508 CYCLES OF ORGANIC AND INORGANIC SUBSTANCES 



academic, only a moment's thought is needed to appreciate their 

 relevance to fertility, for example. In the past this has been 

 regarded in terms of basic nutrients, water movements, etc., but 

 there are good reasons now for taking into consideration also the 

 available metabolites: first, because (in some instances at least) 

 we know that phosphates and nitrates alone may be insufficient, 

 but even more because in many important respects fertility alone 

 — overall production, indeed — may be irrelevant, however great. 

 I have touched previously (1956) on this aspect, which assumes 

 greater importance as we learn more about the nutritional needs 

 of particular organisms and their tendencies to select some foods 

 and /or avoid others. Rich phytoplankton production alone does 

 not govern the successful growth and spat-fall of oysters, but 

 rather the production of a \ery small number of specific algae on 

 which alone the oysters can thrive (e.g., Davis and Guillard, 1958) : 

 these in turn may depend on specific chemical requirements being 

 met (e.g., Hutner et al. 1958; McLaughlin, 1958), or particular 

 harmful ectocrines not reaching dangerous concentrations (Davis 

 and Guillard, 1958; Loosanoff, in discussion of Wilson, 1958, p. 

 99). To the oyster, all the other species are weeds! Plaice production 

 again does not depend merely on the gross success of local in\'erte- 

 brates but rather, at a critical postlarval phase, on the availability 

 of a few, perhaps only one or two, zooplankton species (Shelbourne, 

 1957). We can now expect these in turn to have their specific 

 needs ! Yet another example is pro\'ided by the work of Shivaishi 

 and Provasoli (1959) on the factors required for the growth of 

 Tigriopiis japonicus. As my colleague. Air. Steele, has so rightly 

 said (1958), emphasis on measuring the grosser aspects of pro- 

 ductivity has tended to distract attention from specific compo- 

 sition. Yet, species selection is as fundamental a process in con- 

 temporary ecology as in that large-scale ecology which is evolution 

 itself. Indeed, along with the evolution (in both senses) of free 

 metabolites has proceeded the evolution of new ecological niches 

 and links — and often enough their occupation must ha\'e created 

 yet further possibilities of evolution. Doubtless nowhere is this 

 more evident than in aquatic media, as Waterman (1958, p. 348, 

 commenting on Alargalef) has pointed out. 



