730 



Fishery Bulletin 102(4) 



resembled adults by this size. No humeral spots were 

 noted in the two individuals examined. 



Discussion 



Finescale menhaden larvae resemble the larvae of other 

 clupeids (Houde and Fore, 1973; Jones et al., 1978; 

 McGowan and Berry, 1983; Hettler, 1984: Ditty et al., 

 1994) in having elongate, slender bodies, light pigmen- 

 tation, and a small head lacking spines. They have a 

 long, straight gut, often with striations along the hind- 

 gut, posteriorly placed dorsal and anal fins, and the 

 vent is always posterior to the dorsal fin base (Jones 

 et al., 1978). Hettler (1984) discussed the separation of 

 individual species of Brevoortia, and Ditty et al. (1994) 

 presented a synopsis of characters to separate clupeid 

 larvae (<15 mm) based on meristic, morphometries, 

 and pigmentation. Finescale menhaden have 43-44 

 vertebrae, whereas gulf menhaden have 44-46. Yel- 

 lowfin menhaden are reported to have 45-47 vertebrae 

 (Houde and Swanson, 1975). The number of vertebrae, 

 which should approximate the number of myomeres 

 in larvae much smaller than those collected in this 

 study, in conjunction with pigment differences have been 

 shown to be useful in separating clupeid species com- 

 plexes (Ditty et al., 1994). In the western gulf, counts of 

 43-44 vertebrae (=myomeres) would separate finescale 

 menhaden from other clupeid larvae such as Sardinella 

 aurita (45-47 vertebrae; Ditty et al., 1994), Etrumeus 

 teres (48-50 vertebrae; Fahay, 1983), and Opisthonema 

 oglinum (45-46 vertebrae; Richards et al., 1974). Spe- 

 cies from the western gulf with similar vertebral counts 

 (Harengula jaguana, 39-42; Houde et al., 1974; and 

 Jenkinsia lamprotaenia, 39-42; Powles, 1977) can be 

 distinguished from finescale menhaden by their larger 

 PAL:SL ratio (>85% at 15 mm for Harengula vs. <85 l 7c 

 for finescale menhaden. Table 1) and fewer anal rays 

 (13-14 for Jenkinsia vs. 18-24 for finescale menhaden, 

 Table 3). Although vertebral counts were used suc- 

 cessfully in distinguishing finescale menhaden from 

 gulf menhaden, the time necessary to clear and stain 

 larvae makes this method impractical for distinguishing 

 between large numbers of menhaden. 



In larval and prejuvenile stages, finescale and gulf 

 menhaden are morphologically very similar. Propor- 

 tional body measurements overlapped too greatly to 

 reliably distinguish the two species. Only the presence 

 of medial predorsal pigment prior to transformation, 

 stellate melanophores at the pelvic fin base, and the 

 size at transformation were useful characters in distin- 

 guishing the two species. Hettler (1984) noted that gulf 

 menhaden lack paired melanophores in the predorsal 

 region until initiation of transformation. Pigment at the 

 pelvic fin base appears to be a diagnostic character for 

 the small-scale menhadens because Houde and Swan- 

 son (1975) also reported a similar feature in yellowfin 

 menhaden as small as 12.3 mm. Although the presence 

 of this pigment at the pelvic fin base is proposed to 

 be diagnostic for the small-scale menhadens (present 



study), Hettler's (1984) illustration of a 16.5-mm gulf 

 menhaden shows this pigment. Pigmentation descrip- 

 tions for developing gulf menhaden have not specifi- 

 cally addressed melanophores at the pelvic fin insertion 

 (Hettler, 1984). Finescale menhaden transform at a 

 smaller size (17-19 mm) than any of the other Gulf of 

 Mexico menhadens. Gulf menhaden did not complete 

 transformation until around 25 mm, which is in agree- 

 ment with the reported lengths of 25-28 mm for both 

 laboratory reared and wild-caught individuals ( Suttkus. 

 1956; Hettler. 1984). Yellowfin menhaden reach trans- 

 formation at an intermediate size (20-23 mm; Houde 

 and Swanson, 1975). 



Even as adults, finescale menhaden very closely re- 

 semble gulf menhaden (Hoese and Moore. 1977) and few 

 reliable characters effectively separate them. Only the 

 absence of striations on the margin of the operculum, a 

 single humeral spot (with no hint of trailing spots along 

 the lateral margins I, and more scale rows (60-77 in fi- 

 nescale vs. 36-50 in gulf menhaden; Hoese and Moore, 

 1977; McEachran and Fechhelm, 1998) distinguish fin- 

 escale from gulf menhaden. All other meristics overlap 

 greatly; i.e., counts of dorsal-fin rays, anal-fin rays, 

 pectoral-fin rays, pelvic-fin rays, gill-raker counts, and 

 ventral scutes. Although externally similar, significant 

 differences in internal structure between finescale and 

 gulf menhaden have been documented. Finescale men- 

 haden have fewer branchiospinules and shorter inter- 

 mediate gill rakers than gulf menhaden and, as such, 

 filter mainly zooplankton from the water column; gulf 

 menhaden, in contrast, feed primarily on phvtoplankton 

 and detritus (Castillo-Rivera et al., 1996). 



Based on length-frequency differences seen between 

 the two species (Fig. 1), the reported spawning season 

 for finescale menhaden along the middle Texas coast 

 could be extended to late May. We still do not know of 

 characters that would distinguish finescale menhaden 

 eggs, and yolksac. preflexion, and flexion larvae from 

 other species in the genus Brevoortia. In order to fully 

 describe the development of finescale menhaden, labo- 

 ratory spawning and rearing experiments are needed 

 to fully describe these early-life stages. Houde (1973) 

 presented relatively simple rearing techniques that al- 

 low descriptions of the developmental stages of larval 

 fish (from egg through transformation of larvae to the 

 juvenile stage). These methods have been used success- 

 fully for Atlantic, gulf, and yellowfin menhaden, and 

 presumably, finescale menhaden could be reared with 

 these same techniques if their eggs could be obtained. 

 The rearing of finescale menhaden would also allow the 

 effectiveness of the proposed pigment characters used 

 to separate finescale menhaden from gulf menhaden to 

 be tested. 



Acknowledgments 



This work was performed with funding from the Coastal 

 Bend Bays & Estuaries Program under contract 0203. 

 The Texas Parks and Wildlife Department, Coastal 



