During May, 2001, we investigated whether there was an association between the abundance of rotifers occupying lobules of the epiphytic liverwort Frullania eboracensis and sex expression of the plants. At our field site in southwestern New York, isolated patches of F. eboracensis (referred to as "plants" and assumed to be unique genets) were marked on nine different trees (N=81 plants). Portions of the plants were removed, and examined microscopically in the laboratory. Presence of sexual structures (antheridial branches and sporophytes) was noted, and the numbers of rotifers and lobules were counted. Plants were also measured for size. Plants for whom no sexual structures were found were sampled at least twice more in June to confirm lack of sex-expression. Of the 81 plants, there were 44 females, 17 males, and 20 of unknown sex. Using an analysis of variance followed by Tukey multiple comparison tests, we found that unknown plants were significantly smaller than sex-expressing plants and that the ratio of rotifers to lobules was significantly higher in plants of unknown sex than in either male or female plants. Plants of unknown sex had a mean ratio ( 1 standard error) of 0.83 (0.15) rotifers:lobules. Male and female plants had a mean ratio of 0.38 ( 0.04) rotifers:lobules. Within sexes, there was no significant correlation between rotifer abundance and size nor between rotifer abundance and the number of sexual structures observed on the samples. Our results provide evidence that this plant-animal interaction has ecological significance to the taxa involved.

Key words: Bdelloid rotifer, bryophyte, Frullania eboracensis, liverwort, plant-animal interaction, sex ratio