Gall Wasp 



143 



be done in every instance. The galls indicate closely similar 

 physiologies for all varieties, but in most cases can be dis- 

 tinguished. Where the insects show closest relations then the 

 galls are most similar! This is the case with a series com- 

 prising varieties tiibercidatrix, similis, arefacta, and colora- 

 densis ; similarly with a series leading from coloradensis to 

 tumida and tuasatchensis ; again with a series including cali- 

 fornica versicolor, and melanderi; and with the distinct and 

 compact series of rubriderma, sierranensis , and descansonis. 

 These more closely related varieties might be considered in 

 groups as subspecies if we had occasion to use the terminology. 



In such series of varieties the developments of characters 

 appear to proceed in a continuous, geographic direction, as is 

 discussed under sierranensis. Orthogenesis and other fright- 

 ful words loom large, but need further investigation. 



Two of these series of species present the curious situation 

 of ranges which cross, as we explain further under 

 sierra7iensis . 



This species alone will serve to decide the validity of 

 Kieffer's term, Lytorhodites Kieffer (1902, Bull. Soc. Metz, 

 (2), X, p. 96). This genus was established to include species 

 of Dipolepis, "Rhodites'\ which show a scutellum without 

 fovese, the abdomen microscopically reticulate, and the radial 

 cell more or less open. There is certainly no correlation be- 

 tween the occurrence of these characters among species of 

 rose gall makers, and the nature of the radial cell has been the 

 single character used to delimit the genus. Arefacta, one of 

 the varieties of the present species, is type of the genus (des- 

 ignated by Rohwer and Fagan, 1918, Proc. U.S. Nat. Mus., 

 LIII, p. 370). In this species, tubercidatrix, occur varieties 

 with the radial cell of the female regularly open ; in several 

 other varieties, such as calif ornica, individuals have the cell 

 either open or closed, usually closed. Sometimes a single indi- 

 vidual will present different conditions on the two wings, that 

 is, belonging to one genus on one wing, to another on the other 

 wing! The males of several of the varieties have the radial 

 cell always closed. This, in connection with the entire lack 

 of other correlated characters for the genus, confirms our 

 previous surmise (Kinsey, 1920, Bull. Amer. Mus. Nat. Hist., 

 XLII, p. 392) that Lytorhodites cannot be maintained as a 

 genus distinct from Diplolepis. 



Of 4,060 individuals bred, 1,858, or about 46 per cent are 



