Get access

Nested distributions of bat flies (Diptera: Streblidae) on Neotropical bats: artifact and specificity in host-parasite studies


  • Bruce D. Patterson,

  • Carl W. Dick,

  • Katharina Dittmar

B. D. Patterson ( and C. W. Dick, Dept of Zoology, 1400 S. Lake Shore Dr., Field Museum of Natural History, Chicago, IL 60605, USA. – K. Dittmar, Dept of Biological Sciences, State Univ. of New York at Buffalo, 109 Cooke Hall, Buffalo, NY 14260, USA.


We examined the structure of ectoparasitic bat fly infestations on 31 well-sampled bat species, representing 4 Neotropical families. Sample sizes varied from 22 to 1057 bats per species, and bat species were infested by 4 to 27 bat fly species. Individual bats supported smaller infracommunities (the set of parasites co-occurring on an individual host), ranging from 1 to 5 fly species in size, and no bat species had more than 6 bat fly species characteristically associated with it (its primary fly species). Nestedness analyses used system temperature (BINMATNEST algorithm) because it is particularly well-suited for analysis of interaction networks, where parasite records may be nested among hosts and host individuals simultaneously nested among parasites. Most species exhibited very low system temperatures (mean 3.14°; range 0.14–12.28°). Simulations showed that nested structure for all 31 species was significantly stronger than simulated values under 2 of the 3 null hypotheses, and about half the species were also nested under the more stringent conditions of the third null hypothesis. Yet this structure disappears when analyses are restricted to “primary” associations of fly species (flies on their customary host species), which exclude records thought to be atypical, transient, or potential contaminants. Despite comprising a small fraction of total parasite records, such anomalies represent a considerable part of the statistical state-space, offering the illusion of significant ecological structure. Only well understood and well documented systems can make distinctions between primary and other occurrence records. Generally, nestedness appears best developed in host-parasite systems where infestations are long-term and accumulate over time. Dynamic, short-term infestations by highly mobile parasites like bat flies may appear to be nested, but such structure is better understood in terms of host specificity and accidental occurrences than in terms of prevalence, persistence, or hierarchical niche relations of the flies.