1/6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 52 



has teeth in both jaws; Manta (or Ceratoptera) has teeth confined 

 to the lower jaw, and Ceratobatis has teeth only in the upper jaw. 

 The species also differ in size and the character of the dorsal spine. 

 While a width of twenty feet or more may be attained by some, 

 others become sexually mature when four feet wide. In most of 

 them the tail is short and the dorsal spine characteristic of Sting- 

 rays is obsolete, but it is asserted to be well developed in the Mobula 

 giorna. 



The number of species of Devil-fishes is uncertain. In 1870 seven 

 species were recognized, five of the genus Dicerobatis (Mobula)^ and 

 two of Ceratoptera (Manta). One representing a new generic type 

 (Ceratobatis) was added in 1897. One of gigantic size, generally 

 supposed to be Manta vampyrus, has been observed at many places. 

 Whether there are more than one species is uncertain.- There is a 

 discrepancy in the length of the tail assigned to some. Most of the 

 giants have a tail nearly as long as the body, but one referred to by 

 Hill, about fifteen feet wide, had a tail only two feet long.^ The spe- 

 cies of Mobula dififer. The M. giorna of the Mediterranean is said 

 to have a tail about three times longer than the width of the disk; 

 the M. japonica one "nearly thrice as long as the body," and the 

 M. olfersii of Brazil and the Caribbean Sea one about as long as the 

 disk and much less than its width. The Ceratobatis robertsii has the 

 tail not much less than twice the length of the disk (620: 350), but 

 considerably less than its width (620:780).* 



One species — the true Devil-fish of the United States, Manta 

 vampyrus — is not uncommon in the warm American waters and ap- 

 pears on the South Carolina coast in summer in "shoals." 



The Manta vampyrus has a body or disk nearly twice as wide as 

 long, and a tail about 6/10 as long as the body ; the body and tail are 

 rough from the development of small tubercles which extend almost 

 everywhere; the band of teeth (confined to the lower jaw) extends 

 over almost the whole width of the jaw and is composed of about a 



* Three nominal species were described later — Dicerobatis draco. Giinther, 

 1872; D. moiis/niin Khinzinger, 1871, and Cephaloptcra tarapacana, Philippi, 

 1894. 



'This subject will be considered in a future article. 

 ' The tail may have been decurtated in youth. 



* The figure in Day's Fishes of India (1878, p. 745), which he "surmises" 

 may represent "Ceratoptera chrenbergii" is nothing but an illustration of a not 

 uncommon monstrosity of an ordinary Ray (Rata) with free anterior exten- 

 sions of the pectorals, resulting from arrest of development. (See Proc. U. S. 

 Nat. Museum, 1895, PP- 195-198.) 



