the 25.0 surface and of greater than 320 m. in the 26. G surface are indicative of 

 anti-cyclonic flow due to the deflective effect of the islands. In both cases the 

 centers of the cells were closer to the islands than during cruise 10. 



Salinity 



A cell of high-salinity water in the well-defined density gradient was 

 again the most prominent feature of the salinity cross sections (fig. 30). The 

 upper 35.0 °/oo isohaline, which approximated its upper limit over the entire area 

 except east and south of Hawaii and at stations 2 and I4, coincided approximately 

 with the 24.0 sigma-t surface. 



In the area east of Hawaii the salinity maximum was between 34.8 and 

 35.0 °/oo, indicating that water of lower salinity was moving into the area. The 

 station plots indicate that it penetrated at least as far as station 26. 



The surface salinities (fig. 31) were 0.1 to 0.2 °/oo higher than on 

 cruise 10 except in the channel between Maui and Lanai , where a small cell of low- 

 salinity water again indicated runoff, and in the section east of Hawaii, where the 

 pattern was as confused as, and probably the result of, the current pattern (fig. 

 24). At station 29 there was a cell of low-salinity water at the surface which be- 

 cause of the easterly flow indicated by the geostrophic currents could be ascribed 

 to the runoff from the windv/ard coast of Hawaii. At the adjoining station (28) the 

 cell with salinity greater than 34.8 °/oo broke the surface. Considered in conjunc- 

 tion with the dome in the 78.0°F. isotherm, this would indicate divergence in spite 

 of the "high" in the dynamic topography. 



Inorganic Phosphate 



In the surface layers of the ocean there is a continual loss of the dis- 

 solved nutrient salts from the euphotic^ b^'cause of the production of organic mate- 

 rial and a general downward movement of the particulate matter thus formed. In the 

 open ocean the only processes by which nutrients can be returned to the surface 

 layers are vertical diffusion, convective overturn, and divergence (upwelling). To 

 determine whether mixing or divergence of sufficient magnitude to cause enrichment 

 of the surface layers was induced by the complex flow around the islands, inorganic 

 phosphate analyses were made during cruise 12. Phosphate analyses were made be- 

 cause they are more adaptable to shipboard work than those for nitrate, the other 

 major dissolved nutrient salt. 



Only the cross sections of inorganic phosphate (fig. 32) are included, 

 since they are adequate to depict the variations. The 0.6-(ig atoms/1 isopleth 

 formed the upper boundary of the well-defined gradient. It was between the 24.8 

 and 25.2 sigma-t surfaces and at a depth of I50 ra. or greater, except at station 5, 

 where it was at only 52 m. Although the 0.4-|jig atoms/1 surface has been drawn in, 

 the distribution above the 0.6-^ig atoms/1 level was very irregular and bore little 

 resemblance to the other features. This was probably because the lower limit of 

 the accuracy of the method of analysis was being approached (v.ooster and Rakestraw 

 1951). 



DI3GUS3I0K 



The marked similarities of the circulation patterns of the three cruises 

 have been pointed out repeatedly in the discussion of the results. These quasi- 

 permanent features are more understandable if the nature of the island barrier aind 

 the incident currents are explained and their interaction discussed piecemeal. 



The first impression given by the chart of the windward or high islands 

 of the Hawaiian Archipelago is of a series of vertical obstacles lying in a south- 

 east to northwest direction. However, when a cross section of the islands is con- 

 structed (fig. 33), they are seen to resemble a solid barrier having four small 



15 



