REGIONAL FERTILITY 219 



small in the lower layer. The mean values of the phosphate concentrations at o and 20 m. 

 on the one hand and 80 and 100 m. on the other have been chosen as representative of 

 these two levels. The method appears to be applicable to an area such as this in which 

 the surface waters close inshore are replenished by upwelling of nutrient salts in a zone 

 where vertical mixing is extensive within 100 m. of the surface, and in which these 

 waters drift away from the coast and acquire thermal stratification as they enter upon 

 oceanic conditions. It is not applicable, however, to counter- currents introduced into 

 the upwelling area from the open ocean. 



Phosphate data have been studied by this method at three localities, at Cape Carranza, 

 Antofagasta and San Juan, but elsewhere the catches of phytoplankton were small and 

 our preliminary measurements unrepresentative. In Tables XIV and XV the percentage 

 figures expressing depletion for stations of different phytoplankton concentration have 

 been averaged. The curve in Fig. 62 shows that depletion increases with the phyto- 

 plankton concentration, and the fact that five points out of seven lie on a straight line 

 suggests a direct relation between the two. The relatively big depletion (21 per cent) of 

 phosphate corresponding to the thirteen stations where the volume of phytoplankton 

 averaged less than 25 c.c, would be explained if phytoplankton at these stations had 

 been grazed heavily by herbivorous zooplankton; it is not wished, however, to stress 

 the accuracy of this curve whose straightforwardness was unexpected. 



Conditions on the west coast are thus seen to fall into a series, grading from localities 

 of minimal cool water, nutrient salts and plankton, to localities of active upwelling, 

 rich nutrient salts and rich plankton. At Pichidanque Bay, Arica and Callao in July 

 the weather had been calm for a considerable time and the surface layers were re- 

 latively impoverished of phytoplankton ; at the first phosphate was negligible, at the 

 others it was not estimated. At Caldera such a period of calm had recently been broken 

 by southerly winds and upwelling, with the result that phosphate at the surface had 

 reached medium values but phytoplankton had not had time to develop far. At Anto- 

 fagasta where upwelling seemed to have been in progress for longer, the largest catches 

 were taken at 7-15 miles from the coast, and the most recently upwelled water lying 

 inshore had a small diatom content. At Cape Carranza, San Juan and the lines off 

 northern Peru, wherever phosphate was examined it was rich, and heavy catches of 

 plankton were taken at all of them. Considering the movement in the current the 

 accommodation of the plankton to the hydrological conditions is remarkable. 



The uniformity claimed for this area by many writers might lead to the supposition 

 that it is uniformly fertile over its entire length. The differences in the abundance of 

 phytoplankton, noted above in the separate localities, would be attributable to tem- 

 porary changes in hydrological conditions. In support of this view is the fact that up- 

 welling was marked in all the localities and that they are therefore all productive. The 

 majority of evidence is, however, against this view. The apparent permanency of the 

 centres of major upwelling, together with the other facts from which the existence of 

 large anticyclonic swirls has been inferred, and lastly the adjustment of the phyto- 

 plankton to the recognized upwelling centres, go to show that certain localities are 



DXIII 15 



