120 Propositions for rendering the Nomenclature of 



defence for it is, that it is impossible now to identify the species to which the 

 name was anciently applied. But it is certain that if any traveller will take 

 the trouble to collect the vernacular names used by the modern Greeks and 

 Italians for the Vertebrata and Mollusca of southern Europe, the meaning of 

 the ancient names may in most cases be determined with the greatest preci- 

 sion. It has been well remarked that a Cretan fisher-boy is a far better com- 

 mentator on Aristotle's ' History of Animals' than a British or German scho- 

 lar. The use however of ancient names, when correctly applied, is most de- 

 sirable, for " in framing scientific terms, the appropriation of old words is 

 preferable to the formation of new ones*." 



/. Adjective generic names. — The names of genera are, in all cases, essen- 

 tially substantive, and hence adjective terms cannot be employed for them 

 without doing violence to grammar. The generic names Jlians, Criniger, 

 Cursorius, Nitidula, &c. are examples of this incorrect usage. 



m. Hybrid names. — Compound words, whose component parts are taken 

 from two different languages, are great deformities in nomenclature, and na- 

 turalists should be especially guarded not to introduce any more such terms 

 into zoology, which furnishes too many examples of them already. We have 

 them compounded of Greek and Latin, as Dendrofalco, Gymnocorvus, Mo- 

 noculus, Arborophila, fiavigaster ; Greek and French, as Jacamaralcyon, Ja- 

 camerops ; and Greek and English, as Bullochoides, Gilbertsocrinites. 



n. Names closely resembling other names already used. — By Rule 10 it was 

 laid down, that when a name is introduced which is identical with one pre- 

 viously used, the later one should be changed. Some authors have extended 

 the same principle to cases where the later name, when correctly written, only 

 approaches in form, without wholly coinciding with the earlier. We do not, 

 however, think it advisable to make this law imperative, first, because of the 

 vast extent of our nomenclature, which renders it highly difficult to find a 

 name which shall not bear more or less resemblance in sound to some other ; 

 and, secondly, because of the impossibility of fixing a limit to the degree of 

 approximation beyond which such a law should cease to operate. We con- 

 tent ourselves, therefore, with putting forth this proposition merely as a re- 

 commendation to naturalists, in selecting generic names, to avoid such as too 

 closely approximate words already adopted. So with respect to species, the 

 judicious naturalist will aim at variety of designation, and will not, for ex- 

 ample, call a species virens or virescens in a genus which already possesses a 

 viridis. 



o. Corrupted words. — In the construction of compound Latin words, there 

 are certain grammatical rules which have been known and acted on for two 

 thousand years, and which a naturalist is bound to acquaint himself Avith be- 

 fore he tries his skill in coining zoological terms. One of the chief of these 

 rules is, that in compounding words all the radical or essential parts of the 

 constituent members must be retained, and no change made except in the 

 variable terminations. But several generic names have been lately introduced 

 which run counter to this rule, and form most unsightly objects to all who are 

 conversant with the spirit of the Latin language. A name made up of the 

 first half of one word and the last half of another, is as deformed a monster 

 in nomenclature as a Mermaid or a Centaur would be in zoology ; yet we find 

 examples in the names Corcorax (from Corvus and Pyrrhocorax), Cypsnagra 



* Whewell, Phil. Ind. Sc. v. i. p. lxvii. 



