168 Test of a Digestion Optimization Model: Effects of Costs of Feeding on Digestive Parameters Scott R. McWilliams* [1996]). For example, yellow-rumped warblers (Dendroica co- ronata) that are switched quickly from a low- to a high-protein William H. Karasov Department of Wildlife Ecology, University of Wisconsin — diet increase enzyme (aminopeptidase) activity, amino acid uptake, and extraction efﬁciency of an amino acid (Aﬁk and Madison, Madison, Wisconsin 53706 Karasov 1995; Aﬁk et al. 1995). These changes are not, however, associated with any changes in gut morphology (Aﬁk and Kara- Accepted by G.K.S. 9/12/97 sov 1995), although diet quality has affected gut morphology in other situations (see, e.g., Piersma et al. 1993). Effects of diet quantity on avian digestive parameters have ABSTRACT not been studied often, but the available evidence indicates that increased food intake primarily causes changes in gut We tested predictions of a chemical reactor model of digestion by manipulating the short-term costs of feeding and then mea- morphology that allow digesta retention time and extraction efﬁciency to remain constant (Karasov 1996). However, such suring the effect on digestive parameters. We compared resi- dence time of digesta and extraction efﬁciency of glucose in modulation at the level of gut morphology requires time (per- haps as long as 2–3 mo; Redig 1989). Birds may often experi- cold-acclimated waxwings (Bombycilla cedrorum) feeding ad lib. and in birds whose costs of feeding were increased through ence short-term changes in food quantity that occur faster than the time scale required for changes in gut morphology. For the addition of intervals of time when they received no food. Such a feeding schedule simulated the ecological situation in example, frugivorous birds during migration may one day en- counter preferred fruits that are ubiquitous, allowing relatively which a frugivorous bird like a waxwing encounters food in patches and experiences nonfeeding periods as it searches for constant food intake, whereas the next day their preferred fruits may be patchily distributed and require much travel time be- new preferred food patches. None of the results were consistent with the predictions of the optimal digestion model: extraction tween patches. In such situations, a bird’s pattern of food intake may differ from day to day. efﬁciency was independent of costs of feeding, and residence times did not increase as costs of feeding increased. This empir- We know little about how short-term changes in food intake affect digestive performance in wild birds. Theoretical ical evidence on the passage of digesta in waxwings suggests that movement of digesta in the guts of birds is much more optimality models make explicit predictions about how an animal’s digestive parameters should respond to short-term complex than movement of material in an ideal chemical reac- tor. Tests of the optimal digestion model have involved manip- changes in food intake (Penry and Jumars 1986, 1987; Martı B - nez del Rio and Karasov 1990; Martı B nez del Rio et al. 1994), ulating food quality or the costs of feeding, and the conclusions are similar: compensatory modulation of retention time or although our study provides the ﬁrst such empirical test of the models. The models predict that when costs of food digesta mixing and not rate of hydrolysis and absorption seem most important in maintaining the remarkably constant diges- acquisition are increased, for example, by adding intervals of time when the birds receive no food, the food should be tive efﬁciency. held longer in the intestine, and thus nutrients in the food will be more thoroughly digested (Fig. 1). It is important to note that if retention time and extraction efﬁciency are Introduction modulated as predicted by the model, determining the Digestive parameters at many organizational levels are inﬂu- proﬁtability (e.g., energetic gain divided by the energetic enced by diet quality and quantity (reviewed by Karasov costs) of a given food type is complicated, because digestive efﬁciency is not ﬁxed but is instead conditional on the costs of acquiring the food. *To whom all correspondence should be addressed. Present address: Depart- We tested predictions of the model by manipulating the ment of Natural Resources Sciences, 237 Woodward Hall, 9 East Alumni Ave- short-term costs of feeding and then measuring the effect on nue, University of Rhode Island, Kingston, Rhode Island 02881; E-mail: srmcwill@uriacc.uri.edu. retention time and digestive efﬁciency. For the test, we used cedar waxwings (Bombycilla cedrorum) fed a semisynthetic diet Physiological Zoology 71(2):168–178. 1998.  1998 by The University of Chicago. All rights reserved. 0031-935X/98/7102-96116$03.00 rich in glucose. Cedar waxwings are ideal candidates for testing 9g12$$mr08 03-02-98 14:05:29 pza UC: PHYS ZOO