untitled

recorded. If competition has been operating we would expect that either the popu-

lation, or reproductive rate, or growth rate of the other species would increase.

Forsyth and Hickling (1998) showed from an incidental removal experiment

through hunting that Himalayan tahr (Hemitragus jemlahicus) are associated with

declining populations of chamois (Rupicapra rupicapra). Competition appears to occur

through behavioral interference, with the larger tahr excluding chamois. Another

experiment, illustrated in Fig. 9.3, examined the competitive effect of voles (Microtus

townsendii) on deermice (Peromyscus maniculatus). Deermice normally live in forests

but one race on the west coast of Canada can also live in grassland, the normal

habitat of voles. Redfield et al. (1977) removed voles from three plots and compared

the population response of the deermice there with that on two control areas. On

one control there were no deermice, on the other 4.7/ ha. All the removal areas showed

increases in deermouse numbers, one going from 7.8/ ha before removal to 62.5/ha

2 years later. At the end of the study, when the workers stopped removing voles,

these animals recolonized, reaching densities of 109/ ha, while deermice numbers

dropped to 9.4/ ha. In another experiment, instead of removing voles, Redfield dis-

rupted the social organization of the voles by altering the sex ratio so that there was

a shortage of females, but the density remained similar to the controls. In this area

deermice numbers increased from nearly zero to 34/ ha. This result suggests that it

was interference competitiondue to aggression from female voles that excluded the

deermice because the density and food supply remained the same.

A similar type of experiment was conducted on desert rodents in Arizona by Munger

and Brown (1981). They excluded larger species from experimental plots while smaller

species were allowed to enter. Plots were surrounded by a fence, and access was con-

trolled by holes cut to allow only the smaller species to enter. There were two types

of small rodents, those that ate seeds (granivores) and those that ate a variety of other

foods as well (omnivores). Munger and Brown predicted that if there was exploita-

tion competition for seeds between the large and small granivores then the latter should

increase in number in the experimental plots, while the omnivore populations

should stay the same; if, however, the increased density of granivores was an

artifact of the experiment (e.g. by excluding predators) then the number of small

omnivores should also increase. Figure 9.4 shows that after a 1-year delay small

granivores reached and maintained densities that averaged 3.5 times higher on the

removal plots than on the controls, but the small omnivores did not show any significant

COMPETITION AND FACILITATION BETWEEN SPECIES 139

20

10

0

60

40

20

(^01973) 1973 1974 1975
Present Removed Recolonized
Area C Area F
Peromyscus
(numbers / ha)
Microtus
Fig. 9.3Population
densities of deermice
(Peromyscus) on two
areas from which voles
(Microtus) were
removed in the years
1973 and 1974. On area
F only deermice were
monitored after voles
were allowed to
recolonize. Deermice
were absent or in low
numbers before and
after the vole removal,
but high during the
removal. (Data from
Redfield et al. 1977.)