Largo iteas of food which can not fit into the stomach as they are, for ex- 

 ample scombroids, needlefish, hairboaks, barracuda, and so forth, are put away bent 

 double. From the shape such fish are in it cannot be determined whether they were 

 swallowed head first or tail first. 



In many cases where, as described above, the heads are lost but the bodies are * 

 still comparatively whole a basis is provided for making a count of the number of 

 individuals . 



It is not known whether or not there is a point of satiation in the feeding of 

 the yellowfin, but in fish which may be thought to have eaten well one may find in 

 a specimen 137 em in length three specimens of Auxis sp. about 30 cm in length 

 along with various other items. 



A total of 3 fish, 2 on the first cruise and 1 on the second, were taken with 

 their stomachs completely empty. This is somewhat over 5% of the whole number of 

 specimens. 



It is a question whether or not the yellowfin is fully capable of selecting 

 its food because, as will be related below, when they are feeding on demersal forms 

 one sometimes finds pebbles, decayed leaves of trees, and so forth in the stomachs. 

 These cannot be considered to have any significance as food for the yellowfin, and 

 are Judged to be clearly matter which has entered the stomachs together vd.th the 

 bottom-dwelling organisms. 



The following paragraphs are in explanation of some of the terras used in Figure 

 3. 



The unidentifiable fish are those in which it is only possible to tell by the 

 skeletal remains that they are fish, but in which it is impossible to tell what 

 kind of fish they are. Consequently the number of specimens cannot be determined 

 either. 



Unidentifiable matter is that concerning which it is absolutely Impossible to 

 tell whether it is of animal or vegetable origin or whether it is detritus from the 

 sea-bottom . 



Cases wore frequently met with in which, although the family name of the sped'* 

 men was known, the genus and species could not be ascertained, and in such cases we 

 were forced to identify the specimen only by the family name. For example, the 

 fishes of the family Carangidae are all lumped together under the designation 

 "carangid". However, where the generic and specific names are known the specimens 

 are treated separately. 



The following is a general consideration of the food habits of yellowfin tuna 

 in the Celebes Sea. 



The most important natural food of yellowfin tuna in the Celebes Sea is fish, 

 followed by mollusks (principally squids), with crustaceans. Jellyfish, and so 

 forth also appearing in the stomach contents. A more detailed examination, however, 

 reveals that squids are the most important single item of food . It is thought that 

 all of the squid are of one species, perhaps the tobiika f Symplectoteuthis 

 ovalaniensis ^ (?), but in so many cases the head and tentacles were missing or di- 

 gestion had proceeded to such an extetit that it was impossible to determine the 

 species . 



.2 



