124 HYMENOPTERA. 



localities are very small, about the size of P. cressoni. The 

 largest specimens are from Canada, both males and females 

 but the size does not depend especially upon the locality, for 

 there are a number of very small Canadian specimens as 

 well as large southern ones. In face marks the males vary 

 greatly ; some having the typical lateral marks remaining 

 broad to the upper edge of the antennal sockets and then 

 tapering off as in figure (98); others ending almost abruptly 

 on a level with the antennal sockets as in figures (93-95); 

 while still others have longer, less tapering marks as in 

 figures (96, 99, 100); and between all of these are all sorts 

 of intermediates, some of which are shown in the series of 

 figures (93-100). There is no uniformity whatever in these 

 variations, either according to the size of the specimen, the 

 locality in which it was taken, or any other distinguishable 

 factor. There is simply an exceedingly strong tendency to 

 vary thruout the entire species, and series from practically 

 any locality will show almost any of the modifications. In 

 thoracic markings also, there is no regularity ; often the 

 collar is entirely black ; usually the tegulae and wing bases 

 are black, and the tibiae and tarsi may have almost any 

 amount of yellow from a mere spot on each tibia to entirely 

 yellow tibiae and tarsi both. In some specimens the labrum 

 and mandibles are yellow as in P. zizice, but as a rule they 

 are dark. In puncturation the same sort of variation is 

 found, within certain limits ; the punctures of the head and 

 thorax are considerably more dense in some specimens than 

 in others ; and the basal segments of the abdomen vary from 

 absolutely smooth to quite conspicuously, altho sparsely, 

 punctured. Upon none of the above characters can distinct 

 separation be based, because they are entirely untrustworthy. 

 In this species far more than in any other is there a ten- 

 dency toward variation in the genitalia. The only place in 

 which this is exhibited clearly is in the seventh ventral plate, 

 for the eighth is so simple as to allow of almost none. In 

 the former a series of variations is found including all forms 

 intermediate between the two most extreme ones shown in 

 figures (69 and 72), the former having very broad, short. 



