distribution of the tuna oatah is shown by the solid line in ^igo 3e<, <i/hen 

 this is compared with Fig,?a„ the temporatuT^es for the tims when the largest 

 catches were made in each nonth are distributed around :-o'nters of remarkably 

 uniform value, with the mode at a little over 18°Co Then taking 50-lOOm 

 as the fishing level the surface temperature!? were corrected to the tempera- 

 tures at these depths by subtracting 0„5°G in January i, 0°C in February^ 

 OobOC^in tiarch, 1.5°C in April, 1.5"C in Wy 2o5°G in June, 3o5°C in July, 

 and 1 C in December, The distribution of "tiis catch in relation to the 

 temperatures obtained m this manner for the bO-lOOm level is shown by the 

 broken line in FigcSa^ Exactly as seen at lOOraj, the most favorable tem- 

 perature is slightly ever 18°C„ Next, changing the point of view, I plot- 

 ted the best catch for each month against the water temperatures during 11 

 months covered by two fishing seasons from liarch 1929 to May 1930 as shown 

 in FigoSb, obtaining exactly the same results as in JFig„3ao Using the same 

 method 1 converted the surface temperatures for the catches in the three- 

 year survey carried out by the Shisuoka Prefecture Fisheries Experiment 

 Station in 1929„ 1930, and 1931*' and plotted them against the catchj the 

 mode remained almost fixed at a little over 18 C throughout the seasono 



If_we look at the monthly tuna fishing situation as given in lakayama 

 and Ando ' we see that catches at 18°-19°C have been most numerous in 

 January, February, Iiaarch^, April, the first half of May^, the latter half of 

 November., and in December,, but in Msy^ June^ and July^ mxjnths when a good 

 deal of fish is taken, (the main fishing grounds being the central and 

 southern portions of the Northeastern Sea Area) it is understood that be- 

 cause of the rise in temperature of the surface waters the favorable fishing 

 temperatures seen at the surface are apparent rather than realo 



Conclusionc, Suitable temperatures for catching tun^j.,, ....^.ch in the past 

 have been thought to cover a remarkably broad range judging only by the 

 surface temperatures, are restricted to an unexpectedly narrow range when 

 the depth at which the fish are caught is taken into account or when a 

 correction is made for seasonal rises in the surface temperature o The 

 studies described above indicate that it is probably correct to consider 

 that in the so. chern part of the Northeastern Area and in the Zunan Area 

 the favorable ■• imperature for tuna fishing is generally 18°- 19^Co-'-^-' 

 There are, hc-'/everj not a few examples of catches being made at rather 

 remarkably low temperatures in the Oyashio current ar-za. and sometimes at 

 remarkably hich temperatures in the Satsunan and Ogasawara areas so these 

 areas are excluded from consideration hereo This paper, in brief j, is 

 intended to point out that in discuss 3ng the favorable temperature in any 

 sea area one should first- of all take into account the temperatures at the 

 depths at which fish are actually being taken^ This means that in scouting 



for tuna fishing grounds in general it is not sufficient to take only 

 surface observations (although this is fairly satisfactory in the winter) 

 and observations of the middle layers are necessary o In case only surface 

 ^obssurvatijons <!can be made it is best to calruTfi-^ ^ an approximation by making 



12) 



An attempt using similar methods to ascertain the favorable tempera- 

 tures for iiiackerel catches in the Japan Sea revealed that they were 

 centered around a^--"oximately 14°-16°C<, 



