628 Notes on some South American Fishes. 



slightly behind veutrals ; anal much behind dorsal. Snout pointed. A 

 silvery lateral band from above the eye to the end of the middle caudal rays. 

 Above and below this are brown bands extending the whole length to end of 

 caudal, the lower one forward to tip of snout. Male with a third dark band 

 from the base of the pectoral to the tip of the first anal rays. 



Types : 18 specimens (one a male ?) ; up to 23 mm. long. Lower Amazonas. 



This species is much more elongate than any other of this genus 

 known to me. In shape it approaches more the species of Ciiara- 

 codon, but the teeth are certainly in but a single series. In only 

 one specimen I observed the semblance of a single pointed tooth 

 behind the outer row of from three to four-pointed incisors. 



Poecilia vivipera parse var. nov. 



P. surinamensis and unimaculata Val. P. schneideri C. & V. 



Head 3| ; D. 7 in male, 6 in female ; A. 8 — 9 ; scales 26. Eye about 3 in 

 head, much more than half interorbital. 



Colo7- of females. — Dorsal dotted with black ; scales with dark border. A 

 narrow vertical dark bar two scales high on the sixth or seventh series of 

 scales from the head. This rarely absent. A more or less distinct black 

 blotch on lower posterior part of abdomen. 



Color of male. — Dorsal with two or three oblique spots between each two 

 rays. The rays elevated, reaching in some individuals to the caudal. Traces 

 of an oblique dark bar from middle of root of caudal up and back. Some- 

 times with a short dark streak between two of the upper rays of the caudal 

 and a similar vertical streak between two of the posterior dorsal rays. 



About 150 specimens, taken in the ditches of the Rua das Mon- 

 gubas of Para, probably belong to this species. The greatest dif- 

 ferences between them and the typical vivipera lie in the fact that 

 the dorsal in the female is inserted above the last ray of the anal, 

 and that the anal in the female has 6 rays, in the male 7. If these 

 characters should prove constant, these specimens represent a dis- 

 tinct species. 



A number (12) of other males differ in being slightly darker. The 

 vertical humeral spot is less distinct, but the abdominal spot is much 

 more conspicuous and well circumscribed. There is in addition a 

 dark spot on the root of the caudal, but little smaller than the 

 abdominal spot. The oblique bar of the caudal is conspicuous in 

 these, and there is in addition a straight band along the upper part 

 of the caudal. These are so conspicuously marked that had any 

 females been taken at the same time I should have considered them 

 a distinct species. 



