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

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

MRT = ([Sigma]/[m.sub.i][t.sub.i])/[Sigma] [m.sub.i]

where [m.sub.i] is the number of pollen grains excreted at the ith defecation at time [t.sub.i] after ingesting the pollen dose. In an ideal CSTR-PFR tract, MRT is equal to MGT plus the mean retention time in the stomach (Martinez del Rio et al. 1994). We plotted the natural logarithm of the number of pollen grains excreted against the length time since ingestion, and used the terminal portion of this curve to estimate the slope (k). The inverse of k estimates the retention time in the CSTR (Karasov and Cork 1996) and was used to determine whether or not the gastrointestinal tract of our bats fits a CSTR-PFR model. We also calculated the degree of longitudinal mixing in the reactor by estimating the vessel dispersion number (D/[[micro]liter]), using the formula

[[Sigma].sup.2]/[MRT.sup.2] = 2(D/[[micro]liter]) 8 [(D/[[micro]liter]).sup.2]

where

[[Sigma].sup.2] = [[Sigma] [([t.sub.i] - MRT).sup.2][m.sub.i]]/[Sigma] [m.sub.i].

No mixing is indicated by a D/[[micro]liter] value of zero, and the value approaches infinity when there is mixed flow. Values within this range indicate small (0.002), intermediate (0.025), or large amounts (0.200) of longitudinal mixing (Levenspiel 1972).

Because retention time in the gut increases with body [mass.sup.0.25] in birds and other vertebrates (Karasov 1990), we corrected for body mass by dividing MRT and MGT by body [mass.sup.0.25] to conduct interspecific comparisons.

RESULTS

Extraction efficiency and time of gut processing

Pseudobombax ellipticum pollen digestive processing. - The percentage of empty pollen grains differed significantly among the three species of bats ([ILLUSTRATION FOR FIGURE 2 OMITTED]; one-way ANOVA: F = 16.70, P = 0.0002, df = 2, 15). Anoura geoffroyi was the most efficient species at emptying pollen grains, followed by Artibeus jamaicensis and Sturnira lilium (Tukey's hsd test, P [less than] 0.05).

Mean gut retention times did not differ significantly among A. geoffroyi, A. jamaicensis, and S. lilium ([ILLUSTRATION FOR FIGURE 3 OMITTED]; one-way ANOVA: F = 2.54, P = 0.118, df = 2, 15), and minimum gut transit times were significantly different only between A. jamaicensis and S. lilium ([ILLUSTRATION FOR FIGURE 3 OMITTED]; log-transformed data, one-way ANOVA: F = 8.14, P = 0.004, df = 2, 15, Tukey's hsd test, P [less than] 0.05). Longer MGTs were detected in A. jamaicensis, followed by A. geoffroyi and S. lilium. No significant differences in MRT and MGT were found among the three species of bats on a mass-specific basis ([ILLUSTRATION FOR FIGURE 3 OMITTED]; one-way ANOVA, MRT: F = 0.09, P = 0.917, df 2, 15; MGT: F = 0.49, P = 0.623, df = 2,

15).

Although neither MRT nor MGT was significantly related to the mean percentage of empty grains in any of the species (MRT: A. geoffroyi, r = 0.320, A. jamaicensis, r = 0.392 S. lilium, r = 0.447; MGT: A. geoffroyi, r = 0.453, A. jamaicensis, r = 0.059, S. lilium, r = 0.320; all P [greater than] 0.112, df = 1,4), in general, the percentage of empty grains increased asymptotically with time spent in the gut in individuals of the three species [ILLUSTRATION FOR FIGURE 4 OMITTED].