In general, the water types are relatively 

 uniform from east to west. There is evidence 

 of seasonal movement toward and away from the 

 Equator, and the area covered by each water 

 type varies seasonally. These movements are 

 interpreted as reflecting seasonal changes in the 

 location and intensity of the current system. 



An approximate relation is shown between 

 the occurrence of tunas and the location of vari- 

 ous surface water types. The albacore in their 

 feeding stage appear in the western K (Kuroshio) 

 type, while albacore in the spawning stage are 

 found in the southern N (North Equatorial Cur- 

 rent) surface water type. Similarly, bigeye, 

 yellowfin, and striped mar lin appear to be asso- 

 ciated with specific surface water types, often 

 appearing in water of two different types without 

 being abundant in an intermediate third type 

 which geographically separates the two regions 

 of abundance. 



Yuen, Heeny S. H. 



Experiments on the feeding behavior of 

 skipjack at sea. /Conference Paper 

 IV - 37 



These experiments were conducted at 

 sea from the Charles H. Gilbert, research ves- 

 sel of the Bureau of Commercial Fisheries Bio- 

 logical Laboratory, Honolulu, on skipjack 

 schools which had been attracted to the ship by 



chumming with live bait and were being hooked 

 by pole and line. The influence of water sprays, 

 dead bait, various species of live bait, and glit- 

 ter (a material obtained from artist 1 s supply 

 shops) on the feeding response of skipjack was 

 investigated. Response was measured by catch 

 rate (number caught per hook per minute) and 

 by feeding attack rate (number of attacks on bait 

 or hooks per fish per second). The latter statis- 

 tic was obtained from 16 mm. color film shot 

 with a cine-camera during the experiments from 

 the underwater viewing port at the stern of the 

 Charles H. Gilbert. Water sprays seemed to 

 increase the responses when anchovy (Stolephor- 

 ous purpureus), mountain bass (Kuhlia sandvi- 



censis), tilapia (Tilapia m o s s a m b i c a), and 

 goatfish (Mulloidichthys samoensis) were used 

 as bait, to decrease the catch rate when mullet 

 (M ugil longimanus ) was used, and to h a v e no 

 effect when silversides (Pranesus insularum) 

 were used. The sprays may change the bait's 

 behavior rather than stimulate skipjack directly. 

 When different bait species were used, caranx 

 (Caranx mate) and anchovy elicited greater re- 

 sponses than silversides, and equal responses 

 resulted from alternating anchovy and topminnow 

 (Limia vittata), tilapia and mullet, and tilapia 

 and mountain bass. Live bait was more effec - 

 tive than dead bait, and glitter seemed to raise 

 the attack rate slightly but not the catch rate. 

 Skipjack move toward shiny objects, but the 

 movement of the food seems to provide the 

 stimulus necessaryto result in a grasping of the 

 food. 



