104 vSMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 52 



III 



The only species whose habits are known are several of the genus 

 Cottus} These have been referred by some authors to two genera. 

 Cottiis and Uranidea, but they are so very closely related that what 

 is true of one may be predicated for the other. They agree in all 

 structural details and size as well as appearance and have only been 

 distinguished because Cottus has four soft rays to each ventral fin, 

 while Uranidea has only three; the former includes all the European 

 and most of the American species, while the latter, so far as known, 

 is confined to America. Inasmuch, however, as individuals from the 

 same pond may differ in the number of ventral rays, and even the 

 same individual may have four rays in one ventral and three in the 

 other, the groups must be reunited under the name Cottus." 



The species are so similar in most characters that they can only 

 be distinguished by a close, critical examination. The differences 

 are mainly in the trend and character of the large preopercular 

 spine, the number and condition of the rudimentary spines, the 

 number of rays (especially anal), the relative size of the head and 

 other parts, the presence or absence of palatine teeth (of less sig- 

 nificance than in most groups), the spinescence or smoothness of the 

 skin, the size of the mouth, the character of the nostrils, and the 

 color. According to Jordan and Evermann, there are twenty-two 

 species of Cottus and nine of Uranidea found in the United States 

 and Canada, but no two original investigators, at present at least, 

 would agree as to the exact number. The species are nearly un: 

 form in size, most of them attaining a length of about three to five 

 inches, few less, and few reaching a length of seven inches, or, quite 

 excejitionally, a little more. 



There are no such sexual differences in the Cottines as occur 

 among the Myoxocephalines, although the sexes are readily distin- 

 guishable by the great development of a genital papilla in the male 

 and its absence in the female ; there may also be a difference in the 

 size of the head (it being broader in the mature males than in the 

 females), in the devcli)])mcnt of teeth on the palatine bones, the 



'A singular case of nomenclature occurs in Picvost's article "De la Genera- 

 tion chez le Sechot (Mulus gobio)." This name occurs only in connection 

 with the title, but is reproduced in the reprint of the article in the Annales des 

 Sciences Naturelles (xix) in 1830. Mulus may have originated as a printer's 

 mistake for Cottus; it could scarcely have been meant for a new generic 

 name. 



"For extent of variation in number of rays, see appendix to this article. 



