Behavioral and color variations between rat lines developed for differential alcohol sensitivity.

The AT and ANT rat lines, outbred for differential sensitivity to ethanol-induced motor impairment, also show a difference in their sober behavior. It is not manifested in an escapable-shock test or an amphetamine-stress test, but is shown as significantly more activity by the alcohol-insensitive ATs in a low-stress (33 degrees C) modification of the forced-swimming test. The correlation between alcohol sensitivity and swimming-test activity is, however, not significant in unselected Long Evans rats. Differences in coat color have also developed in the AT and ANT lines; it was possible to estimate whether these changes are independent of ethanol sensitivity with computer simulations. They showed, for example, that the probability of the observed loss of the agouti color in the ATs being by chance is about 0.53 and the probability is nearly 0.50 for there being a spurious line difference, i.e., unrelated to ethanol sensitivity, in which only line had lost the agouti allele. More generally, these and other simulations showed that permanently maintaining selected lines is not the optimal method for finding genetically-based factors related to ethanol sensitivity or other characteristics for which lines have been developed. Nor is either revitalization or replicate lines optimal. The best method apparently would be perpetual restarting of lines.