Fishes of the Western North Atlantic 405 



copepod, Bomolochus eminens Wilson, was found on the gills of specimens taken at Tor- 

 tugas {134: 370). 



Relationship. Longley discovered that the type material of S. anchovia^ from Rio 

 de Janeiro, Brazil, in Paris, consists of two species that are separable by the number 

 of gill rakers on the lower limb of the first arch (57: 7); he reported that one ex- 

 ample, 1 16 mm SL, has 105 rakers, while the other four, 114— 148 mm SL, have 

 160—178. Although Longley's counts are probably not entirely accurate, the difference 

 may be accepted as valid. Since the specimen with the smallest number of rakers was 

 mentioned first in the original description, Longley accepted it as the type of S. an- 

 chovia\ this specimen was taken by M. Delalande, the other four by M. d'Orbigny and 

 M. Gay; Longley assigned the other four specimens to brasiliensis (p. 409) because 

 they had many more rakers and a more strongly compressed body than the above- 

 mentioned type of anchovia, specimens of approximately equal length having been 

 considered. 



The number of gill rakers given for this "type" by Longley (105) corresponds 

 approximately to the number counted in the specimens herein described (85—100), 

 and since Longley gave no other character that disagrees with these specimens, anchovia 

 has been accepted as available for this species, but with considerable doubt. This doubt 

 results from an apparent absence among many Brazilian specimens of fish that have as 

 low a gill raker count as the northern specimens of anchovia herein described. Indeed, 

 there are slender specimens from Brazil with fewer rakers than in brasiliensis, as herein 

 understood, but the raker counts and the relative pectoral fin length are somewhat 

 intermediate between those for anchovia of the eastern United States coast and the 

 braziliensis from Trinidad and elsewhere. Therefore, they may be at least subspecifically 

 distinct from eastern United States specimens described here. If they should prove to 

 be specifically distinct, then the South American species would stand as anchovia while 

 Poey's name, pseudo-hispanica, may be available for the northern species. However, 

 this cannot be definitely determined from the material at hand. In fact, specimens from 

 Cuba, presumably identified as pseudoharengus by Poey, seem to belong to brasiliensis 

 rather than to anchovia. 



In the northern part of the range, S. anchovia., most frequently listed in the past 

 as Clupanodon pseudohispanicus, has frequently been confused with Clupea harengus (the 

 common Atlantic herring). This has led to some confusion, as pointed out by Longley 

 (^7: 6) and others. The two species, indeed, resemble each other superficially in having 

 approximately an equally elongated body, but anchovia is easily distinguishable from 

 C. harengus by the more anterior position of its dorsal fin, absence of vomerine teeth, 

 fewer ventral scutes, fewer vertebrae, a smaller eye, and the enlargement of the last 

 two anal rays. These differences are all shown in detail in the descriptions. 



S. anchovia of the American coast probably was incorrectly synonomized with S. 

 aurita of Europe and elsewhere by Regan in his revision of the genus {106: 378). This 

 tentative conclusion was reached by Longley after he had examined the type material 

 of both species in Paris (see above). From Longley's data it appears that aurita grows 



