Botanists have known since the late 19th century that plants that self-fertilize are more likely to have an annual life cycle, while perennials are more likely to be out-crossers. This may be because poor conditions during the reproductive period will have greater consequences for annual plants than perennials if only outcrossing is possible. Self-pollination relieves this pressure, and indeed, many annuals occupy unpredictable and marginal environments. Conversely, perennials often occur in very competitive environments where high quality offspring are a higher priority.
The occurrence of both annual and perennial populations within the same species is rare enough that only one such species has been examined for its rates of selfing versus outcrossing: a wild rice in which the annual plants were primarily self-fertilized, while the perennials had a mixed mating system.
In a new article published in Annals of Botany, lead author Yue Ma and colleagues studied 21 populations of Incarvillea sinensis, a species made up of both annual and perennial individuals. Of the 21 populations, 16 contained only one life cycle or the other, while five were a mix of both. For each population, the researchers took measurements related to size and floral characteristics, as well as using microsatellite loci to estimate outcrossing rates. Within the mixed populations, they studied pollinator visitation rates and fruit and seed set.
In accordance with theory, the perennial populations outcrossed at a significantly higher rate than did the annuals, a difference that was observed in mixed populations as well. Perennial plants produced more flowers that were larger than those on the annual plants, indicating greater resource allocation to attracting pollinators. In mixed populations, annual plants had a higher fruit and seed set than perennials.
One of the floral measurements taken, that of the separation between stigma and anther (herkogamy), showed clear differences between annual and perennial plants, with a smaller separation found in annuals. This strongly suggests that differences in mating systems have a genetic basis.
“[W]hat is significant about our findings is that the mating system and life history association is maintained over a broad geographical range encompassing striking environmental variation. We were unable to find a single case in which either annual populations were even moderately outcrossing or perennial populations had high selfing rates,” write the authors.