No. 3555 ANGELFISHES—SCHULTZ 5 
of the spiny dorsal fin, replacing the two middle predorsal vertical 
bars on P. dumerilii. Thus, on the basis of coloration P. dumerilir 
can always be distinguished from P. scalare and P. altum. The 
number of vertebrae averages slightly fewer in P. dumeriliz than in 
the other two species (table 1) and there are also slightly fewer soft 
dorsal and soft anal rays and fewer oblique scale rows (from rear 
of head to midbase of caudal fin) (table 2) than in P. scalare and 
TABLE 2.—Counts recorded for Pterophyllum dumerilii 
Localities Dorsal fin rays Pectoral fin rays 
XE XII XIII 18 19 20 21 22 23 24 10 11 
British Guiana 3 Wat) val 2 
Rio Negro-Amazon 1 8 1 Sieve erly leben 27710 
Rio Negro (types of leopoldi) | 4 20 2 Ce cin a5 2 2 
Mouth of Amazon 3 1 4 i 383 
Holotype of dwmerilii 1 1 1 
Localities Anal fin rays Oblique scale rows 
VI 19 20 21 22 23 24 25 26 27 28] 26 27 28 29 30 31 32.533 
British Guiana 3 2) LS al 
Rio Negro-Amazon 10 2 Deno Tee 83 Th. PP Se mall 1 
Rio Negro (types of leopoldi) Zag) 1 3 Ty a ile aa) eantsh nt Srenttl 
Mouth of Amazon 4 1 2 1 iy <3} 
Holotype of dwmerilii 1 1 1 
P. altum (table 3); however, these differences are not great enough 
to enable me to identify all specimens on the basis of fin ray and 
scale counts, even if the dorsal and anal rays are added to the 
oblique scale rows to form a character index (table 4). 
An examination of table 2, shows that the specimens of P. dumerilia 
collected near Belém, at the mouth of the Amazon have a slightly 
greater number of fin rays and scales than do specimens further up 
the Amazon—Rio Negro systems. 
The counts made on the holotype of P. dumerilw from Parad (now 
Belém) with XII,23 dorsal, VI,24 anal, 11 pectoral fin rays, and 30 
scale rows along the side, fit into the higher end of the frequency 
distributions (table 2) for P. dumerilii; however, since the frequency 
distributions partly overlap for the various localities and the number 
of specimens are not sufficient to establish the extent of variability 
among populations, if such exist, I am recognizing only a single 
species, P. dumerilii. 
P. dumerilit has been collected in the Amazon basin and in the 
Rupununi River of British Guiana (see open circles in map, pl. 1). 
