9780521861724htl 1..2

that are considerably larger than their predators, particularly in marine sys-

tems. This situation arisesparticularlywhere modularorganisms,suchas cnidaria,

are involved, whose true ‘mean individual’ body mass is difficult to assess.

However, it seems likely that many instances of predators being smaller than

their prey in summary webs relate to species averaging effects, because whilst

links are portrayed as being between species pairs, this is not the same as there

being feeding interactions betweenallindividuals of those species. All that is

required to obtain a feeding interaction between species is that individuals from

some part of the size spectrum of the consumer population feed on individuals

from some part of the size spectrum of the resource population. In addition, if

the positions and shapes of the feeding niches themselves are a function of

predator size, and there is a frequency distribution of predator sizes that make

up a population, the feeding niche of the species (or population) is itself the

(a)

log

10

prey body-mass (mg)

–6

–4

–2

0

2

4

–3 –2 –1 0 1 2

–2

–3

–1

0

1

2

log 10 predator body-mass (mg)

(b) (c)

2.5

2

1.5

1

0.5

(^00) 0.5 1 1.5 2 2.5
Mean length of consumer (cm)
Mean length of resource (cm)
–6 –4 –2 0 2 4
Figure 6.4Species-averaging
effects in food webs.
(a) Community matrix for
Skipwith Pond with species
ordered according to body length
(mm), with black circles
representing feeding links and
white circles representing the
average size across all prey species
for each predator. (b) Community
matrix for the Broadstone Stream
food web, with species ordered
according to body mass, with
circles representing feeding links,
and the area of each circle
corresponding toper capita
interaction strength (log-
transformed). (c) Individual
predation events for the
Broadstone Stream food web
(n¼1825 observations). The solid
diagonal lines in all three plots
represent the 1:1 line.
106 G. WOODWARD AND P. WARREN