Quality matters. A commentary on: Pollen limitation and xenia effects in a cultivated mass-flowering tree, Macadamia integrifolia (Proteaceae)

Abstract

This article comments on:

Stephen J. Trueman, Wiebke Kämper, Joel Nichols, Steven M. Ogbourne, David Hawkes, Trent Peters, Shahla Hosseini Bai and Helen M. Wallace, Pollen limitation and xenia effects in a cultivated mass-flowering tree, Macadamia integrifolia (Proteaceae), Annals of Botany, Volume 129, Issue 2, 1 February 2022, Pages 135–146 https://doi.org/10.1093/aob/mcab112

Despite the importance and prevalence of pollen limitation, its adequate measurement is often difficult; the study by Trueman et al. (2021) addresses two of the main issues: reallocation of resources and pollen quality. Pollen limitation, or the reduction in female fitness due to low quantity and/or quality of pollen received, can affect a variety of processes including population demographics and growth, selection on reproductive traits and, in the case of agriculture, crop yield. Not only is pollen limitation important, but it also is common; meta-analyses have shown that 63 % of plant species studied are pollen limited (Knight et al., 2005), with supplemental pollen addition increasing fruit set by an average of 63 % (Bennett et al., 2020). Unfortunately, accurately measuring pollen limitation can be difficult. Pollen limitation is generally measured by comparing the fruit/seed production from naturally pollinated flowers with that of flowers that have been supplemented with outcrossed pollen (Fig. 1). If the supplemental treatment has higher fruit/seed production than the naturally pollinated flowers, the plant is said to be pollen limited. While this sounds simple in theory, in practice there are two main issues: reallocation of resources may lead to an overestimate of pollen limitation and low pollen quality may lead to an underestimate of pollen limitation. Trueman et al. (2021) have addressed both of these issues in the economically important macadamia (Macadamia integrifolia, M. tetraphylla and hybrids).

Fig. 1.

To measure pollen limitation, the fruit/seed production of flowers that have been supplemented with outcross pollen (shown on the right) is compared with production under natural pollination (shown on the left). This method confounds pollen quantity and quality since even as the total amount of pollen increases after supplementation, the proportion of high quality outcross pollen (green) to low quality self-pollen (orange) increases as well.

Macadamia is a sub-tropical tree species that originated in Australia. In masting years, a single tree can produce 200 000–400 000 bee-pollinated, partially self-incompatible flowers that are each capable of producing a one-seeded fruit. Although not grown agriculturally until the mid-19th century, the seeds were probably used as a food source by the indigenous people of Australia due to their high oil content (Nock et al., 2019). Because of its recent domestication and the clonal propagation of cultivars, cultivated trees differ little from wild populations (although there is evidence of a genetic bottleneck; Nock et al., 2019), making it useful for better understanding the evolution and ecology of masting and pollen limitation.

The first major difficulty with measuring pollen limitation, especially in large, long-lived species such as macadamia, is that plants may reallocate resources between flowers or years, exaggerating the appearance of pollen limitation (Zimmerman and Pyke, 1988). If only a few flowers on a plant are supplemented with pollen, the plant may reallocate resources to developing those fruits, increasing the difference between supplemental and naturally pollinated flowers, and exaggerating the amount of pollen limitation. Thus, for long-lived species, accurate measures of pollen limitation require a large portion of the flowers to be supplemented with pollen over several years. While Trueman et al. (2021) did not apply pollination treatments across years, they were able to use a relatively quick technique to add supplemental outcross (between cultivar) pollen to approx. 70 % of the flowers on a plant, offering a rare insight on pollen limitation in a tree species (Knight et al., 2006).

The second difficulty with measuring pollen limitation is that pollen might be limited in both quantity and quality, and it can be difficult to distinguish between the two causes (Aizen and Harder, 2007). The quality of pollen received may limit female fitness in a variety of ways including: heterospecific pollen clogging the stigma, self-pollination (especially for self-incompatible species or those with high inbreeding depression) and even maternal choice. Unfortunately, because most pollen limitation studies use outcrossed pollen for the supplemental treatment, the effect of quantity and quality are confounded (Fig. 1). Trueman et al. (2021) addressed the issue of quality in two ways. First, they used experimental trees in areas that differed in their likely pollen quality. Second, they used paternity analysis to directly examine how the identity of the pollen parent influenced the quality of the seeds produced.

To test for pollen limitation in both quantity and quality of pollen, Trueman et al. (2021) compared control and supplemental plants in different parts of a field – plants in the middle of the same cultivar (where most pollen is likely to be from the same cultivar and therefore low quality) and plants on the border between two cultivars (where more pollen is likely to be from the other cultivar and therefore higher quality). This allowed an estimation of the importance of the quality of pollen received. To mimic agricultural growing conditions, they supplemented natural pollinators with honey-bee hives at the recommended density. The addition of the honey-bee hives makes their results even more unexpected – supplemental pollination increased fruit production by 33–92 %, which is above average (52 nd–69 th percentile across species; Bennett et al., 2018). More to the point, for both cultivars tested, plants in the area with high pollen quality had less pollen limitation than those in the low pollen quality area (33→92 % and 32→ >51 % pollen limitation). Given that their pollinator observations showed similar pollinator abundance in these locations, this decline in female fitness probably stems primarily from pollen limitation in quality.

The identity of the pollen affected not only the quantity of offspring, but also their quality. Using a paternity analysis, Trueman et al. (2021) were able to determine whether the identity of the pollen parent affected indicators of progeny quality, in particular seed size and oil content. Unsurprisingly, selfing led to lower quality seeds. More interestingly, particular maternal × paternal crosses produced higher quality progeny, suggesting that it is not only outcross vs. self pollen that can lead to pollen quality limitation but also the identity of both of the parents involved. Thus, to some extent, female fitness was limited not just by the amount of pollen, or the amount of outcrossed pollen, but also by the amount of the best outcrossed pollen.

This study provides insights on several aspects of floral evolution and pollen limitation. First, this study shows that there may be a high cost of producing many flowers through a decrease in the quality of pollen received. This cost should be considered when examining the evolution of masting since it would affect economies of scale, which is thought to be the ultimate driver of the evolution of masting. Additionally, at least for this species, the strong pollen limitation during a masting year with added pollinators suggests that it is unlikely that masting evolved in response to economies of scale for increased pollination. Second, this study shows that even with abundant managed pollinators, pollen limitation in both quantity and quality may still be quite severe. This finding is even more concerning given the possibility of current and future declines in native pollinators due to increasing human disturbance and climate change (Potts et al., 2010; Bennett et al., 2020). Continued research on pollen limitation across a variety of habitats and species is vital, especially as we begin seeing more changes in the environment. Third, this study shows that the effect of pollen quality, which is often ignored, can have a strong, negative influence on female fitness. Because few other studies have specifically examined the effect of the quality of pollen, it is unclear how important pollen limitation in quality is generally, making it ripe for future research. Fourth, outcrossed vs. selfed pollen may not be a specific enough comparison when thinking about how pollen quality limits female function. This will be especially true in species with self-incompatibility, or separate sexes (where hermaphrodites or males may vary in male resource allocation and therefore quality). Future studies are needed using both empirical methods for examining paternity effects and modelling studies to examine their importance in population-level dynamics.