Notes Ecology, 96(2), 2015, pp. 587–593 Ó 2015 by the Ecological Society of America Pollen limitation and flower abortion in a wind-pollinated, masting tree 1,2,5 1,3 4 1 IAN S. PEARSE, WALTER D. KOENIG, KYLE A. FUNK, AND MARIO B. PESENDORFER 1Cornell Lab of Ornithology, 159 Sapsucker Woods Road, Ithaca, New York 14850 USA 2Illinois Natural History Survey, 1816 S. Oaks Street, Champaign, Illinois 61820 USA 3Department of Neurobiology and Behavior, Cornell University, Ithaca, New York 14853 USA 4School of Biological Sciences, University of Nebraska, Lincoln, Nebraska 68588 USA Abstract. Pollen limitation is a key assumption of theories that explain mast seeding, which is common among wind-pollinated and woody plants. In particular, the pollen coupling hypothesis and pollination Moran effect hypothesis assume pollen limitation as a factor that synchronizes seed crops across individuals. The existence of pollen limitation has not, however, been unambiguously demonstrated in wind-pollinated, masting trees. We conducted a two-year pollen supplementation experiment on a masting oak species, Quercus lobata. Supplemental pollen increased acorn set in one year but not in the other, supporting the importance of pollen coupling and pollination Moran effect models of mast seeding. We also tracked the fate of female flowers over five years and found that the vast majority of flowers were aborted for reasons unrelated to pollination, even in the presence of excess pollen. Pollen limitation can reduce annual seed set in a wind-pollinated tree, but factors other than pollen limitation cause the majority of flower abortion. Key words: anemophily; flower abortion; perennial plants; pollen; seed set; wind pollination. INTRODUCTION seed predators. The resource pulses generated by Pollen limitation, the production of less than the masting events can also have a cascading effect through food webs such that even those organisms that do not maximum seed potential due to inadequate pollen directly rely on mast seeds as a resource are still affected receipt by ovules, is common among animal-pollinated by the variability in seed set (Jones et al. 1998). plants but has received little attention in wind-pollinated Two often cited drivers of masting are the internal plants (Koenig and Ashley 2003, Ashman et al. 2004, resource dynamics of plants and pollination success Knight et al. 2006). Despite this, pollen limitation has (Isagi et al. 1997, Satake and Iwasa 2000, Rees et al. become central to the theoretical mechanisms regarding 2002, Koenig et al. 2014). Resource dynamics are central masting, the synchronous and intermittent production to many theories of masting (Isagi et al. 1997, Rees et al. of seeds. Theories to explain masting generally rely on 2002). Their importance is supported by numerous pollen limitation to explain the within-population studies that have reported mast years being followed synchrony of seed set (Satake and Iwasa 2000, 2002, by a year of particularly poor seed set (Sork et al. 1993, Koenig et al. 2014). In addition, pollen limititation is Pearse et al. 2014), a trend typically attributed to often considered a negative consequence of small resource limitation in nonmast years. Resource dynamics population sizes or fragmentation (Knapp et al. 2001, alone, however, cannot account for the high synchrony Wilcock and Neiland 2002, Jump and Pen˜ uelas 2006). in seed set among plants within a region, since each plant Surprisingly, however, pollen limitation has rarely been could presumably maintain its own unique schedule of experimentally demonstrated in natural populations of high and low seed production (Satake and Iwasa 2000). wind-pollinated plants (Koenig and Ashley 2003). Pollen limitation is one factor that could maintain Mast seeding is common among both insect- and synchrony in seed set. There are two key mechanistic wind-pollinated plants (Kelly 1994, Kelly and Sork hypotheses that link pollination success to masting 2002) and is particularly important to populations of (Satake and Iwasa 2000, Koenig et al. 2014). First, the pollen coupling hypothesis suggests that pollen could Manuscript received 12 February 2014, revised 11 June 2014, synchronize seed crops between individuals because accepted 18 July 2014. Corresponding Editor: T. M. Palmer. plants that flower in synchrony with conspecifics will 5 E-mail: [email protected] all produce a larger seed crop in that year. In turn, 587 588 NOTES Ecology, Vol. 96, No. 2 depleted resources from that masting event will limit height) and accessibility. Pollen supplementation exper- flower production for the entire population in the iments were conducted in 2012 and 2013. In each year, following year (Satake and Iwasa 2000). Second, the pollen-supplemented and control (open-pollinated) pollination Moran effect hypothesis suggests that branches were compared within trees. weather events in some years will limit pollination and Within-plant comparisons may lead to overestimation result in a poor seed set, and, in other years, resource of pollen limitation when pollen-supplemented flowers availability may limit seed set (Koenig et al. 2014). Both compete for resources with control flowers (Knight et al. of these hypotheses predict pollen limitation in masting 2006) We attempted to minimize this effect by choosing plants in some years and resource limitation in others. branches that were separated by at least 5 m of vascular We tested for pollen limitation in valley oak (Quercus tissue. Because trees are often modular in their lobata), a masting, wind-pollinated tree, using pollen carbohydrate use (Hoch 2005), these distant branches supplementation experiments in two consecutive years. were unlikely to compete for resources in the same way In addition, we followed the fate of female flowers over as neighboring flowers on smaller plants. five years and related this to the interannual variation in Prior to leaf set, we chose two branches of roughly seed crops. The pollen limitation hypothesis predicts equal size on each tree and randomly assigned one that pollen supplementation will enhance reproduction, branch to each treatment (control or pollen supplemen- but not necessarily in all years. Pollen limitation also tation). Each branch had 100–1000 leaves. We counted predicts that the total seed crop of a given year will be the number of female flowers and leaves present on each more closely related to the successful maturation of branch in late spring (May–June) and counted acorns on flowers into fruits than to the initial abundance of each branch the following September. From this, we flowers, so the size of a seed crop in a given year may be calculated the number of acorns and flowers per 100 unrelated to the abundance of female flowers in that leaves, as an estimate of acorn density (APL) and flower year. density (FPL) on each branch. As female flowers METHODS develop from leaf axils in oaks (Sharp and Sprague Study system and site 1967), standardization of flowers and acorns to the number of leaves accounts for differences in branch size We conducted our study at Hastings Reservation, and architecture. located 40 km southeast of Monterey in central coastal We added conspecific pollen to branches in slightly California. Pollination supplementation experiments different ways in 2012 and 2013. In 2012, we tied a were conducted on Haystack Hill (36.3851148 N, plastic mesh bag with 3-cm mesh size around pollination 121.561906 W) within the reserve. This site is a 8 and control branches. Each week, starting at the first �patchwork of native and nonnative grassland, oak week of flowering, we collected bulk Q. lobata catkins savanna, and chaparral (Griffin 1990). Three oak from trees outside of the experimental population. Each species, Quercus lobata, Q. douglasii, and Q. agrifolia pollen collection contained pollen from at least five are common; all are wind pollinated. Two of those individuals, but the identity of those individuals changed species (Q. lobata and Q. douglasii) can hybridize, but throughout the season as different individuals began and hybrids constitute a small percentage of trees in the area ceased flowering. Each week those bulk-collected catkins (Abraham et al. 2011). were divided evenly into 30 aliquots each containing We used valley oak (Q. lobata) to experimentally test for pollen limitation. As a masting species, Q. lobata ;20 catkins, and those catkins were placed on the pollen exhibits high interannual variation in seed set and high supplementation bagged branch. The control branch synchrony in seed set over large areas (Koenig et al. received no catkins. 1994, Koenig and Knops 2013). Interannual variation in In a subset of supplementation and control treatments (N 9 of each), we assayed pollen receipt during the acorn crop is correlated with spring temperatures ¼ (Koenig et al. 1996), suggesting that pollination, which preflowering period of the branches by taping a happens in spring, is involved in masting dynamics. microscope slide with a thin layer of vacuum grease to Likewise, trees that flower outside of the population pollen supplementation and control branches and then mean flowering time tend to produce fewer acorns than recording the pollen caught in the grease. Grease slides those that flower synchronously with other trees were left in trees for one week, during which time, the (Koenig et al. 2012). This leads to a trend where years ones on pollen supplementation branches received one with low synchrony in flowering have low acorn crops application of catkins. We also included a third (Koenig et al. 2014). treatment (pollen exclusion) on a third randomly chosen branch on each tree by placing a cotton pollen-exclusion Pollen supplementation experiments bag over the branch and taping the bag onto the main We chose 30 mature Q. lobata individuals within our stem of the branch in order to limit air and pollen flow study site based on size ( .25 cm diameter at breast into the bag. All bags (mesh bags and pollen exclusion February 2015 NOTES 589 bags) were removed in June, one month after trees had (pollen supplementation or control).