MAIIIXE FAUNA OF ZAXZIBAR. 



295 



1904.] 



am assured by Pi-of. Mcintosh that the case of E. norvegica 

 adduced as a parallel is quite unproved. The collection of nine 

 specimens together, all of which have large gills, makes the theory 

 so unlikely that it has not been thought worth while to resort to 

 microscopic examination to determine that all are of the female 

 sex. The correlation of gill-complexity with body size noted above, 

 mentioned by Gravier for his U. perrieri and common to most 

 species of the genus, affords a sufficient explanation without the 

 need of any special theory. 



The compound setae vary in the shape of the articulated pieces, 

 which may be almost that of an equilateral tiiangle or, as is more 

 usual, somewhat elongated -^ - 

 this vtvriation. 



Text-fio-. 45 gives the extremes of 



Text-fi2-. 45. 



%350 



The extremes of forms of the compound sette of E. afm, from speciineiis from 

 Funafuti and from Hulule, Male Atoll. X 350. 



It is well known that differences in the shape are usual between 

 the setae of anterior and of posterior feet, but such extremes as are 

 here figured are never met with on the same worm. Of E. afra 

 nine specimens were specially examined in this respect. Of these 

 two showed a distinct difference in the breadth of the end piece 

 of the compound setse of anterior and posterioi- feet, in the other 

 seven no variation in those of the same individual was noticeable. 

 We have to deal with a second variation of a distinct kind, between 

 individuals rather than between the metameric parts of the same 

 worm. Of fifteen specimens examined in detail, eight have setae 

 which approximate to the elongated type, four those which ax-o 



