222 PROCEEDINGS OF UNITED STATES NATIONAL MUSEUM. 



species had been more than once described under different names by 

 Eaflnesqne and Le Sueur. Of these names, Lepomis palUdus Eaf. for 

 the large-mouthed BLick Bass, Micropenis dolomieu Lac. for the south- 

 ern, and Bodianns achk/an Eaf. for the northern variety of the small- 

 mouth have priority over the others. All these, therefore, antedate any 

 precise definition of the name sahnoides. 



The question as to wliether a specific name, at first loosely applied 

 and afterwards precisely fixed, shall claim priority from its first use or 

 not, has been differently answered by different writers, and has perhaps 

 never been settled by general usage. I suppose that the amount of 

 doubt or confusion arising from its use or rejection enters with most 

 writers as an element. The name sahnoides, left unsettled by Lacepede, 

 has been generally received by writers, in consequence of the supposed 

 precision given to it by Cuvier. We have seen, however, that both 

 species were included by Cuvier under one name, and that we must look 

 farther for real restriction of the species. The first distinct use of the 

 name sahnoides for any particular species is by Holbrook, for the large- 

 mouthed form. On the basis of the first unquestionable restriction, the 

 name, if used at all, must be applied to that species. Forty years pre- 

 vious to this restriction, however, the specific name j^a?ZiV7?fs was conferred 

 on the same fish by Eafinesque. 



In the writings of nearly all the older naturalists, as well as in many 

 of the later ones, we find descriptions of species which are really 

 generic in theu" value, and which, as our knowledge of species becomes 

 greater, cannot be disposed of with certainty or even with any high 

 degree of probability, for absolute certainty rarely accompanies any 

 identification. 



In the absence or impossibility of any general rule regarding such 

 cases, the following supposed examples will illustrate what seems to the 

 present writer a fair method of treating them. 



Let us suppose that the genus Micropferus contains two well-marked 

 species; that to one of these the name sahnoides was early applied; that 

 next the names dolomiei and paUidus were applied to the two respect- 

 ively, and that subsequently the name salmoides was restricted to the one 

 called paUidus. 



Now if (1) the original sahnoides were definitely a complex species, 

 distinctly including both, we may hold its author to be a " conservative" 

 writer, and that the subsequent restriction, like the restriction of a 

 genus, is a change of view or the elimination of an error. In this case, 

 the name sahnoides should be retained, dating its priority from its orig- 

 inal use, and applying to the species palUdus. 



If (2) the original sahnoides be not complex, but simply uncertain, tlie 

 probabilities being undeniably in favor of its identity with paUidus 

 rather than with dolomiei, itshouldbe adopted instead of j7rt//</7/^s^ Abso- 

 lute cprtaiuty of identification cannot be expected of many names older 

 than the present generation, and eacli writer must judge for himself of 



