Skip to content

How many species of plant are there?

It should be a simple question. After all, we have been describing, naming, and studying species of plants for 500 years, and the whole system of nomenclature and classification was developed in order to list plants. Estimate range widely, from 200,000 to nearly 300,000 or even 400,000 [also here] and up. There are about one million plant names, but many of these are synonyms, and many are also not species, but subspecies or varieties.

It turns out that the number harder to identify than usually thought or reported, and this has become clear because the Missouri Botanical Garden, the Royal Botanical Gardens at Kew, the New York Botanical Garden, the Global Compositae Checklist in New Zealand, International Legume Database & Information Service, the International Organization for Plant Information, and the International Plant Names Index have all collaborated on putting together a single comprehensive and synonymised list of all plant species.

Called The Plant List, it lists 298,900 accepted species, 477,601 synonyms, and 263,925 as-yet unresolved names. It’s a nontrivial achievement. Here, for instance, is the list of all species names in the genus Rosa (roses). There’s 885 of them, mostly subspecific varieties.

As taxonomy becomes crucial for our biodiversity conservation efforts in the face of major ecological shifts with global warming, such lists, more than a simple literature search but with added value from a critical analysis of each, will become a major tool.

Here’s the press release and here’s a Grauniad article complete with cheesy pun.

16 Comments

  1. Susan Silberstein Susan Silberstein

    Now for something completely different: I don’t know how many plants there are, but I’m pretty sure there are 42 Canadian actors. How do I know this? We get lots (that is, a bunch) of TV shows made in Canada, many of which are science fiction. It seems a surprising number of them feature people from the same small pool of talent. An informal survey conducted by me has determined that the magic number discovered by Douglas Adams is also the number of those Canadians. A coincidence? I don’t think so.

  2. Wait a minute. Are you counting subspecies as species? I suppose you should be asking, How many subspecies of plants are there?

  3. Jim Thomerson Jim Thomerson

    I do fish taxonomy, and have had semi serious hobby involvement with various plant groups. It is clear to me that botanists face different challenges and think about them differently than I think about fish.

  4. I am shocked by not only the number of plants, but also the descrepency in numbers. Crazy to think about how many plants there are and how many we are still waiting to learn more about.

  5. Stewart Hinsley Stewart Hinsley

    I estimate about 500,000 vascular plant species. (There are lot of cryptic species – cytotypes – that are not generally recognised by botanists.)

    The Plant List turns out to be neither complete nor accurate; it’s a snapshot of a work in progress. Quality will vary between families – I expect that it’s pretty good for Poaceae, Asteraceae and Fabaceae – but has a significant number of omissions, and some remaining duplications.

  6. Jim Thomerson Jim Thomerson

    Plants should be much better known than animals, because they are so much easier to catch. LOL

  7. John Harshman John Harshman

    Plants? I can’t even say how many species of birds there are, though I’m almost sure there’s fewer than 10,000. Main problems: numbers vary with species concept; some populations have yet to be found; some populations have been found but have yet to be recognized as distinct from others, cryptic species being a major case; and some allopatric populations may be separate species, but how do you tell? But at least botanists and ornithologists can all be happy that they aren’t entomologists.

    • BioBob BioBob

      “how do you tell? ”
      It seems pretty clear to me that one can only make definitive species determinations via some form of statistical genetic analysis with decent sample size. It would seem we also have a ways to go to understand how best this would work. At least that’s my conclusion after frustrating searches for phenotypic based divisions in insects. I can’t imagine plants would be any easier to deal with.

      • John Harshman John Harshman

        I’m afraid statistical genetic analysis will tell you nothing about the species status of allopatric populations, at least if your preferred species concept is the biological species. It may tell you that the populations have been allopatric for a very long time (or not), but it won’t tell you whether they would interbreed if given the chance. Of course, if you had a good understanding of the phenotypic and behavioral consequences of all genes, you could decide based on genetics if the populations would interbreed, but this is far beyond our current abilities.

  8. BioBob BioBob

    I would think that sufficient (to maintain single species) but rare crosses between ‘allopatric’ populations would be reflected in the lack of discontinuity in some segment of a genetic marker normal curve mapping in the entire population. I am no population geneticist, so I am speaking only from a general statistical point of view. At any rate, that is why I mention sample size as an issue. Clearly this is more probable in winged insects than it might be in sessile plants without appropriate dispersal mechanisms.

    It seems to me that we are stuck at present waiting for automated genetic testing apparatus to be implemented, but I do expect that development sooner rather than later. Clearly morphological systematics has its limitations.

    • John Harshman John Harshman

      I think the main thing that makes your idea wrong is that you don’t need crosses to maintain a single species. All you need is the failure of population divergence to create a barrier to interbreeding. The time needed for such a barrier to arise, in the absence of any contact between populations, can be highly variable, but can sometimes be many millions of years.

      Again, genetic testing will not tell you if two populations are separate species. Not even whole genomes can do that. What it can tell you is whether the populations have interbred, how recently, and how much.

      But that can’t tell you if they’re separate species. If they were isolated 10 million years ago, there will be considerable genetic divergence even if individuals, given a chance, would gladly interbreed. If, on the other hand, they were isolated 1000 years ago but a barrier to interbreeding has arisen since then, they will probably show very little differentiation, and perhaps in fact you will detect none. But in the first case, you have one species, and in the second, you have two.

      • BioBob BioBob

        There is considerable difference between theory and reality, and I am not sure in what realm our respective viewpoints are in here. My experience tells me that in the real world genetic drift is inevitable and relatively rapid in ALL populations. Your concept assumes the alternate hypothesis that populations can perhaps remain relatively genetically constant, or at least enough so that separated populations exposed to differing environments remain alike. I think your hypothesis fails in reality. We like to think that the world patchiness is relatively large scale. However insect models tell us that the environmental mosaic is quite small scale as far as organisms are concerned and therefore genetic pressure differential as well.

        It has been our experience with insect chemical resistance that populations diverge quite universally and rapidly, taking as little as 25 generations to develop chemical challenge resistance. Most environments are NOT stable and always are changing and all organisms respond accordingly. When environments are physically constant, there is the flowering of biological jostling which partitions the trophic environment forcing increasing specialization and additionally genetic change. Therefore, my conclusion has always been that genetic change is an inevitable and constant force with either physical or biological environmental change with a small mosaic such that any physical separation of populations will inevitably result in differential genetic change. Natural selection enforces the retention of this directed adaptation, in effect making “species want to be born” with a seeming universal innate rejection of ‘otherness’.

        Assuming relatively quick and large genetic drift, the insect populations model indicates quite rapid species generation in the case of separate populations without genetic exchange. Presumably some genetic discontinuity also reflects this lack of exchange.

        I think we ALWAYS assume that morphological differences of a suitable nature identify species in a way similar to genetic discontinuity. I have never seen a taxonomist rear 2 or more supposedly separate species where they could mate and produce viable offspring and thus demonstrate one way or another the verity of their cladistic assertions.

        Likewise, it is common for human error in species splitting to occur even when it seems quite patent and the science is ‘settled’. A case in point is the recent notice taken of viable hybridization of Polar Bears and Grizzly presumably due to environmental change. We do the best we can. So we assume separate species with supposed adequate morphological difference just as we would assume species with adequate genetic difference. My point was that when morphological differences are inadequate, genetic differences will have to serve in their place. The reality is that only the individual species ‘knows’ if it is indeed separate, and we generally simply make the assumption.

        I typed this rather long response quickly – pleas excuse all errors of which I assume are plentiful, lol.

      • John Harshman John Harshman

        All I can say is that our empirical experience suggests that you are not correct here. Drift will eventually cause speciation, but it takes much longer than selection would, and populations can accumulate millions of years worth of divergence in the meantime. Note that nothing is standing still here. Neutral evolution is continuing in both populations at its usual pace. There just isn’t any necessary consequence for genetic compatibility. There are many such cases known. And laboratory experiments confirm this impression.

        Now, your example of pesticide resistance isn’t drift at all. It’s selection. And selection can cause rapid speciation. However, it doesn’t have to. There are many cases known in which it apparently hasn’t, even though long periods of isolation have passed.

        One potentially important difference between us: you’re talking about insects and I’m talking (though that may not be apparent) about vertebrates, especially birds. To translate between our perspectives, you would need to apply some kind of time multiplier. In vertebrates the breeding experiments you mention happen all the time, accidentally, in zoos. Species of ducks separated for as much as 30 million years (maybe more) can produce viable, fertile offspring. Of course we still call them separate species, because in vertebrates at least a more important criterion is whether they would do so in the wild, enough for significant introgression. Isolated populations of a single species can have several percent mtDNA divergence with no discernible effect.

        I also question the assumption that separate species will be easily recognized by clear morphological differences. There are a great many cryptic species. All that’s necessary is that they be able to distinguish each other, and the cues may not be readily apparent to human senses. One can however often check these cues without actual breeding experiments. In birds, song response is used as one such method. If males in one population fail to respond to respond to songs of another as they would to rivals, that’s evidence that the populations are separate species.

        I’ll agree with you to this extent: large genetic differentiation is a good reason to investigate whether two populations are separate species.

  9. BioBob BioBob

    Thanks for you reply. I sense that you consider genetic drift as something more significant than I do. To me it is only short term genetic noise or random genetic change is some portion of DNA we might consider “junk” at this point. I see a hypothetical population gene pool which reflects selection as the driving force in almost all particulars, often reflecting all kinds of directed selection we can not envision or discern as well as that portion we do. I don’t suppose we are anywhere near capable of confirming this in the case of insects.

    I am more than happy to set aside consideration of how long allopatric populations need be separated or our understanding of the actuality of that separation but it would seem there IS some basic taxon differential as well as considerable difficulty validating that separation.

    I hope you did not conclude that I make the assumption that species are easily discerned via morpological means since my experience is quite the opposite. This is why I suggest that automated genetic statistical population analysis would be the final recourse employed to determine species. It’s not an especially appetizing prospect to me but entomologists have a somewhat larger job to do in many cases. Similar to the automation currently underway in astronomy, I expect a similar solution will be used to identify species out of necessity, with specialists trying to confirm the immense resulting findings of that process in the same manner as astronomers. At the current rate of species description in insects, the total numbers of evolving new species could exceed the rate of description of existing ones, LOL.

    In birds you may have the luxury of investigating population differentials as species. Entomologist will perhaps just assume as usual with most species in a first pass approximation ;D

Comments are closed.

Optimized by Optimole