of the Island of Trinidad, W. I. 55 



Caudal forked, with the lobes of equal size. 

 Pectorals moderate, with the external rays longest. 



The species of this genus have been included by all previous 

 naturalists, who have had occasion to treat of the family to which 

 they belong, under the group which has received from Artedi 

 the generic name of Tetragonopterns. Yalenciennes had, indeed, 

 in the Histoire Naturelle des Poissons, observed that his atten- 

 tion had been aiTCsted by the great difference of form which exist- 

 ed between these fishes, and those species for which the Artedian 

 genus was originally chiefly instituted ; but as the variation of 

 form was unaccompanied by any corresponding anatomical 

 characters, he did not deem it advisable to elevate the section 

 to the rank of a genus. 



We, however, believe, that so great a variation in form as 

 that which exists between the species of the first and second 

 sections of Yalenciennes, even if unaccompanied by any other 

 difierence, is of generic value ; and we have therefore separated 

 them into two distinct genera, retaining Tetragonopterns as the 

 generic name of the rhomboidal, or sub-orbicular species, and to 

 the elongated oval species we give the name of Poecilurichthys, 

 in allusion to the bands or spots which are almost always 

 branded on the tail. "We derive the name from the Greek 

 ne;x<Aa«, Variegated, Oyf «, tail, and ix^^h fish ; it is there- 

 fore sufficiently distinct in signification, as well as in sound, 

 from Poecilichthys of Agassiz. 



Another group, which is represented in the island of Trinidad, 

 has the lateral line abruptly terminated at a considerable dis- 

 tance from the caudal fin ; in other respects it has the greatest 

 affinity to Poecilurichthys, having, like the typical species of 

 that genus, two rows of cuspidate teeth in the upper jaw, and 

 one in the lower. We have not been able to study it with 

 much care, as we have only two specimens of a single species, 

 but when it has been dissected, and the pharyngeal bones exam- 

 ined, it will in all probability be found to present additional 



5 



