occurrence of yellowfin and skipjack in each 

 nnonth of the period 1951-59 in waters between 

 23° and 34° N., regardless of the quantity of 

 tuna taken, and (b) the northern limit of the 

 isotherm 21° C. at a depth of 10 m. as 

 obtained from records of CalCOFI cruises, 

 except in 1959 where it shows the surface 

 21° C. isothern-i as obtained from nnonthly 

 mean sea surface temperature charts of the 

 Bureau of Comnnercial Fisheries. The tem- 

 perature record is incomplete between 23° 

 and 25° N. because most CalCOFI cruises 

 did not extend so far south and the information 

 was not readily available otherwise. It is 

 incomplete or ambiguous to a nnuch smaller 

 extent in waters farther north, as indicated 

 in figure 14. 



The 21° C. isotherm was chosen because 

 its distribution was found to correspond more 

 closely than that of any other with the distri- 

 bution of tuna, although the 20° C. isotherm 

 corresponded nearly as well. The agreement 

 is actually very good; there are exceptions 

 which could be explained by the fact that there 

 are generally more observations on tuna, per 

 1 -degree rectangle and month, than on tem- 

 perature. The northern limit of the isotherm 

 in August 1957 is doubtful and could have 

 agreed much more closely with the skipjack 

 limit than appears in the figure. There is one 

 completely anomalous record of yellowfin in 

 February 1954. The negative temperature 

 anomaly of 1955 and positive anomaly of 

 1957-59, and the ways in which they respec- 

 tively restricted and extended the range of 

 the tunas at the northern limit of their distri- 

 bution, are evident. 



The upper part of figure 14 shows monthly 

 mean zooplankton volumes, averaged for all 

 available stations between 25° and 29° N. 

 (CalCOFI data: Thrailkill, 1959, and unpub- 

 lished). The seasonal peak of zooplankton 

 sometimes precedes and sometimes coincides 

 with the entry of tuna into the area; zooplank- 

 ton volumes tend to be lower in warm than 

 in cool years and are therefore not positively 

 related to northward penetration of tuna. 



The more important question--what deter- 

 mines changes in abundance of tuna within 

 the area of its occurrence at a given time?-- 

 cannot yet be answered. Table 4 summarizes 

 attempts to correlate mean yellowfin and 

 skipjack abundance in 1 -degree rectangles 

 with surface temperature and zooplankton 

 standing crop in the same rectangles in the 

 same months. The only months worth con- 

 sidering were July, August, and September, 

 because CalCOFI cruises (the main source 

 of oceanographic data for the area) are seldom 

 naade during the later part of the tropical 

 tuna season; even in the above-mentioned 

 months such cruises usually only cover the 



area north of 25 N. No meaning can be read 

 into table 4. For the months considered it is 

 probable that the tuna data are much more 

 representative of the sampled 1-degree rec- 

 tangles than are the oceanographic data. There 

 are not many rectangles available in any 

 month, and it must be borne in nnind that 

 5 percent of such an array of correlation 

 coefficients are expected to be misleading. 



Cruise TO-59-2 included a detailed oce- 

 anographic survey of waters south of 26° N., 

 with a great variety of physical, chemical, 

 and biological measurements made at many 

 stations for comparison with contemporaneous 

 data on tuna abundance. This was in August 

 when the fishery would normally have been 

 active, but in the outcome it was inactive 

 because of a dispute about prices. Similarly, 

 cruise TO-60-1 was intended to throw light 

 on the abundance and distribution of tuna 

 in the season and area in which they occur 

 before they arrive in Baja California (fig. 2). 

 Neither set of cruise results has been fully 

 analyzed. 



Middle America 



The region considered is the offing of the 

 Pacific coast of southern Mexico (Mexico 

 excluding Baja California) and Central 

 America. Recent surface temperature anoma- 

 lies in this region were discussed above and 

 shown in figure 5. 



Figure 5 compares the known distribution 

 but not abundance of yellowfin and skipjack 

 tuna in the years 1955-59 in the fourth quarter. 

 In the cold year 1955 and the average year 

 1956 skipjack together with yellowfin occurred 

 along the whole coast; as temperatures rose 

 in 1957 and 1958 skipjack were no longer 

 found in the thermal-equatorial region, al- 

 though yellowfin were; when temperatures 

 dropped slightly in 1959 the range of the 

 skipjack extended slightly farther into the 

 thermal-equatorial region than in the 2 former 

 years, although not nearly as much as in 

 1955 and 1956, and the range of the yellowfin 

 still remained about the same as in all former 

 years. 



From this and similar distribution charts 

 for other quarters of the years 1955-59, the 

 conclusion is drawn that skipjack probably 

 avoid surface waters of very high tempera- 

 ture; the data have not been sufficiently 

 analyzed to show the limiting surface tem- 

 perature but it is probably about 83 F. or 

 28° C. Yellowfin may also be affected by high 

 temperatures, not in their area of distribution 

 but in their abundance within that area; an 

 incomplete study of records of abundance by 

 1-degree rectangles and quarters over the 

 period 1955-59 suggests that yellowfin were 



39 



