and behavioural) differences to determining the presence of an interaction can
be seen more explicitly in data from experimental feeding trials, where intra-
specific size ranges of predators and prey were limited, but predators were
exposed to a wide range of prey species that differed substantially in size
(Fig. 6.3). Intriguingly, although prey sizes spanned almost four orders of mag-
nitude, the spread of prey sizes taken was wide and there was little evidence of
systematic changes in the likelihood of prey being consumed along the size axis,
within the limits of prey size taken. The interspersion of species that were
consumed with those that were not indicates size was not a good predictor of
whether individual species will be consumed within the prey range taken. Put
another way, feeding was not contiguous on the size axis. For the eleven pred-
ators examined here, the proportion of prey species (whose sizes lay between
the largest and smallest prey taken) that were actually eaten ranged from 0.27 to
0.86. So, whilst at one level there are clearly quite subtle responses of predators
to size differences among prey, it is important to bear in mind that most of these
relationships come from intraspecific size variation. Using these relationships
to scale up and generalize across diverse prey types might be more complex.
The prey-size niche of individual predators does not explain all the variation
in prey choice, but seems to have certain basic characteristics that recur across
many predator types. The first is that individual predators tend to take preyFigure 6.3Results of laboratory feeding trials, in which 11 species of freshwater
invertebrate predators were presented, in separate, replicated trials, with a range of prey
species of different sizes (Warren, unpublished data). Each vertical line of points is the set
of results for a single predator species (plotted at the average predator size used in the
trials). Each potential predator:prey interaction is represented by a semicircle: the left-
hand semicircles, in black, indicate a prey species that was eaten, and the open semicircle
to the right indicates a prey species upon which predation did not occur. The horizontal
bars are the mean prey sizes taken by each predator.104 G. WOODWARD AND P. WARREN