RANDALL: SURVEY OF CIGUATERA AT MARSHALL ISLANDS 



"% 





Figure 51. — Balistoides viridescens, 456 mm SL, 4.5 kg, Tetiaroa, Society Islands. 



It must be emphasized that liver and viscera of 

 the suspect fishes were used in the mongoose feed- 

 ing tests (except for sharks) and not flesh. Because 

 of the much higher level of ciguatoxin in the inter- 

 nal organs than in muscle tissue, low-level toxic- 

 ity (indicated by mongoose reactions of 1 or 2) from 

 liver and viscera would probably not result in a 

 detectable level of toxin if flesh from these fishes 

 had been used in the tests. When the percentage of 

 toxicity is computed for the reactions 3-5 (it is this 

 level at which a human eating the flesh of these 

 species might be expected to fall ill with cigua- 

 tera), the percentage of toxic fishes drops to 16.2 

 for Enewetak and 1.4 for Bikini. 



When one considers the effort directed almost 

 entirely to the worst offenders in ciguatera, the 

 level of toxicity at Enewetak must be regarded as 

 relatively low and that of Bikini decidedly so. Most 

 of these fishes are avoided as adults by islanders in 

 Oceania regardless of the area of capture. There- 

 fore it is concluded that the returning residents to 

 Enewetak and Bikini need not fear at this time 

 any unexpected threat of ciguatera at their atolls. 



Only eight species of fishes produced reactions of 

 4 or 5 in the test animals; that is, severe illness 



or death: Lycodontis javanicus, Cephalopholis 

 argus, Epinephelus hoedtii, E. microdon, Plec- 

 tropomus leopardus, Aprion virescens, Lutjanus 

 bohar, and Lethrinus kallopterus. Had more 

 specimens of Sphyraena barracuda, Caranx ig- 

 nobilis, and Cheilinus undulatus, particularly of 

 large size (none of the specimens taken during this 

 survey approached the maximum size), been col- 

 lected, then they may be expected to be included in 

 the above list (in view of their reputations for 

 causing ciguatera in other areas). 



The moray Lycodontis javanicus was clearly the 

 most toxic of all the species tested, with all indi- 

 viduals producing a reaction of 2 or more in mon- 

 gooses and one-third of them the lethal 5. 



Randall (1958) analyzed the kinds of fishes 

 which have caused ciguatera in terms of habitat, 

 mode of life, and food habits. These species are 

 shore fishes associated with reefs. Usually they 

 are bottom-dwelling generally in < 60 m, but they 

 may be semipelagic open-water forms that range 

 into the reef habitat to feed. They may be car- 

 nivorous or they may feed on benthic algae or 

 detritus. Of the carnivores, those that prey heavily 

 on reef fishes are the most prone to be poisonous. 



245 



