254 SPOLIA ZEYLANICA. 



The ventral setae of segments II.-V., differ somewhat from the 

 rest. The shaft is only very sUghtly curved in the usual S-shape ; 

 the distal prong of the forked end is nearly twice as long as the 

 proximal ; the prongs are about equal in thickness at their base, or 

 the proximal is perhaps a little thinner ; the angle between the 

 prongs is narrower than in the setse of the posterior bundles. The 

 nodulus is either exactly at or shghtly proximal to the middle of the 

 shaft. In length they are from 123 to 128 pi. The number per 

 bundle is four or five. 



In the segments from the sixth onwards, the distal prong is 

 slightly longer than the proximal, and only one-half or two-thirds as 

 thick at its base ; the nodulus is either shghtly or very markedly 

 distal to the middle of the shaft (distal : proximal : : 41 : 46 or 41 : 57). 

 In length they are from 87 to 98 [x. Towards the anterior end of the 

 animal there are four, or occasionally five, setae per bundle ; further 

 back the number sinks to three, and then to two. 



The alimentary canal shows but little differentiation throughout 

 its extent. The pharynx is cihated ; its floor is Uned by long 

 columnar cells, while the cells which form its roof are shorter ; 

 sections show a pair of small recesses dorsolaterally in this region. 

 There are no definite septal glands ; but a few large deeply staining 

 gland cells occur in connection with the alimentary tract in segments 

 III., IV., and V. Chloragogen cells begin in segment VI. There is 

 no stomach. 



The dorsal vessel, covered with chloragogen cells, courses along the 



ventral side of the intestine, a Httle to the right of the middle line. 



The ventral vessel is situated in a corresponding position on the left. 



The cerebral ganglion is widely indented in front, but is not 



indented behind. 



Sexual organs were not present, nor was asexual reproduction in 

 progress in any of the specimens. 



The present form would seem to be one of the best marked species 

 of the genus, and the characters of the posterior end appear to be 

 quite distinctive. It is of course true, as has been pointed out by 

 Michaelsen (6), that the gill-bearing hinder end of species of Dero 

 (including Aulophorus) is liable to varj^ considerably according to 

 whether it has or has not been comj)letely developed after the 

 process of fission ; and Bousfield (4) had previously draAvn attention 

 to the great differences which exist between the conditions of 

 contraction and full expansion. 



In the present case however the distinctive features of the hinder 

 end are due to a characteristic difference of type ; and it is impos- 

 sible to refer the peculiarities of the species to difference in degree 

 of development or to differences of contraction or expansion. The 

 chief j)eculiarities are, that for the greater part of their extent the 

 gills are long ridges onh^ ; and that Avhile in the case of the posterior 

 gills the ridges dissociate themselves from the wall of the fossa so as 



