48
M. E. SOLOMON
laboratory model illustrated in Fig. 17. The curves tended to be sigmoid in the rising phase (Fig. 18). In a subsequent review, Holling
(1961) expressed the opinion that “the functional response curves of
vertebrate predators in general seem to have an S-shaped rise to a
plateau”. In support of this contention he quoted Leopold (1933) as
suggesting that vertebrate predators attack scarce prey by chance
but develop the ability to find a greater proportion where the prey
become abundant. He referred to the experiments of De Ruiter (1952)
in which tame jays (Garrulus glandarius) were shown to develop a
lively interest in twig-like objects after stick-like caterpillars had been
placed among sticks in which the birds had previously soon lost interest.
He also referred to the proposal of L. Tinbergen that woodland birds
preying on a variety of insects developed a “searching image” of any
acceptable species that became abundant. An account of the work of
Tinbergen and his colleagues on this subject has now appeared in
English (Tinbergen, 1960; Mook et al., 1960) and deserves close study
by ecologists. The following is part of Tinbergen’s summary. “When a
new species appears in the environment, its risk is low at first, and then
increases suddenly. A detailed examination of this phenomenon leads
to the hypothesis that tits (Parus spp.) when searching for prey concentrate on one or a few species at a time, and that, by a kind of
learning process, they adopt ‘specific searching images’ for these species.
The main factors that determine whether or not this process will take
place are discussed. . . . Among these, density of the prey species is
important. It was found that the relation between the density of a prey
species and its percentage in the food cannot be explained from probability of encounters alone. At low densities, consumption is lower
than would be expected on that basis. At moderate densities it is
unexpectedly high, and at high densities it falls again below expectation. The difference in risk between low and high densities is explained
by assuming that the birds do not adopt a specific searching image for a
species of prey that is scarce. The decrease in risk at high densities is
supposed to be due to the fact that, in order to obtain a sufficiently
varied diet, the birds stop using a searching image when the species
concerned forms more than a certain critical percentage in the total
food.”
Another study of the density-relationships between birds and their
insect prey is summarized by Mook (1963). He gives curves for the
responses of the bay-breasted warbler to changes in the density of
larvae of the spruce budworm. The curve for the functional response
is identical in general form to the sigmoid plateau curves for Holling’s
small mammals (Fig. 1 8 ~ ) ,
although the concave-upward aspect of the
slope is again not firmly established. The curve of the numerical re-
M. E. SOLOMON
laboratory model illustrated in Fig. 17. The curves tended to be sigmoid in the rising phase (Fig. 18). In a subsequent review, Holling
(1961) expressed the opinion that “the functional response curves of
vertebrate predators in general seem to have an S-shaped rise to a
plateau”. In support of this contention he quoted Leopold (1933) as
suggesting that vertebrate predators attack scarce prey by chance
but develop the ability to find a greater proportion where the prey
become abundant. He referred to the experiments of De Ruiter (1952)
in which tame jays (Garrulus glandarius) were shown to develop a
lively interest in twig-like objects after stick-like caterpillars had been
placed among sticks in which the birds had previously soon lost interest.
He also referred to the proposal of L. Tinbergen that woodland birds
preying on a variety of insects developed a “searching image” of any
acceptable species that became abundant. An account of the work of
Tinbergen and his colleagues on this subject has now appeared in
English (Tinbergen, 1960; Mook et al., 1960) and deserves close study
by ecologists. The following is part of Tinbergen’s summary. “When a
new species appears in the environment, its risk is low at first, and then
increases suddenly. A detailed examination of this phenomenon leads
to the hypothesis that tits (Parus spp.) when searching for prey concentrate on one or a few species at a time, and that, by a kind of
learning process, they adopt ‘specific searching images’ for these species.
The main factors that determine whether or not this process will take
place are discussed. . . . Among these, density of the prey species is
important. It was found that the relation between the density of a prey
species and its percentage in the food cannot be explained from probability of encounters alone. At low densities, consumption is lower
than would be expected on that basis. At moderate densities it is
unexpectedly high, and at high densities it falls again below expectation. The difference in risk between low and high densities is explained
by assuming that the birds do not adopt a specific searching image for a
species of prey that is scarce. The decrease in risk at high densities is
supposed to be due to the fact that, in order to obtain a sufficiently
varied diet, the birds stop using a searching image when the species
concerned forms more than a certain critical percentage in the total
food.”
Another study of the density-relationships between birds and their
insect prey is summarized by Mook (1963). He gives curves for the
responses of the bay-breasted warbler to changes in the density of
larvae of the spruce budworm. The curve for the functional response
is identical in general form to the sigmoid plateau curves for Holling’s
small mammals (Fig. 1 8 ~ ) ,
although the concave-upward aspect of the
slope is again not firmly established. The curve of the numerical re-