Dee and Parrish: Reproductive and trophic ecology of Myripristis amaena 



525 



50 100 150 200 250 300 350 400 450 500 



Whole wet body weight (g) 

 Figure 7 



Size-frequency distribution of population numbers 

 and population egg production of the brick 

 soldierfish, Myripristis amaena, at Johnston Atoll 

 (JA). (A) Distribution of whole wet body weights of 

 all 855 specimens measured from recreational 

 catches and scientific collections combined. (B) Dis- 

 tribution of wet weights of only the reproductively 

 mature portion of the population up to300g(/!=396). 

 (C) Distribution of egg production by wet weight of 

 spawners ( n =396 ) based on the spawning population 

 distribution in (B) and the fecundity vs body weight 

 relationship developed from gonad samples. The fish 

 population is arbitrarily truncated at 300 g for (B) 

 and (C), which excludes 25 specimens distributed 

 very irregularly over 18 weight classes, and with 

 sizes much larger than those used to develop the size- 

 fecundity relationship. 



prey that was common in adjacent sandy patches. 

 Harmelin-Vivien (1979) reported that polychaetes 

 were the major prey by weight found in M. 

 bowditchae (=murdjan\ Randall and Gueze, 1981) in 

 Madagascar, and she specifically noted that this fish 

 sometimes fed on the bottom. Polychaetes were a 

 minor prey of M. berndti at Oahu (Brecknock, 1969) 

 and were fairly abundant in M. murdjan in the Flores 

 Sea (ter Kuile, 1989). A polychaete was found in a 

 M. amaena specimen from Kona (Hobson, 1974) and 

 in one from our collections at Puako. In our JA col- 

 lections, they were much more common and abundant, 

 producing the second largest IRI value of the major 

 systematic groups (Table 3). The JAtaxa were Seden- 

 taria, which must certainly have been benthic dwell- 

 ers. At Puako, at least five of our 36 specimens of M. 

 berndti and two of our 21 specimens of M. kuntee with 

 prey contained polychaetes. Three specimens of M. 

 berndti contained several polychaete individuals each. 



Gastropods (presumably benthic species) were re- 

 ported by Hobson (1974) in the diets of M. amaena 

 and M. berndti from Kona. A benthic gastropod was 

 found in the gut of a M. berndti at Puako, and they 

 were fairly common in our M. amaena specimens 

 there. Brecknock (1969) found a few gastropods in 

 guts of M. berndti from Oahu and reported foraging 

 by this species on the bottom of aquaria. One M. 

 berndti specimen from Puako contained part of an 

 arm of a bottom-living ophiuroid. 



It seems clear that M. amaena and some other 

 Myripristis species eat some fully benthic taxa. It is 

 unlikely that such prey make up a major part of the 

 diet for more than a very few Myripristis species. 

 However, the ability of these squirrelfishes to employ 

 this feeding mode enables them to exploit a greater 

 range of food resources, at least on an opportunistic 

 basis. Again, the trophic source seems to be local. 



Trophic role of squirrelfishes in tropical 

 communities 



Holocentrids commonly make up a significant portion 

 of the total natural fish community and are important 

 predators throughout their range. In the uninhabited 

 NWHI, nine species of holocentrids, including M. 

 amaena, made up 4.5% of all individuals in the fish 

 community (Norris and Parrish, 1988). In Tulear, Mada- 

 gascar, three species of holocentrids represented 2.2% 

 of the total fish population (Harmelin-Vivien, 1981). 

 At JA, M. amaena alone provided about 2% of all indi- 

 viduals in the fish community. Together with three other 

 species of holocentrids, it accounted for about 3% of 

 the total fish community there (Dee et al. 2 ). 



2 Dee, A. J., D. K. Irons, and J. D. Parrish. 1985. Johnston Atoll re- 

 source survey; a final report of the initial phase ( 19 Jan 1984-20 Jul 

 1985). Project report to U.S. Army Engineer District, Honolulu, 70 p. 



