Jordan and Evermann. — Fishes of North America. 2833 



Page 682. After Gamhiisia affinis add: 



1000(a). GAMBUSIA TRIDEXTIfiER, Carman. 



Head 4i; depth at anal i^; suout short, not as long as eye, narrow, 

 rounded forward, and blunt. D. 7 or 8; A. 10; V. (5; P. 12; scales 28 to 

 30-8; vertebrae 14 + 17. Mouth medium, directed obliquely upward; 

 lower jaw longer than the upper, which is short, narrow, and protractile. 

 Teeth in the outer series larger, strongly hooked, pointed, broadened 

 somewhat toward the apex; inner series very small, in bands, tricuspid 

 as in Poccilia; pharyngeal with a shoulder. Eye large, longer than snout, 

 3 in head. Fins small, excepting the caudal; dorsal smaller than anal 

 and farther back, its origin about midway from occiput to end of caudal, 

 nearly above the hindmost anal ray, 17 or 18 scales from the head; anal 

 origin midway between snout and end of caudal; farther forward on the 

 male, between the ventrals, and the fin is modified to form an intromit- 

 tent organ about + length of entire fish; caudal deep, as long as head, 

 rounded on hinder margin. Scales large, median series on flank as wide 

 as eye. Intestine short. Light olivaceous, yellowish or brownish, with 

 7 or 8 vertical bars of brownish, separated by light or silvery spaces of 

 equal width, on the sides of the caudal portion, edges of scales darker, 

 the centers or median series more or less silvery ; belly and loAver surfjice 

 of head silvery or golden ; i)eritoneum black, showing through abdnominal 

 wall; occiput dark; top of snout light; a dark line between anal and 

 caudal; dorsal with a faint sjiot or group of puneticulations behind the 

 middle near the base ; other fins plain to dark tipped. (Garman.) Isth- 

 mus of Panama, in fresh water {tridentujer, bearing trifid teeth). 



Gambusia tridentujer, Garman, Cyprinodonta, Mem. Mus. Comp. Zool., xix, No. 1, 89, pi. 

 4, fig. 10, 1895, Isthmus of Panama. 



Pages 688 and 689. Ileterandria versicolor and H. occidenfalis are cor- 

 rectly referr(!(l by Garman to the genus Pcccilia. It is not improbable that 

 H. versicolor is tlie same as PwHlia vivipara, Bloch & Schneider. 



Lebistes is doubtless identical with Pmcilta, as is also Acropoccilia. 

 Avropacilia tridcns is probably identical with Padlia dominicensis, as stated 

 by Garman. 



Page 691. Garman wrongly refers Pwcilia hntleri to the synonymy of 

 P. sphenops. 



Most of the Mexican and Central American species are imperfectly 

 known and imperfectly described. Of these Garman refers the following 

 to the synonymy of P. sphenops, whether correctly or not only a study of 

 adequate material can determine: PacUia mexicana, P. ihermalis, P. peten- 

 cnsis, P. dovii, P. coudnana, P. phimheus, P. fasciatus, and P. spilurus. 



Poccilia paromna is referred, perhaps correctly, to the synonymy of P. 

 rittata. 



Page 696. Garman thinks that Pwcilia randepolU is identical with P. 

 reticulata, Peters, which may be described as follows: 



1032. P(E( ILIA RKTK I LATA, Peters. 



D. 7 or 8 ; A. 8 or 9 ; V. 5 ; scales 26 to 28-8. Depth of body ? and length of 

 head nearly | of the length to the base of the caudal. Males rather more 



