AQUATIC INSECTS IN NEW YORK STATE 299 



of three segments, an eighth, ninth, and a minute tenth. When 

 the dorsal aspect of the segments in front of this region is 

 studied, they are found to be divided transversely into two sub- 

 segments, each subsegment bearing a transverse band of setae. 

 If we compare this with the region behind the seventh segment, 

 we find that there are four indistinctly marked transverse divi- 

 sions, each bearing a band of setae, which would seem to indi- 

 cate that we have to do with two segments. But, when the 

 ventral aspect is studied, we find that the transverse divisions 

 separating the segments into subsegments are wanting, and 

 that each transverse line marks the suture between two seg- 

 ments, and, when the portion beyond the seventh segment is 

 examined, there are found two transverse divisions [pl.27, 

 figs. 17 and 19], which proves conclusively that this portion is 

 made up of three segments, making 10 abdominal segments in 

 all. The first transverse division on the dorsal aspect caudad 

 of the seventh segment is the suture dividing the eighth seg- 

 ment into subsegments; the ninth and 10th segments are not 

 divided into subsegments. These conclusions are apparently 

 refuted by plate 28, figure 17, where only a single ventral divi- 

 sion is shown beyond the suture of the seventh segment, but 

 a few sections on either side of the one from which this figure 

 was made show two well marked ventral divisions.^ 



Each segment bears a number of minute hair-like setae. The 

 arrangement of these setae into definite areas and the interre- 

 lation of the areas seem to furnish the best characters for sep- 

 arating the larvae of the various species. It has been found 

 necessary for the sake of exactness and brevity to apply names 

 to these areas [pl.30, fig.l] which are as follows: 



The tergal setae are the double row of transverse setae on 

 the dorsum of each segment, the anterior row being known as 

 the anterior tergal setae [Hg.lat] and the posterior row as the 

 posterior tergal setae [fig.lp^]. 



The supraspiracular setae are the patches of setae between 

 the spiracles and the tergal setae [fig.lss]. 



The infraspiracular setae are the small patches of setae imme- 

 diately beneath the spiracles [fig.lis]. 



The pedal setae are the patches of setae surrounding the 

 thoracic legs and on the protuberances where the prolegs would 

 be situated if they were present [fig.lps]. 



i"The long filiform lateral appendages" referred to by Mr Sanderson 

 as figured by Kolliker, A. Observationes de prima iusectorum genesi 

 adjecta articulatorum evolutionis cum vertebratorum eoniparatii>ue. 

 Observationes de prima insectorum genesi. Ann. des Sci. Nat. (2) 1S43. 

 20:267-99, pi. 12, fig.6, are the deciduous setae represented on pl.27, fig.13; 

 and I can not see how they can have any bearing on the question of the 

 homology of the segments. 



