304 PROCEEDINGS OF THE ACADEMY OF [May, 



enings between the posterior spinnerets. The only difference in 

 Loxosceles is that the colulus has an unpaired origin; but this differ- 

 ence does not hold in all cases, for in Trochosa ( Jaworowski) the colulus 

 has a paired origin, and we have seen that in Agelena the colulus is a 

 paired structure in the adult. Evidently colulus and median spinnerets 

 are homodynamous. 



I have examined the cribellum in Filistata, Hyptiotes, Uloborus and 

 Dictyna. In the adult female of Filistata there is found a transversely 

 oval plate between the anterior spinnerets and somewhat anterior to 

 them. This plate (fig. 34, Plate XIV) shows a pair of spinning plates 

 (Cr. PL), provided with numerous fine pores, and these spinning plates 

 are set in the much larger chitinous frame ; on the anterior portion of 

 this frame are inserted hairs (their insertion points indicated by the 

 small circles), which hairs are plumose like these of the venter and 

 differ from them only in their much smaller size. These relations are 

 essentially such as Bertkau and others have described. But an im- 

 portant relation seems to have been heretofore entirely overlooked. 

 The paired spinning plates, the cribellum proper, are only the un- 

 covered free surfaces of a pair of appendages, the remaining portion 

 of which is sunk below the chitinous frame. Thus when one examines 

 on a cleared balsam mount the cribellum from the ventral surface, 

 one sees at a higher focus of the microscope only the frame and the 

 spinning plates ; but on deeper focus one finds a column of delicate 

 parallel spinning tubules, too fine to be represented on the scale of 

 our drawing, extending from each spinning plate to the anterior edge 

 of the frame. The lateral boundaries of these columns or append- 

 ages are shown in fig. 34 by the lines extending forward (upward) from 

 the spinning plates (Cr. PL) to the upper border of the frame. 



Further, each spinning plate shows a slight line of division at its 

 middle, from which point of indentation a line extends some distance 

 forward; this subdivision is probably due to the presence of a muscle 

 (as it is certainly in some other species), though I could not clearly 

 recognize a muscle. In other words, the spinning plates are merely 

 the free, uncovered apices of a pair of spinnerets, the greater region of 

 which lie covered by the cuticular frame. Or the relation might be 

 represented in another way: the cribellum of Filistata is a pair of 

 spinnerets mesially approximated, their glands opening distally (on 

 the spinning plates), and across and around these spinnerets has 

 developed a cuticular frame. 



This interesting relation in Filistata is further explained by a study 

 of the embryo shown in fig. 33, PI. XIV; no one has previously de- 



