FISHERY BULLETIN: VOL. 71, NO. 1 



few were taken short distances offshore from 

 these locations. 



Numerous field observations were made while 

 diving, and small groups of specimens (2-3) 

 were maintained and casually observed for a few 

 months in 20-gal aquaria. 



Eighty-seven specimens were examined for 

 gonad development. Most were examined super- 

 ficially using a low-power ( 10-30 x) dissecting 

 microscope, and the stage of development of the 

 ovarian portion of the gonad was estimated, fol- 

 lowing the definitions of Smith (1965) and Moe 

 ( 1969) . Sexual maturity (the presence of Stage 

 4 oocytes) was determined on the basis of ovar- 

 ian tissue only, since very small individuals often 

 contained some mature sperm, even though the 

 ovarian tissue was immature. For 22 specimens 

 the gonads were removed, embedded in paraffin, 

 and sectioned. Most specimens had been origi- 

 nally preserved in 10% Formalin' and then 

 transferred to 40% isopropyl alcohol. Gonads 

 to be sectioned were removed from the fish, 

 placed in Bouin's fluid for several days, dehy- 

 drated in ethyl alcohol, celloidin-methyl salicy- 

 late, and xylene, and then embedded in paraffin. 

 Gonads were sectioned at 10 /a, mounted on mi- 

 croscope slides, and stained. Mallory-Heiden- 

 hain stain was used for most slides. This stain 

 was easy to use and yielded good contrast be- 

 tween various tissues within the gonad. 



Stomachs of 31 specimens were removed and 

 their contents examined under a low-power dis- 

 secting microscope. Food items were identified 

 to major group (usually class or order) and 

 counted. The importance of each group was 

 determined by calculating its frequency based 

 upon the total number of food items counted 

 and upon the number of fish containing each 

 food type. 



DISTRIBUTION 



Briggs (1958) gave the range of Serraniculus 

 pumilio as North Carolina to Florida, and the 

 southwestern Gulf of Mexico in shore areas. 

 Subsequent references (Bullis and Thompson, 



" Reference to trade names does not imply endorse- 

 ment by the National Marine Fisheries Service, NOAA. 



1965; Cervigon, 1966) indicate that it occurs 

 along the coast of the western Atlantic Ocean 

 from North Carolina to Guyana. Most collec- 

 tions have come from the Atlantic coast of the 

 southeastern United States and from the eastern 

 Gulf of Mexico as a result of the numerous fish- 

 eries surveys in these areas (Reid, 1954; Spring- 

 er and Bullis, 1956; Springer and Woodburn, 

 1960; Moe and Martin, 1965; Moe et al, 1966; 

 Starck, 1968; Struhsaker, 1969). The species 

 has also been collected in the western Gulf of 

 Mexico off Texas (Ginsburg, 1952; Hildebrand, 

 1954) and in Campeche Bay off Mexico (Hilde- 

 brand, 1955) . More recent collecting by the RV 

 Oregon and RV Pillsbury off the coast of Co- 

 lombia indicates that the species is also common 

 in the Caribbean Sea (C. R. Robins, pers, 

 comm.). Two specimens have been taken by 

 the Pillsbury from off the coast of Honduras. 

 Serraniculus is unrecorded from the Bahamas 

 (Bohlke and Chaplin, 1968) and apparently also 

 absent from the other islands of the West Indies. 

 Numerous species of fishes have similar conti- 

 nental distributions in the western Atlantic, 

 while other species are restricted to the coral 

 reef areas of the islands and portions of the 

 Central and South American coast where the 

 continental shelf is narrow. The factors which 

 prevent the distribution of continental species 

 in the islands are not completely understood, 

 but probably include differing ecological condi- 

 tions, as well as competition from closely related 

 and better adapted island species (Robins, 1971) . 

 Physical barriers are undoubtedly not important 

 since many of the species have pelagic larval 

 stages, and some free-swimming species such as 

 Scomberomorus maculatus are also absent from 

 most of the West Indies. Ecological parameters 

 which may be important are temperature, bot- 

 tom type, salinity, and turbidity. Temperature 

 may not be important for Serraniculus pumilio 

 since it is distributed well into the tropics, but it 

 is apparently more tolerant of low temperatures 

 than are most of the coral reef fishes. Serranic- 

 ulus may prefer continental sediments rather 

 than coral reef debris, but accurate descriptions 

 of substrates where it has been collected are un- 

 available. In general, the continental species, 

 including S. pumilio, are more tolerant of vary- 



236 



