the same latitude as the ridge in the dynamic 

 topography of the deeper isobaric surfaces. 

 In the southern part of the area there are a 

 number of inversions, such as the cell of 

 greater than 5. 5 ml. /I. which is centered at 

 150 m. at 31°00'N. on 147°W. Further indi- 

 cations of the variability of either the oxygen 

 consumption or the mixing in this layer are 

 shown by the lack of similarity between the 

 oxygen profiles and the sigma-t profiles (figs. 

 38-47) and the abrupt departure of the percent- 

 age of saturation isopleths on 165 W. (fig. 72) 

 from the oxygen isopleths in the southern part 

 of the area. 



The layer having an almost uniform 



bat large negative gradient, up to 2. ml. /I. 



per 100 m. , extends down to the 1.5-ml. /I. 



o o 



isopleth on 160 and 165 W. and to the 



1. 0-ml. /I. isopleth over the remainder of the 

 area. The sigma-t (figs. 38-47) and percent- 

 age of saturation isopleths are basically paral- 

 lel to those of dissolved oxygen in this zone, 

 indicating that the rate of consumption and re- 

 plenishment via diffusion and lateral mixing 

 must be in equilibrium. The most interesting 

 feature of this layer is that it also contains the 

 deep salinity minimum. The center of the 

 salinity minimum fluctuates between the 3.0- 

 and the 4. 0-ml. /I. (0 2 ) isolines. This 



variation is probably due to the difficulty of 

 locating the center of the minimum on the sta- 

 tion because of the small salinity gradients. 



The decrease in the dissolved 

 oxygen gradient below the 1. 0- and 1. 5-ml. /I. 

 surfaces and its final shift to a positive gradi- 

 ent resulted in a minimum which, as reported 

 by Sverdrup et al. (1942, p. 729), was about 

 400-500 m. below the deep salinity minimum. 

 Because of the great thickness of the band of 

 water with low oxygen content, the large spa- 

 cing of the bottles at the depth at which it oc- 

 curred, and the fact that it was not reached by 

 the 1 , 000-m. casts on the northern part of the 

 two westernmost transects (stations 7-25), no 

 attempt has been made to contour it. However, 

 sufficient data were available to show that a 

 line drawn through the minimum between sta- 

 tions 7 and 75 would approximate the 1, 000-m. 

 contour. The deeper values were to the north- 

 west and the shallower to the southeast, indi- 

 cating that the depth of the minimum tended to 

 decrease from northwest to southeast. 



The isopleths of the oxygen content 

 at the minimum had a radically different 

 pattern from its depth contours. The lowest 

 values formed a tongue of less than 0.4 ml. /I. 

 which entered the area between stations 62 



and 75 and extended in a west-southwest direction 

 as far as stations 54, 47, and 48. The highest 

 minimal values formed a tongue of greater than 

 1.0 ml. /I. which entered the area between sta- 

 tions 4 and 5 and extended eastward as far as 

 station 28. 



The only sigma-t surface on which 

 there was a significant difference between the 

 sigma-t and the dissolved oxygen contours was 

 the 26.8 surface (fig. 83). On this surface the 

 oxygen decreases in the direction of flow, indi- 

 cating that oxygen is. either being consumed by 

 biological activity or depleted by diffusion. Evi- 

 dence of the latter is shown by the decrease along 

 the 600-m. contour from 3.01 ml. /I. at station 

 47 to 2.06 ml. /I. at station 28, while in contrast 

 the oxygen minimum increases from 0. 37 ml. /I. 

 to 1.02 ml. /I. 



Dissolved Inorganic Phosphat e 



The primary purpose of making 

 dissolved inorganic phosphate determinations was 

 to determine if there had been divergence or con- 

 vective mixing in any part of the area of sufficient 

 magnitude to bring nutrients to the surface layer 

 in excess of the utilization by biological activity. 

 The secondary purpose was to map the vertical 

 distribution of the nutrients. The surface values 

 are shown in figure 84 and the cross sections, 

 except for the 155 W. transect, in figures 85-93. 

 The profile for 155 W. was omitted because the 

 Automatic Servo-Operated Photometer was not 

 functioning properly between stations 30 and 38, 

 and the resulting data were too erratic to be 

 considered reliable. 



Although most of the surface values 

 (fig. 84) were near or below the lower limit of 

 accuracy (about 0. 4 ug. at. /I. ) of the molybdenum- 

 blue method of determining dissolved inorganic 

 phosphate (Wooster and Rakestraw 1951), the 

 consistency of the distribution with the other 

 fields indicates that they portray a valid picture 

 of the distribution. As expected from the Carne - 

 gie data (Sverdrup et al. 1945), the surface phos- 

 phate content decreased from north to south on 

 all sections. 



When the surface phosphate distribution 

 (fig. 84) is compared to the temperature and 

 salinity fields, it is evident that the southerly 

 shift from summer conditions is not entirely the 

 result of convective mixing induced by winter 

 cooling and increased wind mixing. In the north- 

 western part of the area, where the southerly 

 shift of the westerlies is normally the greatest, 

 the surface phosphates were the largest. 



11 



