Can extinction risk help explain plant–pollinator specificity among euglossine bee pollinated plants?


J. D. Ackerman, Dept of Biology, Faculty of Natural Sciences, Univ. of Puerto Rico, PO Box 70377, San Juan, PR 00936-8377, USA. E-mail:


Long-term or lifetime specificity in plant–pollinator relationships is likely a consequence of natural selection to not only enhance the probability of cross-pollination but also to improve pollinator efficiency. Dependency on one or few pollinators involves risk whereas multiple species may reduce the probability of extinction via unreliable pollinator service. We analyzed specificity in terms of factors that may ameliorate risk such as long-term pollinator population stability, abundance and the duration of flowering. Bee population stability indices from seven continuous years of census data, combined with pollinator and flowering phenology data for 37 plant species in Panama, revealed pollinator specificity was not related to pollinator population stability. No relationship existed between the length of a flowering season and population stability of associated pollinators. Further data from 30 years of euglossine monitoring also revealed no relationship between bee abundance and specificity. However, a strong relationship was revealed between length of flowering period and specificity. A longer flowering season was associated with lower specificity and shorter flowering was associated with higher specificity, which is as expected if specificity is the outcome of a sampling problem but not as expected if specificity is accompanied by risk reduction. Plant–pollinator specificity involving euglossine bees is evidently not related to bee population stability, abundance, or length of flowering period, in a manner that we predicted would be associated with reducing the risk of extinction. Variation in population stabilities of euglossines may be insufficient to be a factor in the evolution of plant–pollinator specificity. In the tropics, specificity may be more associated with plant longevity, selection for efficiency or effectiveness, or flowering duration –as a sampling phenomenon, than with reducing dependence on unreliable pollinators.