2 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 119 



The Lebiasinini and Pyrrhulinina are small to moderate-sized, 

 predaceous characids with moderate-sized mouths. The tiny mouths 

 of members of the Nannostomina, however, are adapted for feedmg 

 on very small organisms. They apparently feed primarily on animals 

 that occur on plants, rocks, and other objects and also to some extent 

 on slowly moving, free-swimming organisms. 



Wickler (1957) has described the breeding behavior of certain of 

 these fishes in a preliminary fashion, and there is a very large amount 

 of literature on this subject published by aquarists. However, even 

 though they are excellent aquarium subjects, no detailed behavioral 

 study has been published about them. A breeding pair enters a group 

 of fine-leaved plants or a clump of roots, the male presses its body 

 against that of the female, curves its anal fin around the female's 

 vent, and the female sheds one to a very few eggs. The structm-e of 

 the male's anal fin in several species is modified for aiding fertilization 

 by guiding sperm over the female's vent. At this time the male 

 sheds an undetermined number of sperm and fertilizes the eggs. The 

 eggs are very slightly adhesive and may become attached to plants or 

 fall to the bottom. All the species that I have seen alive (all but 

 Nannostomus digrammus and Nannostomus bifasciatus) lay eggs of 

 about 0.6 to 0.7 mm. The young mature in 8 to 10 weeks and speci- 

 mens may live for as long as 4 years. Their lifespan in aquaria is 

 usually 2 or 3 years. 



Since most of Nannostomina can be bred with relative ease, they 

 would make fine subjects for comparative ethological studies. It 

 woidd be interesting to compare the taxonomic implications of such 

 studies with those derived from the morphological investigation 

 presented here. 



The spotty locality records of the Nannostomina, the frequent 

 damaged condition of the specimens, and the paucity of specimens 

 from many localities have precluded satisfactory use of statistical 

 procedures in analyzing geographical population differences or in 

 recording changes in body proportions correlated with growth. Col- 

 lectors frequently have packed these small, delicate fishes with 

 specimens of other, larger species with the result that the nannosto- 

 minans have been squashed out of shape, making valid measurements 

 impossible. Only specimens in reasonably good condition were 

 measured. For this reason, counts should be considered more re- 

 Uable than measurements in the data presented below. The data 

 available indicate in several instances that many geographically 

 distinct populations of these fishes probably exist. Some of these 

 differences are pointed out in the discussion under each species. It 

 would be foohsh to designate these different population samples as 

 subspecies in view of the inadequate number of specimens and data 



