Pollen digestion by new world bats: effects of processing time and feeding habits

Ecology, Dec, 1998 by L. Gerardo Herrera, Carlos Martinez del Rio

Comparative processing efficiency of the three species of pollen by A. jamaicensis and A. geoffroyi. - In general, the three types of pollen were processed with the same efficiency and at the same rates, with some exceptions. Although A. jamaicensis emptied the same percentage of grains in the three pollen species (one-way ANOVA, F = 3.49, P = 0.057, df = 2, 15), A. geoffroyi emptied fewer grains of H. undatus pollen than of the other two pollen types (one-way ANOVA, F = 10.13, P [less than] 0.0001, df = 2, 20; pairwise T[prime] test for equal variances and unequal sample sizes: H. undatus-P. ellipticum, P [less than] 0.0001; H. undatus-columnar cactus, P = 0.010; P. ellipticum-columnar cactus, P [greater than] 0.100). There were no significant differences among the MRT of the three types of pollen in A. geoffroyi (one-way ANOVA, F = 0.74, P = 0.489, df = 2, 20) and A. jamaicensis (one-way ANOVA, F = 3, P = 0.08, df = 2, 15). However, A. jamaicensis had significantly longer MRT for H. undatus than for P. ellipticum grains when the values were controlled for body [mass.sup.0.25] (one-way ANOVA, F = 10.45, P = 0.001, df = 2, 15; T[prime] test: H. undatus-P. ellipticum, P [less than] 0.010; H. undatus-columnar cactus, P [less than] 0.100; P. ellipticum-columnar cactus; P [greater than] 0.100). Mean retention times controlled for body [mass.sup.0.25] were not significantly different among the three types of pollen in A. geoffroyi (one-way ANOVA, F = 0.68, P = 0.517, df = 2, 19). Mean gut transit times were not significantly different among pollen types in A. geoffroyi even after controlling for body mass (one-way ANOVA; MGT: F = 1.71, P = 0.208, df = 2, 19; MGT/[mass.sup.0.25]: F = 0.88, P = 0.423, df = 2, 19), but H. undatus grains had significantly shorter MGTs than those of P. ellipticum and the columnar cactus in A. jamaicensis (one-way ANOVA; MGT: F = 4.25, P = 0.034, df = 2, 15; T[prime] test: H. undatus-P. ellipticum P [less than] 0.05; H. undatus-columnar cactus, P [less than] 0.05; P. ellipticum-columnar cactus, P [greater than] 0.30; MGT/[mass.sup.0.25]: H. undatus-P. ellipticum, P [less than] 0.05; H. undatus-columnar cactus, P [less than] 0.05; P. ellipticum-columnar cactus, P [greater than] 0.20).

Excretion patterns. - In general, individual F curves resembled qualitatively the excretion curves expected for a CSTR-PFR reactor [ILLUSTRATION FOR FIGURE 6 OMITTED]. The slope of the plot of the log-transformed number of grains excreted (k, where the slope has been multiplied by -1 to make k a positive number) tended to be steepest in A. geoffroyi, followed by S. lilium, L. curasoae, and A. jamaicensis ([ILLUSTRATION FOR FIGURE 7 OMITTED]; Table 1), and the average amount of time pollen grains spent in the stomach (1/k) was 15-75% of MGT, depending upon the species and pollen type (Table 1). In most cases, the sum of MGT and retention time in the stomach was only 68-99% of the MRT (Table 1). The mean amount of longitudinal mixing varied with type of pollen and species of bat; A. jamaicensis had intermediate to nearly large amounts of mixing, S. lilium and L. curasoae had intermediate amounts of mixing, and A. geoffroyi had small to intermediate amounts of mixing (Table 2).