On the problem of trap‐response types of small mammal populations

Abstract

The mark-and-release method is basically useful in the study of population dynamics for small mammals, especially required in the research of long-term successive fluctuations in population parameters ; by this means we can obtain data for the estimation of N, B, P (in the terminology of LESLIE et al, 1952LESLIE et al, , 1953 and home range sizes at regular intervals during a study period, through which the population under study is expected to keep its natural status, suffering hardly so serious disturbance as would be caused by the removal method.However, some difficulties are inherent in the marking method, and the heterogeneity in trappability between marked and unmarked individuals will be most important. In terms of my definition (TANAKA, 1956), the alloresponsive type Seems to occur far more usually than the isoresponsive for all local populations of every species of rats and mice.The earliest paper that adduced a proof for increased trappability of animals after once marked seems to be CHITTY and KEMPSON (1949), and later not a few authors discussed the same problem. In 1956, I proposed three types of trap response in the manner of differentiation of trappability between the marked and the unmarked on the basis of estimated probabilities of capture n (marked) and p (unmarked).The proposal has hereto been supported by but few ecologists, but I am becoming still more certain of its truth by having examined further my data and related results of other workers. Then the further consideration of this problem is here described mainly about its relation to speciation and subspeciation and its effects on the estimation method of population parameters in small mammals. GROUPING POPULATIONS BY THE RESPONSE TYPEAny local populations of small mammals, excepting insectivores for the present, can be grouped under three response types by means of the marking method. A diagrammatic arrangement of the type for rat and mouse populations in terms of interrelation between re-and p-values was exhibited in the previous paper (TANAKA, 1956), and a similar figure, including the further added data (Table 1) which were secured after or had been done before 1956, are also presented here (Fig. 1) ; all the values of n and p were calculated by myself, and a group of open or solid circles belonging to the same species or subspecies is circumscribed with a solid or broken line.

Studies on the effects of bait and sampling intensity on trapping and estimating Wood mice, Apodemus sylvaticus

“…This is in agreement with other Apodemus studies (-) time of initial capture equal to mean time of recapture; (---) change in trapping programmes. (Gurnell, 1972;Tanaka, 1963). No general distinctions between trappability and sex ( Fig.…”

Trap response in woodland rodents

“…Clearly, it would be better to calculate such probabilities of capture from longer trapping periods. Gurnell (1976) reported that the probability of recapture in wood mice was greater than the probability of initial capture from a trapping study spanning three weeks, and this type of relationship has been demonstrated for several wild rodent species (Tanaka, 1963). Fig.…”

“…If the probability of initial capture equals the probability of recapture the regression lines in Fig. 1 should have a slope of 1.0 and pass through the origin; these would represent isoresponsive lines in the terminology of Tanaka (1963). Within each species the two variables are reasonably correlated: the correlation coefficient, r, for voles is 0.80; for mice without spring unmarked males 0.86; for mice with spring unmarked males 0.78.…”

“…The factors affecting trap response are complex (Gurnell, 1976;Kikkawa, 1964) which makes it difficult to draw firm conclusions about the trap response of different species {e.g. see Tanaka, 1963) or different castes or groups (e.g. age and sex) within a species.…”

Censusing Mouse Populations by Means of Tracking

“…Another advantage is that tracking is more independent of marking than is trapping, thereby conforming to the equal catchability assumption of mark-and-sample estimates. Tanaka (1963), Tanton (1965), and Marten (1970b) have reviewed the ways in which unequal catchabilities of marked and unmarked mice can inject a severe bias into markrecapture estimates.…”