Biology Reference
In-Depth Information
grouping provided attack rate
does not increase propor-
tionately with group size; if a
group of N individuals suffer N
times as many attacks as a
singleton, then clearly individuals
will be no safer in a group than on
their own. However, if there are
less than N times as many attacks
at a group of N, then individuals
will still be safer in a group. The
key point of this idea is that even
when there are no other
advantages from grouping, for
example better detection of
predators or communal defence,
individuals will often still be safer
in a group through dilution of the
risk of attack.
100
10
Predicted
1
Observed
0.1
0.01
1
10
100
Number of water skaters in the group
Fig. 6.2
An example of the dilution effect. The
prey are insects called water skaters (
Halobates
robustus
) that sit on the water surface; their
predators are small fish (
Sardinops sagax
). The
fish snap the insects from below, so there is little
possibility that vigilance increases with group size.
The attack rate by the fish was similar for groups
of different sizes, so the attack rate per individual
varies only because of dilution. The 'predicted'
line is what would be expected if the decline in
attack rate with group size is entirely caused by
dilution; this line is very close to the observed.
From Foster and Treherne (1981). Reprinted with
permission from the Nature Publishing Group.
Dilution: evidence
So much for the theory - what
about the evidence for dilution?
Figure 6.2 provides an example
where predator attack rate does
not vary with prey group size, so
here grouping by the prey results
in perfect dilution. In a group of
100, an individual suffers
1/100th the attack rate
compared to being alone. More
often, however, predator attack
rate will increase with group size because larger groups are more conspicuous.
Nevertheless, grouping usually still brings a net dilution advantage. For example, in
the Camargue marshes of the South of France, wild horses are attacked by blood
sucking flies (Tabanidae), which not only remove blood but also transmit bacterial
and viral diseases. During the weeks when these flies are most active, the horses
aggregate into larger groups. Duncan and Vigne (1979) varied group size
experimentally and found that more flies were attracted to larger groups of horses;
nevertheless, attack rate per horse was still lower in a larger group (Table 6.2).
A spectacular example of dilution is the winter aggregations of monarch butterflies
in Mexico, where thousands or millions of individuals assemble in enormous communal
roosts, clothing the trees over an area of up to 3 ha. The monarch is not a very palatable
butterfly but some birds attack them in these roosts. Counts of the remains of depredated
butterflies showed that although larger colonies attracted more predators, predation
rate per individual was lower in a larger colony, so the advantage of dilution outweighed
the disadvantage of greater conspicuousness in a larger roost (Calvert
et al
., 1979).
Grouping can
dilute an
individual prey's
risk of being
attacked
… even when
attack rate per
group increases
with group size
