untitled

resource spectrum use a broader range of resources because they are less abundant.

Some species, for example 2, 3, and 4, overlap while others such as 2 and 5 do not.

Overlap is necessary (but not sufficient) to demonstrate competition. An example

(Fig. 9.7b) is provided by the range of seed sizes eaten by finches in Britain (Newton

1972). In this case we see that, contrary to the theoretical distribution proposed in

Fig. 9.7a, there is a broader range of seed sizes eaten by these finches in the middle

range than by birds eating seeds at the extremes.

So far we have considered only one resource axis, that is, one variable such as seed

size. When we consider two or more axes the picture becomes less clear cut in terms

of overlap. Take two species, 1 and 2, which overlap along two axes, for example

moisture and temperature as in Figs 9.8a and 9.8b. If we plot the outline of the two

species distributions by considering the two axes simultaneously we see that it is pos-

sible for the two distributions to be distinct (Fig. 9.8c) or to overlap (Fig. 9.8d). Which

one occurs depends on whether individuals show complementarity(i.e. individuals

that overlap on one axis do not do so on the other one (Fig. 9.8c), or overlap simul-

taneously on both axes (Fig. 9.8d)).

An example of complementarity is shown in Fig. 9.9. DuBowy (1988) examined

the resource overlap patterns in a community of seven North American dabbling ducks

all of the genus Anas, by plotting habitat overlap against food overlap for pairs of

species. In winter, when it is assumed that resources were limiting, points for pairs

were below the diagonal line (Fig. 9.9a), indicating complementarity: pairs with high

overlap in one dimension had low overlap in the other. In contrast, during summer

species pairs showed high resource overlap in both dimensions (several points are

outside the line), indicating that species fed on the same food at the same place. In

summary, the change in niche of these duck species from summer to winter results

in lower overlap and by implication lower competition at a time when we would expect

that resources would be limiting. Note, however, that neither the lack of resources

nor interspecific competition was demonstrated, merely that the results conform to

what we would predict if competition had been acting.

144 Chapter 9

Frequency

Moisture

(a) (b)

(c) (d)

Sp 1 Sp 2 Sp 1 Sp 2

Wet Dry Warm Cool

Moisture Temperature

Temperature Temperature

1 1

2

2

Frequency

Moisture

Fig. 9.8Hypothetical
frequency distribution
of species 1 and species
2 along two parameter
gradients: (a) moisture;
(b) temperature.
Outline of the species
distributions when
considering the two
parameters
simultaneously shows
niches that can be
either distinct (c), or
overlapping (d).