Table 4. — Number of months (October-March) that 

 the two-dimensional first moment was either 

 to the north or south, or east or west of 

 the general mean position and x^ tests of 

 significance 



Homogeneity X 20.2 not significant 45.1* 

 X^ on totals 0.32 not significant 



example, (1) the albacore might actually swim 

 at the same velocity as that observed for the 

 first moment; (2) the net distance moved by the 

 albacore might be close to zero even though the 

 CPUE first moments exhibit movements in 

 time--the observed apparent movement being 

 induced by a temporally and spatially variable 

 field of catchabllity coefficients over the fish- 

 ing ground; (3) the albacore might tend to be 

 constrained to a horizontal lamina of ocean, 

 swim in this lamina in a sinusoidal pattern, and 

 travel greater distances and consequently at 

 greater velocities than would be deduced from 

 computing this velocity from initial and termi- 

 nal positions; (4) exactly the same situation as 

 in (3) except that the fish might be constrained 

 to move in a vertical rather than in a horizontal 

 lamina; and as a final example (5) the albacore 

 might be constrained to move back and forth 

 over a straight line traversing distances that 

 are greater than those implied by the motion 

 of the moments. At the present stage of our 

 knowledge, the mode or combination of modes 

 of movement that actually apply to the albacore 

 is difficult to deduce. 



The average rate of apparent movement for 

 the peak midwinter fishing period was com- 

 puted for both the longitudinal and latitudinal 

 directions. We computed the longitudinal ve- 

 locity by subtracting the longitudinal moment at 

 the end of March from the longitudinal moment 



at the beginning of December. The difference 

 in degrees of longitude was converted to miles 

 and divided by the number of days (120) to 

 obtain the average longitudinal movement per 

 day. The average latitudinal movement per day 

 was computed in a similar fashion. 



The average longitudinal movement per day 

 is 6.25 miles (11.58 km.), and the average lati- 

 tudinal movement per day is 1.65 miles (3.05 

 km.), giving a resultant velocity of about 6.5 

 miles (12.04 km.) day"' (table 5). A slight in- 

 crease in longitudinal velocity appears over 

 the years, since the velocities for the first 

 3 years (1950, 1951, 1952) are less than the 

 average velocity of 6.25 miles (11.58 km.) 

 day"' ; all the other velocities are greater than 

 the average velocity except that for 1960. 



Relating the velocity of the moment to actual 

 swimming speeds is rather difficult since there 

 are not many critical studies of an average 

 swimming speed for tunas at sea. Some de- 

 tailed studies such as that by Yuen (1966) have 

 shown that swimming speeds of feeding tunas 

 (yellowfin tuna, Thunnus albacares . and skipjack 

 tuna, Katsuwonus pelamis ) are of the order of 

 one body length per second. This would amount 

 to 86 km. day~'--a speed far in excess of 7 

 miles (13 km.) day"'. 



The cause of the discrepancy between the ve- 

 locity of the albacore moments and the expected 

 swimming velocity provides an interesting area 

 of inquiry. It is well known that tunas swim 

 continuously, probably in order to maintain a 

 flow of water over the gills. The continuous 

 swimming is necessary to provide the flow of 

 water over the respiratory surfaces because 

 tunas, unlike most fishes, have no musculature 

 to pump water over the gills. Continuous swim- 

 ming is facilitated by structural modifications 

 that provide well-developed lifting surfaces 

 such as the pectoral fin (Magnuson, 1970). The 

 lifting surfaces help the tuna maintain hydro- 

 static equilibrium. The pectoral fin of the al- 

 bacore is especially well developed, so the 

 albacore may be able to swim at lower veloci- 

 ties than, say, the skipjack or the yellowfin tuna 

 and still be able to maintain hydrostatic equi- 

 librium. Thus the albacore might be adapted to 

 a slower swimming speed than the other tunas. 

 It is unlikely, however, that structural modifi- 

 cation alone could account for the discrepancy 

 between the velocity of apparent movement and 

 the "potential" swimming velocity. 



Another possibility is that tuna soar like 

 birds (Cone, 1962) utilizing the energy in ver- 



21 



