The mechanism of replacement of red squirrels by grey squirrels: A test of the interference competition hypothesis

Introduced American grey squirrels have replaced native red squirrels in most of the range currently occupied in Britain and northern Italy. The mechanisms of the replacement are not yet fully understood. We restated the commonly cited Interference Competition Hypothesis (ICH) that grey squirrels interfere with the behaviour of red squirrels in three possible ways: 1. by direct aggressive interactions; 2. by interrupting red squirrel mating-chases; or 3. by forcing red squirrels to actively avoid areas intensively used by grey squirrels. We compared the activity pattern, behaviour and reproductive performance of red squirrels in two study areas in northern Italy, one with only red squirrels (control area C1), the other with both species (experimental area E1). The following predictions were tested: 1. the total time spent in both intraspecific and interspecific interactions by red squirrels increases in the experimental area. 2. most interspecific interactions are aggressive, with grey squirrels being the dominant species; 3. the proportion of breeding female red squirrels that are unsuccessful at weaning offspring increases in area El; 4. grey squirrels take part and interfere with red squirrel mating-chases, and thereby decrease the reproductive output of red squirrel females. 5. the activity pattern of red squirrels in the mixed-species area is shifted with respect to that in the control area to the hours of the day during which grey squirrels show little activity; and 6. red squirrels will shift their home range (or at least their core-area) when grey squirrel densities increase to avoid interspecific core-area overlap. Our results supported only the first prediction of the ICH: they failed to support all the other predictions. Moreover, the increase in the percentage of active time red squirrels spent interacting with other squirrels in the experimental study area was very small (only 1-2 min/day). Red squirrels did not avoid the woodland patches most intensively used by grey squirrels and the interspecific core-area overlap was similar to red squirrel intraspecific core-area overlap. This suggested that red squirrels avoided spatial overlap with grey squirrels in a similar manner as with conspecifics and that an increase in grey squirrel numbers will augment the intensity of resource competition. We therefore conclude that our results do not lend support to the Interference Competition Hypothesis and that interference competition by grey squirrels cannot explain the large-scale replacement of red by grey squirrels that has occurred in Britain and in Piedmont.