Jumars and Fauchald (1977) discussed the biases of the sampling and sundry
statistical niceties at great length. For example, after deciding that most samples in the
shelf survey done with grabs probably underestimated burrowers, they eliminated
burrowers from the comparisons shown in Figs. 14.13 and 14.14, in order to generate
Fig. 14.15. Examining their graph, they hypothesized that sessile worms are rare
inshore because unstable sediments and turbulent events, which are frequent to depths
as great as 30 m, militate against that lifestyle. Probably it is necessary for successful
adults (the ones that we see in samples) to retain the ability to re-establish themselves
in the bottom and to return to the sediment surface after slumps, migration of sand
waves, and other disturbances. Deeper on the shelf, where sediment stability is higher
and food is abundant, sessile life modes (but not filtering above the sediment) are
more favored. At the greatest depths, where food is least abundant, Jumars and
Fauchald hypothesized that enough food cannot acquired by sitting still. This analysis
is very informative, although in itself it does not prove the hypotheses, only suggests
them. Jumars and Fauchald were clear about that. More deep-sea data were obviously
needed. It is often difficult to commandeer a ship and set off to get needed additional
data. It would have been useful to know something about the proportion of
carnivorous polychaetes, even if they did not fit conveniently on the triangle graphs.
Fig. 14.15 Changes in relative proportions of sessile and discretely motile polychaetes
in samples from different depths in several sampling surveys. Confidence limits
(95%) are for proportion estimates and are based on the binomial distribution.
(^) (After Jumars & Fauchald 1977.)