Fugitive Species: Experiments with Fungi and Some Theoretical Considerations

An experimental ecosystem was constructed and used in studies of species co- existence in a patchy environment. This model system consisted of two species of fungi, Aspergillus nidulans and an unidentified species of Penicillium, competing for space on agar- filled petri dishes. Aspergillus was observed to overrun and infiltrate established colonies of Penicillium, and so was considered competitively dominant, while Penicillium produced a greater number of daughter colonies at an earlier age than did Aspergillus, and so was con- sidered a better colonizer. It was hypothesized that these two species should be able to coexist in a patchy environment where local populations on occasion become extinct, with Penicillium playing the role of a fugitive species. Two different regimes (ANNUAL and PERENNIAL) for the generation of empty patches through the extinction of their resident species were compared with reference to the ease with which coexistence was obtained. Under the ANNUAL regime, where all patches became empty (hence available for colonization) at the same time and where the (n + 1)st generation was seeded with spores from the nth generation alone, coexistence was not obtained. In the PERENNIAL experiments, however, where only a fraction of patches became empty at any one time and where newly emptied patches were seeded from patches of several different ages, coexistence was easily obtained. Temporal differences in the use of single patches occurred under the PERENNIAL transfer regime, with Penicillium colonizing from younger patches in which Aspergillus had not yet exerted its dominance, and Aspergillus colonizing from older patches. Temporal partitioning did not occur in the ANNUAL experiments, and coexistence was not obtained. Since the pattern of disturbance and the life history patterns of the competing species combine to determine the extent to which temporal partitioning can occur, it is evident that for the fugitive species mechanism to be effective, the life history patterns of the species involved and the pattern of environmental disturbance must be well matched. Discussion of these problems is facilitated by the introduction of the notion of "coexistence bandwidth," a theoretical construct which relates in these experiments to the ease with which coexistence can be obtained, and in natural systems to the fragility or robustness of competitive associations in the face of changing environmental conditions.