those of western Atlantic Sardinella sp. (Simpson 

 and Gonzales 1967; Matsuura 1971; Houde and 

 Fore 1973), but their average diameter was 

 greater than for those species. Our B. smithi eggs 

 ranged from 1.21 to 1.34 mm in diameter (mean = 

 1.27 mm) while those of 0. oglinum are 1.10 to 1.28 

 mm (mean = 1.19 mm) (Richards et al. 1974) and 

 those of western Atlantic Sardinella sp. are 1.00 to 

 1.32 mm (mean = 1.12-1.18 mm) (Simpson and 

 Gonzales 1967; Matsuura 1971; Houde and Fore 

 1973). Reintjes (1962) reported B. smithi eggs 

 ranging from 1.15 to 1.48 mm, including both 

 planktonic and artificially fertilized eggs. The 

 eggs of 0. oglinum would not usually occur with 

 those of B. smithi because Opisthonema spawns 

 during spring and summer (Fuss et al 1969; Houde 

 1973a; Richards et al. 1974), while members of the 

 genus Brevoortia are winter spawners (e.g.. 

 Turner 1969; Dahlberg 1970) in waters of south 

 Florida and the Gulf of Mexico. Sardinella eggs 

 could conceivably occur with those of B. smithi, 

 but Sardinella probably spawns farther offshore 

 than does B. smithi. Eggs of B. tyrannus ap- 

 parently are larger than those of B. smithi, the 

 reported diameters ranging from 1.30 to 1.95 mm 

 (Mansueti and Hardy 1967). Brevoortia patronus 

 eggs usually are slightly smaller than B. smithi 

 eggs, the diameters ranging from 1.04 to 1.30 mm 

 (Houde and Fore 1973). Hybrid embryos from ar- 

 tificial fertilization of B. smithi x B. patronus 

 ranged from 1.05 to 1.18 mm (Hettler 1968). 



Larvae of Brevoortia spp. have some distinctive 

 characters that serve to distinguish them from 

 other clupeid larvae with which they may occur. 

 Newly hatched larvae of B. smithi have been pho- 

 tographed by Reintjes (1962), but these pho- 

 tographs fail to show distinguishing characters of 

 larvae in that size range. Hettler (1970a) present- 

 ed illustrations of 7.6- and 11.9-mm TL larvae of B. 

 smithi, but only the 11.9-mm specimen has some 

 characteristics illustrated that help to identify it 

 as a Brevoortia sp. 



Myomere numbers ranged from 45 to 47 in B. 

 smithi, thus separating it from H. jaguana (42 or 

 fewer) (Houde et al. 1974) and E. teres (48 or more) 

 (Houde and Fore 1973). Total myomere counts of 

 B. smithi overlap those of 0. oglinum (45 to 49), 

 Sardinella sp. (45 to 47) (Houde and Fore 1973), and 

 the other species of Brevoortia. Numbers of post- 

 dorsal-preanus myomeres always were fewer than 

 5 in B. smithi larvae longer than 10 mm and never 

 exceeded 6 in smaller larvae (usually 4 or 5). The 

 other identified clupeid genera from this region. 



excepting Etrumeus, have 5 or more (usually 6 to 

 9) postdorsal-preanus myomeres in all length 

 classes, thus serving to distinguish them from 

 Brevoortia larvae. 



Pigmentation of newly hatched Brevoortia lar- 

 vae apparently differs from that of other clupeid 

 genera in its details. Recently hatched B. smithi 

 larvae have pigment on both the dorsal and ventral 

 sides of the notochord tip (Figures 2A-3B), distin- 

 guishing them from other co-occurring clupeid 

 genera, except for some specimens of Harengula 

 (see Houde et al. 1974). Brevoortia tyrannus has 

 pigmentation similar to B. smithi at the notochord 

 tip (Mansueti and Hardy 1967), and we suspect 

 that B. patronus also has this pigment 

 characteristic based on our observations of 

 Brevoortia larvae that were collected in the 

 northern Gulf of Mexico, where B. patronus is 

 known to spawn. Recently hatched larvae of 0. 

 oglinum, Sardinella sp., and E. teres have pigment 

 only on the ventral side of the notochord tip. 



Lateral pigmentation is present on B. smithi 

 larvae as small as 5.2 mm-which is smaller than 

 other clupeids found in their geographical range. 

 At 10 to 12 mm, most of our B. smithi larvae had 

 more than 5 melanophores on their sides, and some 

 had as many as 25. No larvae of Harengula, 

 Opisthonema, Sardinella, or Etrumeus that we 

 have observed have had pigment on the sides until 

 they were at least 15 mm in length. We do not 

 know if B. tyrannus or B. patronus develop lateral 

 pigmentation at sizes as small as B. smithi, but 

 illustrations of B. tyrannus larvae (Mansueti and 

 Hardy 1967) from 8 to 23 mm do not show any such 

 pigment, nor is it mentioned in their text. 



Size at transformation varies among clupeid 

 species. Brevoortia smithi had completed transfor- 

 mation at about the same size as H. jaguana 

 (Houde et al. 1974) and 0. oglinum (Richards et al. 

 1974), at lengths from 20 to 24 mm. However, other 

 species of Brevoortia apparently do not transform 

 until they are of larger size. Brevoortia tyrannus 

 exceeds 30 mm before having a typical juvenile 

 form (Mansueti and Hardy 1967), and the obser- 

 vations and morphological data of Suttkus (1956) 

 suggest that B. patronus does not transform until 

 28 mm or longer. It is possible that our tank-reared 

 B. smithi transformed at smaller sizes than in na- 

 ture, but the seemingly good growth rate and the 

 selection of normal appearing larvae to describe 

 development lead us to believe that B. smithi is 

 transformed at approximately 22 mm. 



672 



