The number of seeds produced by a flower is influenced by how much pollen a pollinator places on the stigma and by the likelihood that pollen fertilizes an ovule that develops into a seed. The relationship between pollen received and seeds produced should vary with changes in abiotic conditions, but those effects have rarely been studied. Such a prediction can be studied within the framework of dose response curves, which are characterized by a maximum response to a dose (maximum seed set) and a half-saturation parameter (dose of pollen received at which half of the maximum response is achieved). Here we asked how water availability affected fruit production, seed set, seed mass, and the dose response curve in Phacelia parryi. Recipient plants were exposed to either a low or a high water availability treatment, while all pollen donors received the high water treatment. In order to generate a wide spectrum of pollen received, hand pollinations were done in order to vary the pollen dose applied to stigmas on each plant. We did 708 hand-pollinations on 30 plants. For each hand-pollinated flower we counted pollen grains on the stigma and seeds produced and calculated seed mass.
Seeds per flower and seeds per fruit were higher in the high water availability treatment, but seed mass did not depend on the water treatment. Thus adding water increased seed production without a tradeoff in terms of lower seed quality as measured by mass. The dose response curve varied greatly even among individuals exposed to a given treatment. Despite that variation, we detected a significant difference between the dose-response curves of the two water treatments. The high water availability individuals had a higher value for the maximum seed production and reached half of its maximum seeds at a lower dose of pollen. Our data show how water availability influences not only average seed production but also the form of how pollen receipt translates into seed production, which could impact selection on floral traits. In our system, plants receiving more water make more seeds regardless of pollen receipt, but in other systems that need not necessarily be true as dose response curves can cross. We are now examining how water availability to donor plants influences pollen quality and the resultant seed production.