BULLETIN OF THE UNITED STATES FISH COMMISSION. 73 



cartilaginous state. Still it is not necessary to prepare the skeleton to 

 determine tlie absence of ossification, for we can establish this easily in 

 unskinned specimens by the flexibility of the jaws. It is very remark- 

 able that this modification of the skeleton is not incompatible with 

 healthly existence, and that it does not prevent the fish in which it is 

 found from attaining a very large size. 



"Those fishes in which ossification is absent are remarkable by reason 

 of the great reduction of the number of teeth, which, although the only 

 parts which become hard by the deposit of calcareous salts, remain how- 

 ever much smaller than in individuals whose skeletons are completely 

 ossified. 



"We can thus understand how such specimens could present char- 

 acters apparently specific, and that they should have been considered 

 by Kaup as types of new species. These considerations have led me to 

 reduce, on an extensive scale, the number of species in the family. 



"So, in the genus Angiiilla, I find but four species: Angmlla vulgaris 

 occurring throughout the northern hemisphere, in the new world as well 

 as the old. Anguilla marmorata and A. mowa of the Indian Ocean, and 

 Anguilla megalostoma of Oceanica. 



"There are at least four distinct types, resulting from the combina- 

 tion of a certain number of characters; but the study of a very large 

 number of specimens belsnging to these four specific types has convinced 

 me that each of these characters may vary independently, and that con- 

 sequently certain individuals exhibit a combination of characters belong- 

 ing to two distinct types. It is therefore impossible to establish clearly- 

 defined barriers separating these four types. 



" The genus AnguUla exhibits, then, a phenomenon which is also found 

 in many other genera, and even in the genus Homo itself, and which 

 can be explained in only two ways: Either these four forms have had 

 a common origin and are merely races, not species, or else they are dis- 

 tinct in origin, and are true species, but have been more or less inter- 

 mingled, and have produced by their mingling intermediate forms which 

 coexist with those which were primitive. Science is not in the position 

 to decide positively between these alternatives." 



11. Number op species of eels in Ameri'a. 



It is the disposition of American ichthyologists, at least, to accept the 

 views of Dareste, and to consider all the eels of the northern hemisiihere 

 as members of one polemori^hic si^ecies. Giinther is inclined to recog- 

 nize three species in i!forth America : one the common eel of Europe, 

 Anguilla vulgaris; one the common American eel, Anguilla. hostoniensis, 

 which he finds also in Japan and China; and the third, Anguilla texana, 

 described and illustrated by Girard, in the Eeport of the United States 

 and Mexican Boundary Survey, under the name of A. texana, which, he 

 remarks, is scarcely specifically' distinct from A. hostoniensis, from which 

 it differs only in the greater development of the lips. The distinction 



