Role of diatoms in copepod production:good, harmless or toxic?

Nun~erous recent studies have challenged the classical view that copepod production in the ocean is primarily based on pelagic diatoms. Kleppel (1993), in particular, has argued that copepods in nature feed mainly on other microplankters (dinoflagellates, ciliates) which prinlarily account for reproduction, and that diatoms alone provide an insufficient diet for reproduction. More recently, several authors have argued that diatoms can be toxic or deleterious to copepods by preventing their eggs from hatching or causing the eggs to hatch into malformed, non-viable nauplii (Ban et al. 1997 and references therein). While there is now general consensus in the literature that some species of diatoms are nutritionally insufficient for copepod reproduction (Stattrup & Jensen 1990, Kleppel 1993, Jonasdottir & Kiarboe 1996), the idea that they are toxic remains controversial. Whether diatoms are toxic or nutritionally incomplete makes a major difference to the impact on pelagic food web dynamics and to our understanding of the basis for fish production in the oceans. Planktonic copepods account for about 80% of the mesozooplankton biomass in the ocean and thus provide the most important link between lower and higher trophic levels in the pelagic food web. Therefore, if toxic, the presence of diatom blooms would reduce copepod population sizes, mesozooplankton secondary production and, hence, food for planktivorous fishes. In this comment we examine the evidence for the toxicity hypothesis, argue that it has not been rigorously tested in an ecologically meaningful way, and propose a possible avenue to follow. Evidence. Several studies have clearly demonstrated that some diatom diets result in lower egg production and/or hatching success and/or a higher frequency of nlalfornled nauplil than do, for example, dinoflagellate and flagellate diets (Stattrup & Jensen 1990, Jonasdottir & Ki0rboe 1996, Ban et al. 1997). However, these observations do not allow us to decide whether the inadequacies of these diatom diets are due to toxicity or nutntional insufficiency. Copepod eggs are typically rich in lipids (Gatten et al. 1980) and essential fatty acids are therefore potentially limiting nutritional components of diatom diets, as demonstrated by both laboratory studies (Jonasdottir & Kiarboe 1996) and field observations (Pond et al. 1996). Deformities, reduced egg production and hatchability are all wellestablished symptoms of essential fatty acid deficient diets in other marine organisms, including crustaceans (reviewed e.g. in Watanabe 1982, Harrison 1990). There are several other possibly limiting constituents (see Harrison 1990). Potential toxic compounds, on the other hand, have not yet been identified, with the exception of a few diatom species that have toxins similar to those isolated from many dinoflagellates (Bates et al. 1993). Probably the strongest evidence from feeding studies that diatoms can have deleterious effects on egg production and hatchability was provided by Chaudron et al. (1996). These authors found, in accordance with other studies, that negative effects on egg hatching in Calanus helgolandicus only occurred after several days of feeding on the examined diatoms. They also found that egg hatching success and the length of the time-lag both decreased with increasing concentration of diatoms and, hence, feeding and egg production. They considered this evidence that an inhibitory compound was accumulating in the gonads of the copepod. However, an alternative explanation is equally well supported by their observations, i.e. that the copepod exhausts its own storage of a limiting