9, and 11; primary productivity in area 3; and 

 zooplankton in areas 2 and 11. 



In areas 2, 9, and 11 the means for chloro- 

 phyll a were the higher in January to June , and 

 the means for zooplankton the higher in July to 

 December. This result is hard to understand; 

 a lag of 6 months between maxima of standing 

 crops of phytoplankton and zooplankton seems 

 excessive for tropical waters (see Forsbergh, 

 1963), and it is not even clear why all three 

 areas should have the chlorophyll maximum at 

 the same season. 



Area 3 had a significantly higher primary 

 productivity in January to June (the principal 

 upwelling season) than in July to December, as 

 expected, but the differences between half-year 

 means of chlorophyll a and zooplankton were 

 not significant. Most of the data on chlorophyll 

 a for January to June were taken late in the 

 upwelling season; therefore, the mean may be 

 too low. The zooplankton maximum in area 3 

 occurred in the middle of the calendar year, 

 and values for other months were symmetrical 

 around it (ThrailkiU, 1963). This distribution 

 of values seems to explain the similarity of the 

 nneans for this property for the 2 half-years. 



Perhaps the most interesting feature of this 

 analysis is that no differences were significant 

 between half-year means for areas 4, 5, 6, cind 



10, except for chlorophyll a_ in area 6. The data 

 from areas 4 and 5 were comparatively abun- 

 dant for each half-year. Surface- enriching 

 physical processes operate seasonally in area 

 5, at least, and surface currents also change 

 seasonally to some extent. 



The periods January to June and July to 

 December are not ideal for revealing dif- 

 ferences that result from the enriching physi- 

 cal processes in area 5, because in the most 

 northern part of this area, the Gulf of Tehuan- 

 tepec, this enrichment occurs from about 

 November throughFebruary (Blackburn, 1962). 

 The biological data for area 5 were too un- 

 evenly distributed by months to permit a mean- 

 ingful comparison between November to April 

 and May to October, but they do permit a com- 

 parison between December to May and June to 

 November; the numbers of observations for 

 these periods are 78 and 64 for chlorophyll a, 

 31 and 41 for productivity, and 77 and 82 for 

 zooplankton. None of the differences between 

 means for these periods was significant at the 

 5-percent level of probability. Removal of the 

 observations made in the eutrophic part of the 

 Gulf of Tehuantepec (north of lat. 14° N., west 

 of long. 94° W.) made the corresponding num- 

 bers 54 and 60 for chlorophyll a, 21 and 39 for 

 productivity, and 59 and 72 for zooplankton; the 

 difference between means for zooplankton was 

 then just significant at the 5-percent level (the 

 mean for December to May was the higher), 

 but the other differences were still not signifi- 

 cant. 



Blackburn (1966) showed previously that 

 regressions of standing crops of herbivores 



and primary carnivores upon standing crop of 

 chlorophyll a, and partial correlation coeffi- 

 cients for various combinations of these three 

 variables, were consistent with the assumption 

 of steady-state conditions in the southern and 

 central parts of area 5 and the eastern part of 

 area 6. The season for these steady-state con- 

 ditions could have been short (a few months 

 preceding May, when the measurements were 

 made) or long (possibly the whole year). It is 

 surprising that such a situation should be found 

 at all in a eutrophic area, for reasons given by 

 Gushing (1959a, b). The scarcity of statistically 

 significant differences between property means 

 for different seasons is likewise surprising, 

 and is again consistent with steady-state con- 

 ditions even though existence of such conditions 

 cannot be proven. Possibly the physical 

 phenomena which produce seasonally eutrophic 

 conditions, such as those in the Gulf of Tehuan- 

 tepec, are too restricted in space and time 

 to have much effect on stainding crops andpri- 

 mary productivity averaged for a large area 

 over a half-year. If it were possible to compare 

 adequately based property means for different 

 periods from those mentioned above, more 

 significant differences might appear. 



Further work on changes within the year 

 obviously is needed very badly in the eastern 

 tropical Pacific. Some of the results obtained 

 so far do not entirely agree with expectation. 

 Very little is known about seasonal cycles of 

 biota in tropical oceans, as distinct from 

 tropical neritic waters (Gushing, 1959a, b; 

 Heinrich, 1962). Information about suchcycles 

 could be most valuable for making forecasts, 

 identifying areas in which various hypotheses 

 about trophic relationships could be tested 

 (e.g., those in which steady-state conditions 

 are assumed to exist), and probably for using 

 in other ways. The eastern tropical Pacific 

 would be a good section of ocean in which to 

 make these studies because of the contrasts it 

 seems to offer between eutrophic and oligo- 

 trophic conditions. 



RELATIONSHIPS WITHIN AND BETWEEN 

 TROPHIC LEVELS 



Production of phytoplankton in the eastern 

 tropical Pacific is more likely to be limited 

 by nutrients than by light, but the identity of 

 the limiting nutrient or nutrients is still in 

 doubt. Inorganic phosphate is the only nutrient 

 that has been measured with much regularity; 

 charts of distribution of this property agree 

 fairly well with those of zooplankton (Reid, 

 1962, and references). These charts show 

 phosphate to be generally present in surface 

 water of the eastern tropical Pacific at a con- 

 centration higher than that which has been 

 found to limit diatom growth (Goldberg, Walker, 

 and Whisenand, 1 95 1). We may presume, there- 

 fore, that some other nutrient, perhaps nitrate, 



12 



