Plant responses induced by herbivore damage can provide fitness benefits, but can also have important costs due to altered interactions with mutualist pollinators. We examined the effects of plant responses to herbivory in a hummingbird-pollinated distylous shrub, Palicourea angustifolia. Through a series of field experiments we investigated whether damage from foliar herbivores leads to a reduction in fruit set, influences floral visitation, or alters floral traits that may influence pollinator preference or pollinator efficiency. Foliar herbivory by a generalist grasshopper led to reduced fruit set in branches that were directly damaged as well as in adjacent undamaged branches on the same plant. Furthermore, herbivory resulted in reduced floral visitation from two common hummingbird species and two bee species. An investigation into the potential mechanisms behind reduced floral visitation in induced plants showed that foliar herbivore damage resulted in shorter styles and lower nectar volumes. This reduction in style length could reduce pollen deposition between different floral morphs that is required for optimal pollination in a distylous plant. We did not detect any differences in the volatile blends released by damaged and undamaged branches, suggesting that foliar herbivore-induced changes in floral morphology and rewards, and not volatile blends, are the primary mechanism mediating changes in visitation. Our results provide novel mechanisms for how plant responses induced by foliar herbivores can lead to ecological costs.
fruitset
To determine the effects of herbivory on fruit set, we experimentally manipulated damage on 39 pairs of plants. On one plant in each pair we selected and bagged two branches with inflorescences at the bud stage that were at approximately the same size, phenological stage, and height on the plant. One of the branches received herbivore damage where one individual of Z. selenesii was placed inside the bag for three days to consume the leaves. In order to prevent the grasshopper from eating the inflorescence, the inflorescence was covered with a second bag during the three days the herbivore was feeding. The damaged branch was subsequently used to measure locally-induced responses to herbivory. The second branch of the same shrub was bagged in the same manner but received no damage and was used to measure systemically-induced responses to herbivory. The second plant in each pair served as an absolute control in which a single branch that was approximately the same size and height was bagged in the same way without placing any herbivore on the plant. After the three-day herbivore treatments, we removed the grasshoppers and the bags and allowed natural pollinator visitation to the inflorescences. We counted the total number of flowers that were aborted and the number that successfully set fruit. The first column (replica) show the number of the replica. The second column shows the treatment of the branch. The third column shows the number of fruit that was developed in the branch. the fourth column shows the number of flowers in the branch. the fifth column shows the fruit set (proportion of flowers that becomes in fruits).
alex_fruitset.csv
morphology
To establish if foliar herbivory causes changes in floral morphology and rewards that could alter floral visitation, we selected another 30 pairs of plants (15 pin and 15 thrum) of P. angustifolia. On each pair, we selected branches with inflorescences at the bud stage to serve as locally-induced, systemically-induced, and control branches and applied grasshopper herbivory treatments as described above. After 3 days, grasshoppers and bags were removed and once the flowers opened we took measurements of corolla length, anther length, style length, and distance from anthers to stigma for one flower randomly chosen from each inflorescence. The first column shows the number of the group of plants (bloque), the second column shows the treatment: L, local induction with Z. selenesii; S systemic induction and C, control. The third column shows the Morph of the branch T for Thrum and P for Pin. columns 4 to 7 show the length of different structures in the flower (Corola, anteras (anthers), estilo (Style) and distancia anteras-estigma(distance anthers-stigma)).
Alex_morphology.csv
nectar
To establish if foliar herbivory causes changes in floral rewards that could alter floral visitation, we selected another 30 pairs of plants (15 pin and 15 thrum) of P. angustifolia. On each pair, we selected branches with inflorescences at the bud stage to serve as locally-induced, systemically-induced, and control branches and applied grasshopper herbivory treatments as described above. After 3 days, grasshoppers and bags were removed. From a subset of the same inflorescences (5 pin and 6 thrum), we also collected nectar on one randomly chosen flower. Nectar volume was measured using 2µL microcapillary pipettes (Drummond Scientific Company, Broomall, PA). The sugar concentration was estimated using a handheld refractometer (Reichert Digital Brix/RI-Chek). Due to low nectar volumes, all samples were diluted in 0.1 mL distilled water prior to taking concentration measurements. The concentration of the floral nectar was then calculated as the measured solution concentration multiplied by the ratio of the final solution volume to the collected nectar volume. The first column shows the flower morph: pin (P) and thrum (T). The second column is the volume in microliters of nectar present in the flower. The third column is the concentration of the nectar in each flower. The fourth column is the treatment of each branch: C control, S systemic induction and L local induction. The fifth column is the block of each pair of plants.
Alex_nectar.csv
totalvisits_summed
To determine if foliar herbivory affected pollinator visitation, we monitored pollinator visits to 26 pairs of plants from the experiment described above. Plant pairs used for the observations were always located within approximately 2 m of each other to allow simultaneous pollinator observations. On each pair of plants, we recorded floral visitors in a series of one to three separate 30-minute continuous observations conducted after the removal of grasshoppers from the leaves. Just prior to each observation, we counted the number of open flowers on inflorescences from each of the three branches. The locally-induced, systemically induced, and control branches from a single plant pair were then observed simultaneously to record the identity of the floral visitors and the number of flowers visited by each individual visitor. The first column shows the name of the pollinator. the second column is the number for each couple of plants. the third column is the treatment of each branch: L local herbivory, S systemic herbivory, and C control. The fourth column is the number of visits of each pollinator. the fifth column is the number of individuals that made that visit and the sixth column is the average of opened flowers at the moment of the observation.
alex_visitsbyID.csv