Diets of foliage-gleaning insectivorous bats in Mexican shade coffee plantations

Background/Question/Methods

Both observational and experimental studies suggest that in the Neotropics, foliage-gleaning insectivorous bats (Subfamily Phyllostominae) are important predators of herbivorous insects.  In a previous exclosure study, we found that excluding bats from coffee plants in a shade coffee plantation in Mexico resulted in an 84% increase in arthropod density when bats were excluded, an effect we attributed to foliage-gleaning bats.  However, information on the diets of these bats is limited.  To better understand the diets of foliage gleaning bats, we used molecular methods to determine insect species.  We captured foliage-gleaning bats in shade coffee plantations and collected fecal samples, which we dried, ground, and from which we extracted DNA.  We then used a universal arthropod primer to amplify a 160-bp segment of the mitochondrial COI gene (the DNA “barcoding” region in animals), producing a “mini-barcode.”  PCR product was cloned into chemically-competent E. colito separate templates, reamplified via colony PCR, sequenced, and then matched to voucher sequences in the Barcode of Life Database.  We considered sequences with 99.3% or greater similarity as being species-level matches, and sequences with a less than 99.3% match to one another were considered to be different Operational Taxonomic Units (OTU's).

Results/Conclusions

At the time of writing, we have recovered 301 arthropod sequences from 22 bat fecal samples representing three species of insectivorous phyllostomids (4 Lonchorhina aurita, 4 Trachops cirrhosus, and 14 Micronycteris microtis).  According to the criteria outlined above, the 301 sequences represent 69 OTU's.  We recovered an average of 3.7±0.5 SE OTU's per sample, with as many as 9 different OTU's recovered from a single bat.  There were only three samples with overlap between bats in OTUs consumed.  We were able to identify over 40% of  OTU's to the genus or species level.  Not surprisingly, many (30%) of these sequences belonged to lepidopteran species, which accords with other studies of bat diet.  However, 33% of OTU's belonged to dipterans, a surprisingly high number given that these species are not known to prey heavily on flies.  In some cases, it is probable that bats ate larvae infected by tachinid parasitoids.  However, approximately half of the dipteran OTU's belong to non-parasitoid families, suggesting that these bats prey more heavily that anticipated on flies.   At the time of writing, we have found no significant effect of bat species, season, or habitat type on dietary diversity or composition.