are not allowed to become obese (Ator, 1991; Kemnitz et al., 1989, 1993; Lane et al., 1992, 
1997; Turturro et al., 1999). For example, rats having dietary restriction sufficient to cause a 
25 percent reduction in body weight compared to controls fed ad libitum lived longer without 
impairment of growth or of routine clinical indices of health (Hubert et al., 2000). Weight 
restriction is best started after the animal has reached maturity. Problems occur only if the 
ration is nutritionally incomplete or unbalanced. 
Restriction of caloric intake, in the context of ensuring a nutritionally balanced diet, is 
recognized in the 1996 ILAR Report (ILAR, 1996) as an accepted practice in long-term housing 
of some species. In the wild, food and water generally are not "freely" available; that is, effort 
(foraging) is required to obtain them. Ethological observations indicate that most species 
have access to food and water only for limited periods of each day (Altman and Altman, 1970; 
Hall, 1965; Hamilton et al., 1976; Lindburg, 1977). Thus, research methods that require 
animals to expend time and energy to obtain food during limited periods each day can be 
compatible with the natural pattern. In fact, USDA regulations permit "task-oriented" access 
to the regular food supply as a means of environmental enrichment for laboratory primates. 
‘TREATS’ VERSUS BALANCED DIET AS FOOD REWARDS 
Although "preferred" food items or "treats" often are used to maintain stable responding, 
balanced pelleted or liquid diets have several advantages over treats, such as sugar pellets or 
sweetened condensed milk. It is important to note that the nutritional status of the animal 
may be better if the majority of calories are obtained from balanced diet rather than treats (i.e., 
even if balanced diet is freely available, animals may eat less of it if they receive a significant 
number of calories from treats). The possibility of dental caries with frequent consumption of 
sugared food is also a disadvantage, particularly when the subjects will serve for many 
months or years. 
SPECIES DIFFERENCES IN WEIGHT REGULATION 
The manner in which food restriction is accomplished and any target weight selected-must be 
carefully considered for the species in question to maintain the animals in good health and to 
adhere to humane standards of care. The reduced weight often seen as a "generic" standard in 
the literature for a variety of species is 80 to 85 percent of a free-feeding weight. The age of 
the subject and the duration that free feeding is permitted, however, are critical determinants 
of whether the "80 percent" rule is a reasonable one for different species. Knowledge of 
nutrient requirements as well as feeding and growth patterns for different species is important 
to determine rational weight control regimens. The goal is to select a weight range that 
permits the reinforcer to maintain responding during the experimental session and maintains 
the animal's physical well-being. Another factor to consider is that a lower weight may be 
necessary early in training but not after performance has been established, even though food 
44 
