180 arousing situations (such as human disturbance, and enforced separation from their young), and 62% of stereotypic mink performed it in the quiet hours after feeding. Post-feeding stereotypy and inactivity were negatively correlated (r = - 0 . 5 0 2 , P < 0.001 ), but stereotypy and normal activity were positively correlated ( r = 0.35, P < 0.001 ).
Stress and nest-boxes in farmed fox L.L. Jeppesen Institute of Population Biology, Universitetsparken 15, 2100 Copenhagen, Denmark
ABSTRACT Farmed foxes are conventionally kept in barren wire cages without a nest-box for most of the year. Only in the breeding period do they have a nest-box, in which they deliver and raise their young. Without a nest-box, foxes are constantly exposed to potentially stressing stimuli from neighbours and farm personnel. With a nest-box attached to the cage all year round, foxes are able to hide at will, and thereby to control their input of stressing visual stimuli. Whether foxes avail themselves of this possibility to control levels of social stress, was examined in an experiment with 100 silver fox vixens. Fifty animals were assigned to the experimental group and kept individually for 2 years in 2-m 2 wire cages from which they had free access to three nest-boxes. The fifty other animals were assigned to the control group and kept without access to nest-boxes but under otherwise identical conditions. Experimental animals had lower base levels of cortisol and eosinophil leukocytes, were more active in an open field, and less fearful towards humans. Experimental animals were less fearful and more aggressive or exploratory no matter how and where behaviour was assessed, whether it was assessed in the cage, during capture or in an open field runway. It is concluded that adult silver fox vixens kept with access to nest-boxes all year round experience less stress than control animals kept without boxes.
Behavioural genetics and domestication J.M. Faure and A.D. Mills INRA, Station de Recherches Avicoles, Nouzilly, 37380 Monnaie, France
ABSTRACT Price defined domestication as "that process by which a population of animals becomes adapted to man and to the captive environment by some combination of genetic change occurring over generations and environmentally induced developmental events reoccurring during each generation". It is
181 implicit in Price's definition that domestication has to a large extent depended on selection for behaviour. This paper discusses behavioural changes which occur during domestication and the role of behavioural genetics in improving animal welfare. Four reasons exist for selecting for behaviour. Firstly, selection for production characteristics frequently occurs in an environment different from the one in which the animals' offspring are subsequently housed. Secondly, when farmers knew each animal individually, there was conscious selection for behavioural traits as animals with "bad characters" were not used for breeding. Today, animals are not known individually and this selection no longer occurs. Thirdly, in the last 40 years husbandry environments have changed too quickly for adaptation to occur through natural selection. Fourthly, selection for reproductive performance results in adaptation only when animals are placed in a very new environment and even then only for a few generations. Certain types of behaviour cause problems in virtually all husbandry systems. Other behaviour patterns are problems only in particular husbandry systems. Selection against behaviour which is problematic in a wide range of husbandry systems offers the most promise. However, efficient selection requires that thousands of animals are scored. Therefore, the behavioural measures used must be simple and cheap.
REFERENCES Price, E.O., 1984. Behavioural aspects o f a n i m a l domestication. Q. Rev. Biol., 59: 1-32.
Effect of genetic selection for dustbathing activity on productive traits in Japanese quail Martina Gerken and J. Petersen InstitutJ~r Tierzuchtwissenschafi, Abt. Kleintierzucht, Endenicher Allee 15, W-5300, Bonn 1, FRG
ABSTRACT As a consequence of increased public concern about animal welfare in intensive housing systems, genetic adaptation ofbehaviour to such systems might receive more attention in the future. Undesirable correlated changes in production traits, however, might interfere with such a selection. Divergent selection was conducted in male Japanese quail for low (N) and high ( H ) dustbathing activity, as measured by the number of dust tosses during one dustbathing bout, for 17 generations. Line C served as a randomly bred control. The following productive traits were individually recorded in all generations in both sexes: body weight at 12 weeks of age, age at first egg, and egg production between 12 and 23 weeks of age. Heritabilities were estimated from combined sire and dam variance components from the sib analyses. Across all generations, heritabilities for dustbathing activity were 0.28_+0.05 (N), 0.32_+0.05 (C) and 0.22_+0.04 (H), and for body weight were 0.50_+0.05 (N), 0.64_+ 0.05 (C) and 0.65 _+0.05 (H). In males, phenotypic and genetic correlations between body weight and dustbathing activity were not significantly different from zero. Observations on females in generations $7 (about 125 hens per line) revealed no significant phenotypic relationship between dustbathing behaviour and body weight, age at first egg and egg production. Across all generations,