BY H. LEIGHTON KESTEVEN. 471 



tortirostris and L. radiale in different genera. Considered as 

 varieties of one type, they may, for the convenience of the 

 monographer, be disregarded. 



'*A number of species have a world-wide distribution, which 

 is doubtless due to their free-swimming or pelagic larvse " (Tryon, 

 op. cit. p. 7). I am inclined to carry this statement further, and 

 to say that the distribution of a species is largely decided by the 

 size of its protoconch. Most of the species with many-whorled 

 protoconchs have a very wide distribution, whilst those with 

 small protoconchs are rather local. There are exceptions to this. 

 L. cornutum, Perry, with a many-whorled apex ranges over a 

 small area, whilst if Tryon is correct in regarding L. loroisii, 

 Petit, as a synonym of L. labiosum, we are presented with a species 

 having a small protoconch and a practically world-wide distribu- 

 tion. 



Conclusions o^ Part ii. 



My conclusions are that the subgeneric names heretofore used 

 under " Triton " are redundant and altogether useless. To quote 

 them conveys no more than would the generic name alone. It 

 is impossible to treat them as full genera; to do so would create 

 confusion. References to several species would have to be sought 

 under two or three genera. After eliminating L. clandestinum, 

 all the species enumerated by Tryon under Simpulum, Cymatium 

 and Guturnium form one section, which it is impossible to divide 

 into groups, if such groups are to be regarded as nomenclatural 

 entities. If from Tryon's " Triton s.st" L. suhdistortum and L. 

 parkinsonianum are withdrawn, and L. ovoideum added, we have 

 a second section ; a third, as already stated, is formed by L. 

 jyarkinsonianum and some of the Australian Tertiary species. 

 Should it be decided advisable to give the sections quotable 

 names, the above are the three which must be first recognised; 

 any further subdivision is to be deprecated. 



I am of the opinion that none of the sections are sufficiently 

 distinct to rank as subgenera, and submit the following arrange- 

 ment of the species of the genus. 



