How does mixed mating persist, when self-pollination ensures a plant won’t reproduce?

Being a seed from self-pollination guarantees failure in Eucalyptus regnans. Yet despite this, new research shows a mixed-mating system. including selfing, is evolutionarily stable.

In Tasmania, and some of Victoria, you can find the tallest flowering plant in the world. It has a variety of names, Mountain Ash and Swamp gum are two. The scientific name makes clear it’s not an Ash (Fraxinus) and references its size: Eucalyptus regnans. The tallest trees are astonishingly tall, being comparable to the tallest redwood trees. It is possible that they could grow even taller than a redwood if left long enough. However, their life cycle needs its forests to burn for the next generation to come through. A new paper by Rod Griffin and colleagues, Life cycle expression of inbreeding depression in Eucalyptus regnans and inter-generational stability of its mixed mating system, examines that life cycle, starting from the flowers.

Eucalyptus regnans
Forest of E.regnans at Mt Field National Park, Tasmania. 250 years old and approx. 75m tall. Photo: R Wiltshire. Click to enlarge.

E. regnans has a mixed mating system in that the flowers can produce seed by swapping pollen with other plants, outcrossing. The flowers are hermaphroditic, meaning they have male and female components, like many flowering plants. If they have male and female parts, you might wonder why they don’t just pollinate themselves. Sometimes they do.

Self-pollination has its drawbacks for plants. It reduces gene flow through a population. It might seem E. regnans has found a way to reduce self-pollination; the flowers are protandrous. The male parts of the flower mature before the female parts. But an E. regnans can have one and a half million flowers. It’s a massive food source for the plant’s insect pollinators. The massive flowering means it’s possible for an insect pollinator to carry pollen around to different parts of the same tree as it forages.

Once the flowers set fruit, they store the seed in wood cases on the tree for up to three years before they fall, but this doesn’t usually help the species. “The species has very small seeds which do not persist in the soil seed bank,” said Rod Griffin.

“For successful germination, they need an open mineral “ash bed” which is produced by fire at irregular intervals. The trees carry several years of seed crop in their woody capsules which open after the fire and produce a huge seed rain.”

The fire clears the old trees and opens the landscape for a new cohort of plants. Prof Griffin said: “The resulting regeneration is more or less an even-aged monoculture (with regard to tree species), though of course fire intensity will vary across the landscape, so there can be a mosaic of patches of different ages.”

Prof Griffin underlined how fire is critical to reproductive success. “If there is no fire for 400 years or more, then there can be a succession to Nothofagus forest, and the eucalypt component is lost.”

The tree population starts from the last batch of seeds in the trees after a fire. So how does inbreeding depression, where the self-pollinated trees lose out to the outcrossed trees, happen? The paper in Annals of Botany publishes the results of a 29-year study, following the trees as they grow up. Though it wasn’t originally designed to take that long.

“These experiments were part of a series of investigations of the reproductive biology of E.regnans designed to underpin design and management of new commercial seed orchards,” said Prof. Griffin.

“We were aware from an earlier study (Eldridge & Griffin 1983) that selfed progeny showed strong inbreeding depression for growth and we wanted to understand how the trees produced selfed seed and the consequences for natural regeneration (a system still widely used in the native eucalypt forests) and in plantations. This wider project was wrapped up in the 1980s, and a number of publications were produced, e.g. Griffin et al. 1987; Griffin & Cotterill 1988; Moran et al. 1989, but fortunately the landowners kept the major field trial for wood production and permitted ongoing access.

“A PhD project investigated the stand structure at age 15 (Hardner & Potts 1997). Prior to harvest we revisited the trial and observed that seed production had started on some trees. This presented a unique opportunity to investigate the expression of inbreeding depression through the full pre-reproductive phase of the life and to compare the mating system with that of the previous generation.”

The key to the success of the outcrossed plants is down to the sheer quantity of seed a Eucalyptus tree produces. Prof Griffin said: “A tree may produce, millions of seed and yet all it requires to do to maintain population continuity is to reproduce itself once!”

E. regnans seedlings en masse
Seedling regeneration of E.regnans 1 year after fire. Tooms E.Tasmania. Photo: M Neyland.

“The combination of intense competitive thinning and inbreeding depression which favours outcrosses, is enough to ensure that the reproductive population is effectively outcrossed irrespective of the proportion of selfed seed produced.”

This intense competition at the seedling stage does carry some cost to the gene pool. Prof Griffin said: “Since all inbred plants are eliminated irrespective of their specific genotypes there is no opportunity to purge deleterious genes and so reduce inbreeding depression over generations (Lande et al. 1994).

“It is probably no co-incidence that the world’s tallest angiosperm has a regeneration system which must favour selection for vigorous height growth. As we show in the paper, if a tree has not achieved a dominant position in the canopy by age 10, it is unlikely to reproduce.”

Botanists are used to working with short-lived species that turn over a few generations in a year. Working with trees is somewhat different. “As tree breeders and forest geneticists, we are quite resigned to working with the timelines dictated by species’ biology!” Prof Griffin said.

“Once we had the trees in the ground and a secure access and management agreement with the landowners, all that was needed was to maintain academic “corporate memory” and to resource the collections and lab work needed for timely collection of the next tranche of data.”

The result is a study that gives scientists a different insight into evolutionary and population biology. “There are many papers presenting models of expectations under different mating system and inbreeding assumptions but rather few empirical studies such as this… especially with woody perennial species,” said Prof Griffin.

“There are also applications in silviculture and tree breeding. Foresters need to know the consequences for plantation productivity of planting a mixture of self and outcross seedlings; breeders should be concerned with minimizing inbreeding in their seed orchards.”

“Our work provides empirical evidence that a mixed mating system can be maintained in a species which also exhibits strong inbreeding depression in the pre-reproductive phase of the life cycle. We are happy to make the raw data available to anyone with an interest in modelling these dynamics.

“The practical lessons in terms of silviculture and breeding are now well established, but if we had the time and resources, there are many biological questions which could be explored. For example, it would be relatively easy to find populations where the relative importance of reproductive assurance could be examined.

“Eucalyptus is a large genus with, in general, a mixed mating system (Byrne 2008), but also strong variation in ability to tolerate and recover from fire. It would be interesting to take species with a range of life history traits and compare the mating system/inbreeding depression to see if evidence for direct selection on the mating system could be detected. In the other large Australia-wide woody genus Acacia there is large inter-specific variation in self fertility (Gibson et al 2011). Why the difference?”