— 413 — 



the possession of observational powers and tecliuical 

 acumen of a very high order, uDforlunately deals with the 

 subject from the point of view of the collector ralher than 

 of the zoologist ; and, it is exceedingly questionable 

 whether the majority of the species and varieties which are 

 enumerated therein bave any real existence in Nature. But 

 little pains bave so far been taken to study the range of varia- 

 tion, and the Columbellidae are variable in the extrême, — 

 of any of the forms during life ; and, most of the material 

 which is available consisls merely of dead, beach-rolled 

 spécimens derived from the miscellaneous refuse of a 

 shell-beach, and quite inadéquate for really scientific 

 research. Hovv^ever, even this unsatisfactory material 

 shows, when a large séries is studied, that many of the 

 so-called species can be readily linked together, and, also, 

 that there is, on the other hand, every probability that 

 the différent islands bave each their own peculiar, more 

 or less well marked, local races or varieties, — although 

 thèse varieties do not as a ruie coincide with the named 

 varieties of the collector. In short, the Pacific fauna bas 

 been treated in quite an artiticial manner, and unfortuna- 

 tely it is by no means alone in this respect. A collector 

 receives a parcel of shelis, — probably the indiscriminate 

 gathering from a shell beach, and he then sets to work 

 to bestow spécifie and varielal names upon every departure 

 from certain quite arbitrary standards which he bas set 

 up in bis own mind : not knowing, or indeed caring, 

 whether the divergences from bis imaginary standard are 

 merely terms in the range of individual variability of a 

 single species, whether they may beassociated with parti- 

 cular conditions of environment, or if they are really 

 expressive of truly genetic distinctness. Now, this is not 

 Science, for absolutely no advantage is attained by the 

 bestowal of names by which, for example, a stout indivi- 

 dual may be differentiated from one of more slender pro- 



