2 TJ. S. BUREAU OF FISHERIES 



can Academy of Arts and Sciences of Boston in 1846. That work 

 enumerates 739 species distributed in 221 genera. As the fishes of 

 the Rocky Mountain region, Mexico, the Pacific coast, and the West 

 Indies were then scarcely known, this list was very incomplete and 

 the nomenclature naturally was imperfect. 



After an elaborate survey of the Pacific coast fish fauna in 1880, 

 David Starr Jordan and the late Charles Henry Gilbert published a 

 Synopsis of the Fishes of North America (1882), including all species 

 known north of the Mexican boundary. The 1,340 species names in 

 that synopsis were arranged in 487 genera; a few other species 

 appeared in the appendix. 



Jordan's Catalogue of the Fishes Known to Inhabit the Waters of 

 North America North of the Tropic of Cancer (1887) was based on 

 this synopsis and includes a number of additional species. 



In the present memoir we have followed implicitly the rules of 

 nomenclature laid down by the International Commission for Zoo- 

 logical Nomenclature, of which the senior author has been a member 

 since 1908. One single case now under discussion (Carcharias) has 

 not been disposed of fully as yet, and the view recommended by 

 Jordan is adopted here tentatively. An important rule set forth by 

 Jordan and Evermann in 1895 has not been acted upon. This rule 

 reads: "We regard all generic names as dift'erent unless actually 

 spelled alike." In accordance with decisions of the commission, 

 another rule, which we have followed, is this: "No change is to be 

 made in current nomenclature unless the case is fairly clear; as, for 

 example, a matter of priority." In general we have followed current 

 practice rather than accept questionable changes. One doubtful 

 opinion should not be exchanged for another. The original orthog- 

 raphy of all names (misprints aside) has been retained. Some 

 supposed misprints (Cypsilurus, Pachynathus, etc.) have been 

 referred to the International Commission, whose verdicts have been 

 followed. 



In accordance with the decision of the International Commission, 

 we have disregarded (unless revived by later authors) the polynomial 

 generic names of Gronow, Klein, Browne, and other able early 

 writers, who failed to adopt the Linnsean code simply because they 

 had not heard of it. This decision rests not on consideration of 

 justice but upon that of permanence. 



The most serious problem in taxonomy is that of the assignment 

 of generic types to the genera of authors who had no conception of 

 types. The "rule of the first reviser" was adopted by the commission 

 at Boston in 1908, and the present writers have endeavored to 

 follow it. 



By far the most complete work on North American fishes is that by 

 Jordan and Evermann, mentioned above — the Fishes of North and 

 Middle America. In that work the authors recognized 1,113 genera, 

 3,263 species, and 133 subspecies — a total of 3,396 species and sub- 

 species. The check list of Fishes and Fishlike Vertebrates of North 

 and Middle America, by Jordan and Evermann, based upon the above 

 and pu})lished in 1896, enumerates 3,100 species arranged in 1,053 

 genera, these being about double those in the Jordan catalogue of 

 1887, the additions being mainly from Mexico, the West Indies, and 

 the deep seas. 



The late modern period in systematic ichthyology is from 1880 to 

 the present time and is especially signalized by the explorations of 



