Print PageThe desert shrub Krameria erecta (littleleaf ratany) belongs to a family of plants associated with a specialized lineage of pollinators, floral oil-collecting bees in the tribe Centridini (Apidae: Apinae). While Krameria species are generally thought to be dependent on Centrid bees for successful pollination, there have been no in-depth studies of which insect species visit and effectively pollinate K. erecta. The breeding system of K. erecta has also not yet been measured in the field.
At our study site in central Arizona, we compared K. erecta fruit set among three pollination treatments: assisted autogamy, geitonogamy, and within-population outcrossing. Each treatment was applied to a single branch on each experimental plant, and all of the flowers on a branch received the same treatment. Branches were bagged to exclude visitors before and after treatment. One branch on each plant was bagged but otherwise unmanipulated, and an additional branch was designated an open-pollinated control. In a separate experiment, we compared pollen tube growth between self and outcross pollinations. Styles were collected 2, 6, 10, and 24 hours after pollen deposition. Pollen tubes were stained with 0.1% aniline blue and examined under UV light.
Results/Conclusions
As expected, K. erecta flowers at our study site were visited by female Centris sp (Centris rhodopus) collecting floral oil. However, flowers were also frequently visited by female Lasioglossum subgenus Dialictus collecting pollen. Scopal loads of Dialictus (n=9) collected on K. erecta flowers indicated some fidelity to Krameria, at least within a single foraging trip. Foraging Dialictus clearly moved pollen from anthers to stigmas within K. erecta flowers, but their potential to vector pollen on a larger scale is unclear.
Both self and outcross pollination treatments initiated fruit set. Quantitative comparisons among self and outcrossed treatments were not possible because of widespread loss of flowers and fruit to herbivory by microlepidopteran larvae. Pollen tube growth was rapid, with pollen tubes reaching the base of the style within 2 to 6 hours. All pollination treatments produced pollen tubes that reached the style base. It appears that, if K. erecta has some degree of self-incompatibility, the mechanism is either late-acting or occurs through variable pollen tube growth rates within the first two hours after pollen deposition.
Our preliminary results suggest that the pollination ecology of K. erecta may be more complex than originally expected, with the potential for successful pollination by insects other than the specialized Centrid bees.
