of the Genera of the Arancoc. 53 



all, then the disjjutation is at least confined to the inter- 

 pretation of the strict letter of the law of priority. 



1'he original spelling, therefore, of each name is given in 

 every case in this revision ; for although it is true that 

 Micromata may offend the classical eye, just as the asso- 

 ciations also gathered round ct-rtain names are swept away 

 and feelings wounded by any alteration in the nomenclature, 

 still sonietliing must be sacrificed for the sake of uniformity, 

 and it is better to sacrifice feelings, which are transient, than 

 to tamper with printed facts, which will, at any rate, outlive 

 authors, sentiments, and associalions. 



The object held in view, then, is to ascertain what is the type 

 species of every group which has ever received a name, and 

 briefly to give the reasons why such a species must be 

 regarded as the type. 



No attemjjt is here made to determine whether this or that 

 generic group ought to be maintained or not, but simply to 

 settle wliat, if a genus is maintained, must be the type 

 species representing that genus, exclusive of any other 

 species. 



This attitude naturally involves the following of some 

 definite system, which shall be consistently applied throughout 

 and no deviation from it admitted on any consideration 

 whatever. 



The Principles of Elimination. 



The system followed in the determination of types where 

 no type has been definitely selected is known as that of 

 " Eliminatiot),^^ by which the last species left in, of those 

 originally included in the genus when first published, becomes 

 the type, supposing the group to be broken up into other 

 genera by the author himself or by subsequent authors. 



If, however, the author himself or another author has 

 definitely selected a ti/pe for the genus, either from all those 

 originally included or from the two or more species left in, 

 the species thus selected is regarded as the type, whether it 

 be the eldest species or not. On no account can a species 

 not origiiiall}- included in the group become the type of the 

 genus, even though added subsequently by the author him- 

 self or definitely selected by that author as the type. 



Species are often eliminated by " implication " in other 

 genera. For instance, supposing three species were originally 

 referred to a certain generic name and an author subsequently 

 founds a genus upon another species not originally include<l 

 but afterwards found to be congeneric with one of the ori- 

 ginal : this original species is then regarded as removed from 



