PROC. EXT. SOC. WASH., VOL. 21, XO. 6, JUXE, 1919 131 



i 



by MacGillivray, 1913, would appear to need revising, and a com- 

 parison with the structures of lower insects would permit the de- 

 termining of their homologies with a fair degree of certainty. 

 For the greater part of the material upon which the present 

 study was based, and for many valuable suggestions, I am deeply 

 indebted to the kindness of Mr. S. A. Rohwer, whose generous 

 assistance has made this work possible. 



In referring to the different abdominal segments of the male, I 

 would count them in the dorsal region, beginning with the basal 

 abdominal tergum (which is usually demarked into two sym- 

 metrical halves), since the sternal region of the first abdominal 

 segment has become lost through atrophy, or through uniting 

 with the hindermost segment of the thorax. The presence of 

 the first abdominal spiracle in the basal segment of the abdomen 

 will serve to differentiate it from the thoracic region, if there is 

 any question as to its identity. For studying the union of the 

 first abdominal segment (propodeum) with the thoracic region, 

 Cephus offers an extremely interesting intermediate condition 

 between the lower and higher types of Hymenoptera; but the dis- 

 cussion of this region can be better taken up elsewhere. 



In most sawflies, the tergum or "tergite" of the eighth abdominal 

 segment (sometimes referred to as the eighth "dorsal segment") 

 is clearly evident as in Figs. 42, 46, 49, 50, 55, 56, etc., where it 

 is labeled "S 4 ." In Oryssus (Fig. 42) and many other sawflies 

 (Fig. 56) it overlaps the terga of the succeeding segments, and may 

 be referred to as the "pseudopygidium." In Tremex (Fig. 49), 

 however, the eighth tergum does not overlap the succeeding ones to 

 any great extent. The sternum or sternite of the eighth seg- 

 ment is labeled "8 s " in the above-mentioned figures. 



In some sawflies, such as Hemitaxonus, etc. (Figs. 50, 56, 57, 

 etc.), the sternum or "sternite" of the eighth segment "8 s " is 

 greatly reduced and becomes so narrow in the mesal region that 

 it is almost divided into two lateral halves. This fact, and the 

 partial overlapping of the eighth sternum by the sternum of the 

 seventh segment, caused Newell, 1918, to disregard the true eighth 

 sternum in her figures of a male sawfly (Dolerus) and to designate 

 the true ninth sternum (labeled "ha" in all figures), incorrectly, 

 as the eighth. If one examines a sawfly such as Tremex (Fig. 49), 

 however, it is very easy to identify the tergites and sternites, 

 since the eighth sternite "8 s " is large, and. the ninth tergite, or 

 pygidium, "9*-," which in Tremex and most siricids is demarked 

 into two halves by a convolution along the mid-dorsal line, is 

 not overlapped to any great extent by the eighth tergite "8 l " 



