FISHERY BULLETIN: VOL. 78, NO. 2 



whereas those that eat mainly benthic crusta- 

 ceans the least. Fishes that feed wholly or primar- 

 ily on plankton are not apt to be toxic. Some mol- 

 lusk and echinoid feeders may cause severe cases 

 of ciguatera. The level of toxicity among benthic 

 herbivores and detritus feeders is consistently 



low. 



The food-habit studies of this survey support 

 these generalizations. Seven of the eight most 

 toxic species are piscivorous. The one other, Leth- 

 rinus kallopterus, appears to feed mainly on 

 echinoids and mollusks. No specimens ofLutjanus 

 fulvus,Epinephelus socialis, and Adioryx spinifer 

 were found to be toxic (although relatively few 

 specimens were collected); these feed more on 

 crustaceans than fishes. Among the herbivores 

 tested, only two individuals of Scarus and one of 

 Kyphosus gave a reaction of 1. A water-soluble 

 toxin as well as ciguatoxin were found in the 

 detritus-feeding surgeonfish Ctenochaetus 

 striatus, but in small amounts. 



The relatively low level of ciguatoxin in sharks 

 is surprising. Because they feed heavily on fishes 

 and are believed to be long-lived, one might expect 

 them to be as ciguatoxic as the larger moray eels. 

 The tropical species of sharks are not as widely 

 eaten as bony fishes. If they were, no doubt more 

 cases of ciguatera would be attributed to them. 

 The species of Carcharhinus appear to prey to a 

 significant degree on pelagic fishes, and when they 

 do feed on reef-dwelling species, they seem to take 

 many plankton-feeding forms. This may in part 

 explain their apparent relatively low level of 

 ciguatoxin. Still another possibility is that sharks 

 may not accumulate as much ciguatoxin in their 

 tissues as bony fishes. 



Because ciguatera can be highly localized to cer- 

 tain reefs or even sectors of reefs, the fish collect- 

 ing was carried out at many different locations at 

 the atolls. No one area was detected as having a 

 notably higher level of toxicity. 



Many of the most dangerous ciguatoxic species 

 are roving predators. Examples are the bar- 

 racudas, jacks, dogtooth tuna, emperors, and, to a 

 lesser extent, the snappers. They can be caught at 

 a different area from which they acquired most of 

 their toxicity. The strong localization of ciguatera 

 occurs more where the level of toxicity is high and 

 the smaller more resident species are poisonous. 

 Our fishing has not been sufficiently extensive to 

 demonstrate minor differences in the incidence of 

 ciguatera with locality. 



At Enewetak most of the fishing was un- 



dertaken on the southern part of the atoll, par- 

 ticularly in the vicinity of Enewetak Island, the 

 largest of the atoll . This island had the largest pop- 

 ulation of Marshallese people before they were 

 evacuated from the atoll, and it is expected that 

 it will have the largest number when all have 

 been repatriated. Also it is in this area that 

 most of the long-term disturbances of the marine 

 environment, such as the dumping of unwanted 

 material, have taken place. It is fortunate that 

 ciguatera, though more in evidence at Enewetak 

 than at Bikini, is not a major problem as might 

 have been predicted from the impact of western 

 man on the atoll. 



ACKNOWLEDGMENTS 



The author and associates acknowledge with 

 gratitude the support of the U.S. Energy Research 

 and Development Administration through con- 

 tract E(26-l)-641 with the University of Hawaii 

 and Bemice P. Bishop Museum, Honolulu. 



The following individuals participated in the 

 fishing program in the Marshall Islands: Gerald S. 

 Akiyama, Bruce A. Carlson, the late David Erlen- 

 kotter. Glen H. Fredholm, James W. Fry, Gregory 

 Gahagan, Gerald Gulden, Walter C. Gutjahr, Guy 

 S. Haywood, George MacGuire, Oliver K. McCaus- 

 land, Rhett M. McNair, Robert F. Meyers, Takeo 

 Okamura, John E. Randall, Robert P. H. Ruther- 

 ford, Arnold Y. Suzumoto, and Gordon W. Tribble. 

 In addition, Phillip B. Lamberson, formerly of the 

 Mid-Pacific Marine Laboratory, Harry J. Miller 

 and Russell E. Miller, former resident personnel of 

 the atoll, assisted in the collecting at Enewetak. 



The testing of the toxicity was carried out by 

 James Murphy and Lambert Yamashita at the 

 Hawaii Institute of Marine Biology of the Univer- 

 sity of Hawaii, under the direction of A. H. Ban- 

 ner, except for the last sampling in May 1978 for 

 which an initial screening was made by Yoshi- 

 tsugi Hokama of the Department of Pathology, 

 University of Hawaii, by radioimmunoassay 

 (Hokama et al. 1977). The higher reactions were 

 then confirmed by mongoose feeding at the Ber- 

 nice P. Bishop Museum by Arnold Y. Suzumoto. 



A. H. Banner and Helen A. Randall provided 

 helpful suggestions and reviewed the manuscript. 



LITERATURE CITED 



ADACHI, R., AND Y. FUKUYO. 



1979. The thecal structure of a marine toxic dinoflagellate 



246 



