Comparison of feeding and growtfi 

 of iarval round Fierring Etrumeus teres 

 and guif meniiaden Brevoortia patronus 



Weihzong Chen 



East China Sea Fisheries Research Institute 



300 Jun Gong Road, Shanghai, Peoples Republic of China 



John J. Govoni 

 Stanley M. Warlen 



Beaufort Laboratory, Southeast Fisheries Science Center 



National Marine Fisheries Service, NOAA, Beaufort, North Carolina 285 1 6 



The round herring Etrumeus teres 

 is one of several clupeid fishes, 

 abundant in continental shelf 

 waters of the Gulf of Mexico, that 

 presently is not commercially ex- 

 ploited by the United States, al- 

 though its sibling species E. white- 

 headi is a fishery resource for South 

 Africa (Roel and Melo 1990). The 

 potential annual yield of this latent 

 resource is estimated as 3.3 x lO"* to 

 4.2 xlO^ metric tons for the east- 

 em Gulf (Houde 1977) and 1.1 x 10^ 

 to 1.1 X 10'' metric tons for the en- 

 tire GuJf (Reintjes 1980). Details 

 relevant to the distribution and 

 population dynamics of round her- 

 ring, including elements of its early 

 life history, are presently sketchy. 

 Houde (1977) reported that round 

 herring in the eastern Gulf of Mex- 

 ico spawn from mid-October to the 

 end of May between the 30 and 200 

 m isobaths. He surmised that there 

 is a major spawning area about 150 

 km west-southwest of Tampa Bay, 

 Florida, and a minor area just north 

 of the Dry Tortugas. Off Texas and 

 Louisiana, spawning occurs from 50 

 to 200 km offshore and may extend 

 to the edge of the continental shelf 

 (Fore 1971). Round herring and 

 another clupeid, the gulf menhaden 

 Brevoortia ipatronus, are sympatric; 

 the latter spawns in inshore waters 

 of the northern Gulf at least as far 

 offshore as 130 km with a focus of 

 spawning off Mississippi between 



mid-October and late March (Christ- 

 mas and Waller 1975). 



Differences between adult round 

 herring and gulf menhaden are so 

 obvious that systematists once re- 

 ferred these two species to separate 

 families, Dussumieriidae and Glu- 

 peidae (Whitehead 1963), but their 

 larvae are morphologically similar 

 with one major exception, their jaw 

 structure. (The misperception that, 

 unlike other clupeids, round herring 

 larvae do not possess a swimblad- 

 der (Fahay 1983) has been perpetu- 

 ated in the literature.) At hatching 

 larvae of both species are about 3.0 

 mm notochord length (NL) and are 

 slender and elongate with a straight 

 alimentary canal and a posterior 

 anus (Houde and Fore 1973). Trans- 

 formation to the juvenile form 

 begins at about 18 mm standard 

 length (SL) (Houde and Fore 1973). 

 Round herring larvae develop teeth 

 on their long, spatulate upper and 

 lower jaws at about 6 mm SL 

 (Houde and Fore 1973); but gulf 

 menhaden do not develop teeth on 

 their shorter, less compressed jaws 

 until they are about 10 mm SL (Hef- 

 tier 1984). 



The diets of the larvae of these 

 species might reflect differences in 

 jaw structure and dentition. In ad- 

 dition, differences in diet quality 

 and quantity may register different 

 growth between these species. 

 While feeding and growth of gulf 



menhaden larvae are docimiented 

 (Govoni et al. 1983, Stoecker and 

 Govoni 1984, Warlen 1988), similar 

 information on the early life history 

 of round herring is unavailable. In 

 this paper, we compare the feeding 

 and growth of larval round herring 

 and gulf menhaden. 



Materials and methods 



Round herring and gulf menhaden 

 larvae used in this study were re- 

 moved from ichthyoplankton collec- 

 tions obtained during two cruises 

 (December 1980 and February 1981) 

 using MOCNESS gear (multiple 

 opening/closing nets and environ- 

 mental sensing system). Three sta- 

 tions were occupied, one each at the 

 18, 91, and 183 m isobaths, along 

 three transects (off Cape San Bias, 

 FL; off the Mississippi Delta, LA; 

 and off Galveston Bay, TX (Sogard 

 et al. 1987)). The objective of the 

 sampling plan was to broadly can- 

 vass the continental shelf of the 

 northern Gulf for larval gulf men- 

 haden and two other species (So- 

 gard et al. 1987); larval round her- 

 ring were collected incidentally. 

 Sampling at three discrete depths 

 (surface, in the middle of the upper 

 mixed layer, and within or below 

 the thermocline) assured the collec- 

 tion of adequate numbers of speci- 

 mens. In addition, larvae from a 

 single collection taken at the Mis- 

 sissippi River plume front (Govoni 

 et al. 1989) in December 1982 were 

 examined to augment gut content 

 data. Larvae were preserved in 5% 

 formalin for food analysis and in 

 70% ethanol for growth studies 

 (Table 1). To provide an indication 

 of true dietary differences between 

 species encountering the same food 

 assortment, only those larvae from 

 a single vertically and horizontally 

 discrete collection (Govoni et al. 

 1986) that produced both species 

 were used for diet comparisons 



Manuscript accepted 25 November 1991. 

 Fishery Bulletin, U.S. 90:183-189 (1992). 



183 



