Under certain conditions, and in certain species, large portions of muroid rodent populations may emigrate. Krebs et a1 (1973) found the population density of field voles in a fenced area rose to three times as great as that in a control situation without a fence, and he concluded that at least two-thirds of the animals would have emigrated if it had been possible. The extensive migrations of the Norwegian lemming are legendary, and have been the subject of several intensive studies (Curry-Lindahl, 1962; Frank, 1962; Myllymaki et a1, 1962). Emigration has also been studied in the great gerbil that is found in Central Asia (Naumov, 1975).
The stimuli that trigger emigration are not well known, but many studies suggest that they include the experience of defeat suffered as a result of increasing population density and frequent fighting. Frank (1962) noted that Norwegian lemmings are quite intolerant and that fighting could account for at least some of the "unrest" that seemed to precede migrations. Clough (quoted by Archer, 1970) has also suggested that lemming migrations may be initiated by increased rates of fighting. Krylov (1974) has reported that the emigration of. Clethrionomys corresponds with the advent of the breeding season and the fall dispersal of the young, at which times fighting is maximal. Helkovska (1974) has reported that emigrators of Microtus arvalis are smaller than normal and that emigration comes at a time of peak population density, both to be expected if emigration is triggered by fighting and defeat. Fairbairn (1978a) found that during the breeding season most emigrating Peromyscus maniculatus were subordinate males. Several laboratory studies may provide an analog of the emigration state. Thiessen and Dawber (1972) found that gerbils defeated by a territorial male would embark upon an "emigration" from the test chamber, even under circumstances in which a large proportion of them were drowned in the attempt. Bowen and Brooks (1978) found that male lemmings defeated by other males exhibited high rates of locomotion in some cases, and they speculated that under field conditions these animals might have emigrated. In the Fairbairn study quoted above there was another type of emigration as well; juveniles and breeding males tended to disperse at the end of the breeding season which is the time that one might expect dominant as well as subordinate males to enter into a reproductive postponement state
The preceding data lead to an hypothesis that emigration is a variant of the reproductive postponement state. One could argue in this case the postponement is in terms of space rather than time, that the animal postpones its reproductive behavior not until the next season, but until it reaches a new habitat. In favor of such an argument is the fact that emigrating voles, Microtus townsendii, are more likely to show submissive behaviors than non-emigrating controls when paired with other conspecifics (Krebs, 1978). On the other hand, Myllymaki et al (1962) found that emigrating lemmings were more likely to show an active defense response rather than flight when they encountered other animals.
The locomotion of emigrating animals is qualitatively different from the exploratory locomotion of other animals. Their locomotion is seemingly purposive, as if guided by some kind of directional mechanism (Myllymaki et a1, 1962). They follow odor trails left by other animals, but do not stop often to investigate them (Myllymaki et al, 1962; Naumov, 1975). Emigrating males of Peromyscus manicu1atus, for example, are more active than non-emigrating males, but they less readily explore an unfamiliar maze (Fairbairn, 1978b). These observations all suggest that the behavior might be understood as something qualitatively different from a reproductive postponement state, perhaps more analogous to the migratory state of birds that is under hormonal control (Meier, 1973}.
By analogy to birds, it might be worthwhile to look for hormonal differences in emigrating muroid rodents that might underlie the difference between their behavior and that of the animals which remain behind. It is known that there are protein differences in these two groups (Krebs et aI, 1973), and that there may be genetic differences, since pairs of siblings migrate more often than would be expected by chance Hilborn, 1975). As far as I know, however, no specific hormone differences have yet been looked for.
There is a theory that spring populations of muroid rodents may, in some cases, be an emigrating or "colonizing" generation, while later generations remain on the original territory, often postponing their reproduction until the following year. This theory has been developed by Anderson (1970) on the basis of earlier data and formulations by Soviet ecologists (Schwarz et al, 1964). If this is the case, it would also suggest that a distinction should be made between the states of reproductive postponement and emigration.