204 
PACIFIC SCIENCE, Vol. IX, April, 1955 
Because of the considerable labor contained in the calculation, the computation is con- 
fined only to the western part of the ocean bounded by two meridians, X = 0 and X = 0.2, 
that is, 24° of longitude apart. Choice of the western part of the ocean for the computation 
was made because the circulation patterns in that section are more complicated and hence 
more interesting. In the central part of the ocean we will have indeed a very slow motion 
approximately in east-west direction, while very diffuse meridional motion will exist close 
to the eastern coast. 
We have a number of gyres in the surface circulation corresponding to those obtained 
in Munk’s (1950) and the author’s (1951) results with respect to mass transport. We have 
a broad gyre with strong western current flowing north in the latitudes between 20° and 
40° N and corresponding to the Kuroshio, or Japan Current. We also notice one boundary 
vortex, but the secondary boundary vortex is not distinct. We have a subtropic gyre with 
the western current flowing south. This corresponds to the Mindanao Current. Of course, 
we have a faint subarctic gyre corresponding to the Oyashio, or the Kurile Current. 
On the surface of the Southern Pacific Ocean, we have western currents flowing north 
a little south of the equator and in the subantarctic latitudes. Between these two we have 
a strong current corresponding to the East Australian Current, though actually this 
current never develops so strongly because of many passages connecting the Southern 
Pacific to the Indian Ocean through the numerous islands and archipelagoes in the 
Australian- Asiatic Mediterranean. Had we not these passages together with the Southern 
Antarctic Circumpolar Ocean, we could have a much stronger western current in the 
South Pacific Ocean than. actually observed. 
It looks also rather strange that we do not have any strong westward flow in the 
latitudes between 5° N and 2° S although actually the northern margin of the South 
Equatorial Current is in this zone. This is because the Equatorial Counter Current appears 
in our theoretical result much broader and much more diffuse than actually observed. 
This is also the same in Munk’s and Hidaka’s results. The theory of the Equatorial Counter 
Current has been attacked and explained by several authors (Montgomery and Palmen, 
1940; Neumann, 1947) in some other ways than ours. 
Evaluation of the Coefficient of Vertical Mixing 
The streamlines in Figure 3a are drawn for an interval A \f = 250 X 10 l0 /D z cm 2 / sec 
of the stream function. The velocity can be determined as the ratio Ai/'/Ax, where Aa is 
the actual distance between two consecutive streamlines. Because these diagrams are not 
drawn in a common scale for the north-south and east-west directions, it would be 
rather laborious to compute the magnitude of current velocity for all parts of the Pacific. 
Still it will be easy to determine when the streamlines run in exactly north-south or 
east-west directions. 
The values of the stream function at several points along the 33° N parallel are com- 
puted as compiled in Table 4. Assuming the Pacific Ocean is 10,000 kilometers across in 
its east-west direction, we obtained the velocity of the Kuroshio at its swiftest zone, which 
is located approximately 55 km. off the coast, to be 329 cm/sec, 219 cm/sec, 165 cm/sec, j! 
and 1 10 cm/sec according as we assume D z — 50 m., 75 m., 100 in . , and 150 m., respectively. 
Actual velocity of the Kuroshio has been estimated at approximately 3 to 5 knots, or 
about 150 to 250 cm/sec in its swiftest zone. From Table 4 we recognize that the computed 
velocity of the Kuroshio, assuming for D z a value between 50 m. and 150 m., agrees with 
