FisViermen also say that the toxicity results from feeding on poisonous crabs. 

 Matsuo also records this. Examination of the stomach contents of poisonous fish 

 collected in the area said to be inhabited by these crabs showed that most of them 

 had been feeding on small giganids (photograph 6 shows the stomach contents of 

 a specimen of dokuhiraa.li [Caranx melampygus]) and no crab fragments could be 

 found. 



Some persons say that fish which eat coral animals are poisonous, and this 

 theory is also recorded by T'atsuo. Among the poisonous fishes there are some, 

 such as the Hepatidae, whose teeth are constructed in such a way that they are 

 thought to eat coral animals, and some have been found to have coral animals in 

 their stomachs, but not all of the fishes cited in this report eat them. If 

 the fish poison originates in the nematooysts of coral polyps, it must be said that 

 the number of species which get the poison directly by eating the polyps is less 

 than the number of those which get it indirectly by feeding on the coral-eating 

 fishes. 



Another theory is that fish which feed on small poisonous species (Matsuo 

 cites the irb'l , pauij, and ajule , all of which are Marshallese names, the species 

 being unidentifiable) become poisonous, and Matsuo was unable to disprove this 

 theory. 



It is clear that there is no definite connection between the feeding habits 

 of any poisonous fish emd its toxicity, and the author was unable to discover 

 any certain relations hip between the ecology and the toxicity of the 45 species 

 reported in this paper. As far as feeding habits are concerned, some feed on 

 coral (hepatids and callyodontids) , some on small fish (carangids), some on 

 demersal shellfish (sparids), and others eat large fish (sphyraenids), so they 

 have nothing in common on that score. Some swim in the surface waters (barra- 

 cuda), sone live in holes in the coral (morays), some swim in the middle layers 

 (hepatids, labrids), and some live on the bottom (lethrinids), so nothing can be 

 deduced from a consideration of their habitats. There seems to be no connection 

 between ecology and toxicity. 



(3) Variations in toxicity depending on the habitat 



The fishermen at Saipan told the author that there is a definite difference 

 in the toxicity of fish of the same species taken north and south of a line 

 drawn straight out from the government pier at that island. Fish taken even 

 one fathom inside of that line are poisonous, they said. They further reported 

 [Page 132] that the fish at Laulau Bay are not poisonous, while those taken around 

 the poisonous weed beds at Tenian and the places where poisonous crabs occur at 

 Saipan are mostly toxic. In order to test these assertions the author selected 

 three stations, taking the fishermen's theories into consideration, and fished 

 at all three for sazanamihagi [ctenochaetus strigosus ], selecting this species 

 for comparison because it was eRsy to obtain" Seventeen specimens were collected 

 and tested, and it was impossible to establish any difference between the results 

 from the three stations. (The results are reported in Chapter II in the section 

 on the sazanam ihagi .) (it is not possible here to give the exact locations of 

 the stations.) 



For the scattered islands of the Marshalls group Matsuo made a detailed 

 record of the variations in toxicity reported by natives from the various islai\ds 

 who visited Jaluit. He listed the native names for fish at sixteen islands and 

 noted the variations in toxicity. The present writer also inquired of the 



81 



