HOLOTHTJBIOIDEA OF THE INDIAN OCEAN. 65 



H. glaberrima in which there were four stone canals, and 

 another specimen which had spicules hke H. lubrica but had 

 only one long stone canal. These results are quite contrary 

 to my observations. 



Similarly there appears to be much variation in the size of 

 the spicules of H. glaberrima a,ndH. lubrica, and also according 

 to Ludwig in regard to the shape of the deposits. In the 

 specimens examined by me the spicules do not show inter- 

 mediate stages hnking up the two species. With regard to 

 the size of the spicules my specimens of H. glaberriina show a 

 length variation from 56 |x to 84 ^, with a mean length of 

 66 ^J- The spicules of H. lubrica vary from 112 \i. to 148 ^, 

 with a mean length of 127 [J. ; that is to say, the spicules of 

 the latter species are nearly twice as long as those of the 

 former. 



In Selenka's original description (37) he gave the length of 

 the spicules of H. glaberrima as 50 ^ and of H. lubrica as 60 ^. 

 Lampert (21) gave the size of the spicules of the former species 

 as 105 ^ and of the latter as 70 \^, exactly the reverse of what 

 I find. Hence the size of the spicules in the two species varies 

 to a very large degree. 



It is seen, then, that all three points of difference between 

 the two species are dependent upon characters which have 

 been shown to be inconstant, and it is possible that Ludwig's 

 opinion that the two forms really belong to the same species 

 may be eventually borne out when a sufficiently long series of 

 specimens has been brought together for examination. In the 

 meantime I follow Lampert and Koehler & Vaney in separating 

 the two species. 



HoLOTHCTRiA ciNERASCENS (Brandt). 

 (Plate IX., fig. 10.) 



Stichopvs {Gymnochirota) cinerascens, Brandt 1835 (8). 



Holofhuria pulchella, Selenka 1867 (37) ; Semper 1868 (38) ; 

 Haacke 1860 (12) ; Ludwig 1881 (25), 1883 (27) : 

 Theel 1886 (42) ; Sluiter 1887 (39) ; BeU 1887 (6). 



K 6(5)13 



