156 REPORT OF NATIONAL MUSEUM, 1898. 



groups, where I have called it " heterology;" that is, the genera of a 

 family may parallel more or less the genera of another; or the families 

 of one order those of another order. It is not, then, remarkable that 

 sometimes one or more species of two or more genera should parallel 

 each other. The reason why the aggregate of appearances is sometimes 

 preferred to single structural differences as definition of the genus, is 

 because it is supposed that the contents of a genus must possess closer 

 "affinity" than species of separate genera. But this is not the case in 

 many instances, and may not be so in any instance where the evolution 

 of generic characters becomes known. Certain it is that in the embry- 

 onic life none of the specific characters appear before the generic. 

 Hence, species may be polyphyletic as regards genera. 



The essential of systematic biology is exact definition. Taxonomy is 

 science only in so far as it is exact. If it be alleged that gradual evo- 

 lution of characters must preclude the possibility of exact definition, I 

 answer that this will only become a practical truth when all intermedi- 

 ate forms have been discovered. And it will never be a practical truth 

 for the botany and zoology of any single geologic epoch or period — as, 

 for instance, the present one — since most of the connecting forms belong 

 to past periods and are not treated of in works in zoology, but in those in 

 paleontology. There is therefore no excuse for the loose and inexact defi- 

 nitions which some taxonomists are accustomed to inflict on the world, 

 which are both unscientific, and serve at the same time to obscure the 

 subject and render it inaccessible to the student. It must be admitted, 

 however, that in a few instances it occurs that transition between two 

 important anatomical characters is discoverable in series of species of 

 the existing period. The special nature of such a case will determine 

 whether definitions with exceptions stated are admissible or not. 



In the matter of nomenclature the author necessarily follows the rule 

 of priority now adopted in all codes, and the definitions of priority 

 contained in the report of the committee of the American Association 

 for the Advancement of Science of 1876. According to these rules, a 

 name can only have a status when accompanied with a description or 

 diagnosis of that to Avhich it is applied, whether species, genus, or divi- 

 sion of higher rank. In the case of species, a recognizable figure is 

 regarded as of equal validity with a description, but a figure possesses 

 no discriminative character as a definition of any higher group. I must 

 here insist on another point, which it has not been thought necessary 

 hitherto to incorporate into any system of rules, since it is one which 

 lies at the basis of all language. That is that names shall be properly 

 spelled. This is necessary on account of the attempts on the part of 

 more or less illiterate persons to construct names for the use of biolo- 

 gists, and on account of the fact that certain other naturalists, not so 

 illiterate, feel bound to adopt their misspellings, and typographical 

 errors as well, on account of the supposed exigencies of the law of 



