TEST OF A REACTOR-BASED DIGESTION OPTIMIZATION MODEL FOR NECTAR-EATING RAINBOW LORIKEETS

We tested predictions of a chemical reactor model of digestion of gluc ose solutions by rainbow lorikeets (Trichoglossus haematodus, 125 g). The optimization criterion was the maximization of net energy gain. Th e predictions were that residence time in the small intestine would be inversely related to sugar concentration, and extraction efficiency w ould be inversely related to sugar concentration and generally below 9 0%. Residence time in the small intestine was indexed by the transit t ime (TT) of the nonabsorbed marker polyethylene glycol (molecular weig ht = 4,000), and extraction efficiency was measured by the inert indic ator ratio technique. Birds were tested when ingesting 0.4 or 1.2 M D- glucose. All predictions were rejected: TT teas independent of concent ration (average = 37 +/- 4 min, n = 7 birds), and extraction efficienc y was uniformly very high (98.0% +/- 0.4%) and also not influenced by glucose concentration. At low and high sugar concentration the rates o f glucose absorption were similar (average = 4 mmol/h). Extraction eff iciency of sucrose solution at one concentration, 1.2 M, was 90.5% +/- 1.0% (n = 6). Stomach-emptying rate appeared slower at high glucose c oncentration, perhaps owing to negative feedback arising from intestin al receptors. Total (mouth-to-anus) mean retention time was independen t of glucose concentration and averaged 88 +/- 12 min. Because the bir ds were not growing, storing fat, or reproducing, perhaps the assumpti on about energy maximization was inappropriate. The birds' responses w ere consistent with the goal of minimizing feeding time.