562 TRANSACTIONS OF THE WAGNER FREE 



as their object the determination (i) of what constitutes a scientific name, and 

 (2) how to determine its priority. 



III. A name to be vaHd must be binomial (if relating to a genus and species) ; 

 must be in Latin form (at least superficially) ; must be defined or suffi- 

 ciently identified, with the object to which it is applied, for recognition in 

 the absence of specimens of that object, and must be printed and pub- 

 lished so as to be accessible to scientific men. In the case of fossils it has 

 been pretty generally conceded that a figure must be given to validate the 

 description and enable it to stand, if in conflict with a subsequent name 

 for the same object accompanied by a figure. (This latter rule is not 

 retroactive, and dates from the Bologna Geological Congress.) 



IV. Binomial nomenclature in moUusks dates from the tenth edition of Linne's 

 Systema Naturae, 1758. 



There is practically no difference of opinion about this point among 

 modern students of mollusks, and there are reasons why it is important that 

 this date should be assented to, but it would require too much space to detail 

 them here. 



The writer, in conformity with the first clause of Rule III, rejects all 

 names not given by binomial authors, or by authors in works not consistently 

 binomial (though trinomials for subspecific names are not inimical to this 

 rule). Such names do not belong to nomenclature and, if used by a binomial 

 author subsequently in a different sense, are not pre-occupied by reason of the 

 non-binomial use; or if used m the same sense, take date only from the epoch 

 of their use binomially. 



The second clause of Rule III is more important than is commonly sup- 

 posed, and has been frequently violated with apparent unconsciousness. The 

 writer holds that no vernacular names whatever should have any standing in 

 scientific nomenclature when proposed, unless (i) they have an accidentally 

 Latin form, or (2) unless they are intentionally modified to give thegi such a 

 form. If vernacular names be admitted in the way in which they have often 

 been used for Mammals, there is no reason why any vernacular term should 

 not claim recognition, to the eventual destruction of Latin nomenclature. The 

 only exception to this lies in the use of proper names of persons and places, 

 which are acceptable when provided with the proper Latin suffixes, which 

 should be attached to the name as a whole (unless it terminates in a silent 

 vowel, which may be retained or omitted as euphony may dictate), without 

 subjecting the name to a Latin transliteration ; unless the said name be one 

 which belongs to a language which does not use the Roman alphabet. Ori- 

 ental and Slavonic names, which use an alphabet which is either not Roman, 

 or Roman without Roman values to the letters, must be transliterated so as to 

 exhibit their phonetic value. The requirement of a Latin form, however, does 

 not involve (i) the modification or rejection of an old-established name because 

 of its not conforming to the rule established long after it was proposed. Thus 



