Getting plant sex wrong (2)

I’ve been reading Horseshoe Crabs and Velvet Worms by Richard Fortey. Overall, I’ve been finding it enjoyable. However, portions of the book cause me to cringe. Here’s an example near the beginning of chapter two:

These Gondwanan coniferous trees, with their relatively large leaves and bright berries, do have a very special appearance, at least to a European accustomed to pines and firs with their dry-looking cones. A botanist would remind me I should really describe the berries as “fleshy peduncles” because they carry exposed seeds at their tips.

What’s wrong with that? Well, the trees (podocarps in New Zealand, specifically totara, Podocarpus totara and rimu, Dacrydium cupressinum) Fortey is talking about do not have berries. He is apparently aware of this, but the way he presents it suggests that these things really are berries and that describing them as “fleshy peduncles” is some kind of obscure scientific quibbling. It isn’t: “fleshy peduncle” is right, “berry” is wrong. This follows an irritatingly predictable tendency in science writing about plants: present a misleading description as though it were true, restrict the correct information to some kind of afterthought, and don’t explain it enough for readers to actually figure out what’s going on. This both misinforms readers and gives the impression that the situation is hopelessly complex and cannot be understood by anyone who doesn’t have a PhD and a labcoat.

So, OK, here’s what a fleshy peduncle is. In conifers, a peduncle is a stem that bears one or more cones. In this case we’re talking about seed cones, analogous to the seed cones you’d see on pine, spruce, Douglas fir, etc. You know, these things. In the podocarps Fortey is discussing, the seed cones are very small, having one or two small leaves each with a single exposed seed. In these species, the peduncles bearing the seed cones are swollen, fleshy, edible, and look somewhat like berries. Why aren’t they berries? A berry is a kind of fruit. Fruits are found only in flowering plants (not conifers) and instead of leaves with exposed seeds they have carpels. Carpels are leaves that have been folded and fused into little chambers with seeds inside of them, protected from the environment.

That is more complicated than just saying “these trees have berries; well, technically they’re fleshy peduncles”. However, it’s also informative. Authors writing about science should aim to inform readers about science rather than reinforcing misconceptions and presenting science in a dismissive and uninformative manner.

Here’s a similar example from near the end of chapter 3:

Despite their apparent simplicity, Porphyra and Bangia have quite complex life histories. Cells of the “weed” contain only one package of genetic information; they are described as haploid. A second phase in the life history of these seaweeds is called the Conchocelis stage, which makes miniature branching plants, some varieties of which inhabit the borings they make inside seashells. These plants are so different from their “parents” that they were once given the separate generic name, Conchocelis, which is now only retained as a label for one stage in the life cycle. Conchocelis plants are the diploid, or the sexual, stage of the red algae. The leafy stage releases gametes that mate with one another, thereby doubling up the genetic content; this produces spores that can germinate into Conchocelis.

Some of the basic ideas here are correct. However, the definitions of “haploid” and “diploid” are misleading, the processes of fertilization and sporulation are conflated, and describing Conchocelis as “the sexual stage” is incorrect.

First, what does “one package of genetic information” mean? A package of genetic information could refer to a few different things. It could possibly refer to a gene, or a chromosome, or a nucleus. When discussing haploidy and diploidy, we’re talking about chromosomes, which are long strands of DNA, each containing many genes as well as regulatory sequences and non-coding “junk” DNA, bound up with proteins. So why not just say “one chromosome”? This would be less ambiguous, but it would also be wrong. One vs. two chromosomes is not the distinction between haploid and diploid cells. Instead, the distinction is how many copies of each chromosome is present. In a haploid cell, each chromosome present as a single copy. In a diploid cell, chromosomes are present in pairs. We could easily reword the second sentence of this quote to say “Cells of the ‘weed’ contain only one copy of each chromosome; they are haploid.”

Second, Fortey says that “the leafy stage releases gametes that mate with one another, thereby doubling up the genetic content; this produces spores”, which conflates fertilization and spore production. The union of two gametes is fertilization, the process by which we move from haploid cells to diploid cells. The cell produced by fertilization is not a spore, it is a zygote. The diploid zygote then goes through several mitotic cell divisions to produce diploid spores. By suggesting that fertilization produces spores, Fortey is simply skipping this stage of the life cycle and implying that fertilization directly produces spores, which is incorrect.

Third, Fortey describes Conchocelis as “the diploid, or the sexual” stage of Porphyra and Bangia. Sex consists of the production of gametes and their subsequent union in fertilization. The Conchocelis stage is not directly involved in this process: it doesn’t produce gametes and it isn’t produced by fertilization. Instead, the diploid spores mentioned above can grow into the Conchocelis stage, and the Conchocelis stage produces haploid spores by meiosis (Porphyra and Bangia produce multiple types of spores, which is rather odd). The Conchocelis stage is part of the whole life cycle in these algae and the whole life cycle includes sex, but that’s as close as it gets. It’s like describing a human liver as “the sexual stage” in humans. A functional liver is a necessary component of the human life cycle and the human life cycle includes sex, but the liver doesn’t have any direct role in sex.

The life cycles of Porphyra and Bangia are fairly complex and difficult to describe clearly and accurately. Fortey had two good choices here: either don’t bother with it since it’s not really necessary for the narrative of the chapter or go into the detail needed to convey what’s going on. Instead he makes a third, bad choice: discuss the topic briefly, confusingly, and incorrectly. All readers are likely to get from this passage is that something weird and confusing is happening.

That said, Fortey is doing far better at botany than Bernd Heinrich, who in The Trees in My Forest repeatedly refers to “flowers” of conifers. This is just wrong. Pines do not have flowers. At least Fortey tells us that the “berries” of podocarps are in fact fleshy peduncles, even if he does so in a rather unhelpful fashion. And maybe we can cut him some slack, since his book isn’t primarily focused on plants, much less trees or conifers in particular. Heinrich, on the other hand, wrote a book about trees, with long discussions of conifers, and gets it completely wrong. He tells us that conifers have flowers and provides no explanation nor any indication that this might not be consistent with what we actually know about botany.

Gonochorism, a neglected term

From 11 Oct 2010.

I’ve been reading a few papers recently on the evolution of sex and it seems to me that an important distinction is often overlooked, the distinction between sex and gonochorism. Theoretical work on the evolution of sex often contrasts gonochoristic sexuality, i.e., having separate male and female individuals, with asexuality. If females have a constant number of offspring in gonochoristically sexual and asexual lineages, the asexual lineage will have twice the reproductive output. Suppose all females have four offspring; a gonochoristically sexual female will have two sons and two daughters, while an asexual female will have four daughters. Those four daughters will have four daughters, while the gonochoristically sexual female’s two daughters will each have two daughters. There are some simplifying assumptions here that may not hold–i.e., the sex ratio is 1:1, males are assumed to make only a genetic and not a material contribution to reproduction, and it is assumed that no physiological reliance on sexuality exists. The second is apparently the most problematic; if males provide food or other resources that can increase the per-female reproductive rate in a gonochoristically sexual species, a competing asexual female will not be able to achieve the theoretical doubling in reproductive rate. This is likely to be a factor to some extent among many mammals and birds, but presumably not in reptiles & insects. The third assumption has also been shown not to hold in some cases. Whiptail lizards (genus Aspidoscelis) have a hormonal reliance on mating behavior, and engage in pseudosexual behavior. This apparently reduces the efficiency of reproduction and prevents asexual whiptails from maintaining an equal per-female reproductive rate compared to their gonochoristically sexual relatives.

In any case, apart from the violations of simplifying assumptions that mitigate the potential doubling reproductive output of asexuals in some cases, the potential reproductive inequality between gonochoristically sexual and asexual lineages presents presents a significant challenge to any account of how sexuality could have evolved and how it could be maintained. Resolving this difficulty has been the focus of extensive research. However, as I suggested initially, I think the problem is in part poorly framed. The reproductive inequality that papers on the evolution of sex seek to address is not specifically one relating to sexuality vs. asexuality, but a result of the division of the sexes into separate individuals, i.e., gonochorism. If we compare hermaphroditic (loosely speaking… in plants this term and ‘gonochoristic’ can be misleading, see the post regarding Pollan’s “Botany of Desire” on PBS) lineages with asexual lineages there isn’t any general theoretical reason to expect a strong reproductive inequality between the two. All offspring in each case will be reproductive. We might expect marginally lower reproduction in hermaphrodites because some of their energy is devoted to male functionality, but, as the biological cliché goes, sperm are cheap; this should not be a major factor. We might still expect asexuality to be favored under certain circumstances, particularly if gamete (or, in flowering plants, pollen) transfer is inefficient. If populations are sparse and encountering other individuals of the species is infrequent or energy-intensive (or, in flowering plants, if pollinators are scarce or inefficient)–i.e., if it is difficult to engage in sexual activity–this should tend to favor asexuality. That aside, in general it seems to me that the arguments already advanced to explain the evolution of sex, which focus in various different ways on the fact that sex encourages genetic diversity, both in terms of numbers of alleles present and in their ability to be assorted in offspring independently of each other. Genetic diversity provides the substrate on which natural selection can act, and is thus a prerequisite for an evolutionary response to selective pressure. If either abiotic or biotic conditions change, a sexual species should be able to adapt to that change much more quickly and effectively.

Sexuality seems to me to be relatively easy to explain. Gonochorism, on the other hand, is very difficult to explain. Compare a gonochoristic and a hermaphroditic lineage; the latter should have twice (more or less) the reproductive output, but both have the genetic advantages of sexuality. The hermaphrodites should win easily. So, why are there so many gonochoristic species? I’m sure there must be literature that addresses this specific point, although I haven’t run across any in my own fairly cursory reading in the field, but I think many miss this point and simply take gonochoristic sexuality and asexuality to be the two available options. They are not.

(Parenthetically, a further confusion that I’m ignoring for present purposes is that between self-fertilization (or self-pollination…) and outcrossing. Self-fertilization is sexual, but has broadly similar detrimental genetic consequences to those asexuality. In plants, the evolutionary dynamics of self-compatibility / self-pollination vs. self-incompatibility are subject to some of the same general considerations as the evolution of asexuality vs. sexuality (which in plants is generally ‘hermaphroditic’, but occasionally ‘gonochoristic’…), but of course it isn’t quite the same. The main complication added is that selfing plants can be either facultatively or obligately selfing… anyways, enough for one day.)

Getting plant sex wrong

From 28 Oct 2009.

Watching “Botany of Desire” on PBS. I’ve generally been a bit ambivalent about Michael Pollan, but about 36 minutes in he veers into “just plain wrong” territory. “Before that [before the evolution of angiosperms] you had this greener, sleepier world where things reproduce usually by cloning, by spores that were genetically identical to their parents”. Regarding cloning–yes, there were a fair number of clonal plants before angiosperms evolved, but there are also plenty of clonal angiosperms. Regarding spores–he is right that sexual recombination isn’t involved in producing spores but: 1) this does not mean the spores are genetically identical to the parent–they aren’t; 2) angiosperm reproduction involves the production of spores as well, so this is not a difference between angiosperms & non-angiosperms.

Brief recap of plant life cycles: the dominant portion of the vascular plant life cycle is the sporophyte (on the other hand, gametophytes are dominant in non-vascular plants: mosses, liverworts, hornworts), which is diploid (has two sets of chromosomes, just like all stages of the human life cycle except sperm & eggs). The diploid sporophyte produces spores by meiosis. Meiosis halves the chromosome number, so the spores are haploid. Whereas the sporophyte has two copies of each gene (excluding the rare cases in which plants have sex chromosomes), each spore has one copy of each gene. It’s the same as the relationship between, for instance, a human male and one of his sperm cells (except, for the sake of nit-picking, that humans have sex chromosomes and most plants don’t), and is not genetic identity. However, whereas human sperm & egg cells do nothing more than unite to form a zygote, plant spores undergo cell divisions to produce gametophytes. Gametophytes are a multicellular haploid stage in a vascular plant’s life cycle, and they produce gametes (sperm and eggs) through mitosis. Gametophytes exist in all plants, but are quite small and dependent on the sporophytes in angiosperms. Pollen grains are male gametophytes. Inside each ovule lives a female gametophyte. When the sperm and eggs produced by gametophytes join in fertilization, we are back at the diploid sporophyte level. Very short version: diploid sporophyte produces spores; spores grow into haploid gametophyes; gametophytes produce sperm & eggs; fertilization takes us back to a diploid sporophyte. Although (with minor exceptions; e.g., the genus Vittaria, some species of which exist solely as gametophytes) all plants produce spores, in no plants is the production of spores itself sexual. The production of spores is part of the sexual process, however; without the gametophytes they give rise to, you can’t get gametes. Lest you think Pollan just misplaced a word or two, he continues: “And then you have this incredible explosion of diversity that happens with this new strategy. It was an incredibly successful strategy. It allowed you [by which he means angiosperms] to move your genes around, it allowed you to evolve much quicker because sex creates variation.” Nope, he didn’t just misplace a word or two, he’s really saying that the key innovation of angiosperms relative to earlier plants was sex.