My thoughts on Ribes hudsonianum varieties
This journal post is to respond to the comment under this post: "I see iNat recognizes the mentioned (FNA) variations, what are your thoughts on splitting the species?"
(The question is in regards to whether or not to follow the practice of recognizing the western US Ribes petiolare as either a distinct species or distinct variety apart from the closely related R. hudsonianum of Canada and the Great Lakes)
I've gone back and forth on it a bit but recently I am more of the opinion that there should be no subspecies or varieties recognized. It is hard to find the exact arguments behind why they should be split but I tracked down this keybreak from an old flora linked here: https://www.biodiversitylibrary.org/page/8010030#page/409/mode/1up
- Racemes 6-12 flowered; bracts 2-5 mm long; leaves firm, villous-pubescent beneath, with a shallow or no basal sinus: -- R. hudsonianum/ var. hudsonianum
- Racemes 25-50 flowered; bracts minute, 1-2 mm long; leaves thin, essentially glabrous, usually with a deep basal sinus: -- R. petiolare/ var. petiolare
I don’t have any experience looking at most of the traits in this key. The majority of my experience with this taxon is by looking at pictures on iNat, so even if there are differences in the bract length and and the leaf traits, I’m not familiar enough with how the species expresses these traits over its range because they are hard to see in photos (I’d have to spend quite a bit of time in the herbarium to make this call).
However, the one trait I can speak to is the flower count of the racemes. It is definitely true that there is an overall geographic correlation regarding flowers per inflorescence, with the more northern and eastern populations having notably short and stout racemes with usually less than 12 flowers, and the more southern populations (in the western US) having long, spikelike racemes with over 50 flowers (such as in this population from near my home, near the absolute southern limit of its distribution). This difference in flower count is also clearly associated with a difference in habitat, as R. hudsonianum seems to be confined to montane riversides in the western US, whereas it seems to favor lower elevation habitats, often near lakes or swamps, in the more northern and eastern populations.
Despite this, the problem with interpreting this broad difference in flower count as valid evidence for a split is that it forms more of a gradient rather than a clear split. The key above implies that there are no plants with between 13-24 flowers per inflorescence, which would be strong evidence for a split if it were true, but it is not. I didn't have to do too much digging to find this population which looks to have ~17 flowers per raceme, comfortably between the two suggested figures in the key. This plant was also observed from southern Canada which is around where we would expect the ranges of the two proposed varieties to overlap.
I would interpret the existence of these specimens with intermediate flower counts in this area as solid evidence that even though there are broad trends of geographically correlated differences between the western US populations and the rest, they don't form a clear cut enough line to justify a nomenclatural split. After all, if we did split, we would need to be able to confidently place any specimen in either one of two boxes, which would be impossible in those grey zones of range overlap. This solution is obviously not perfectly satisfying either given that the plants from the far northern extent of its range are clearly different than the plants from the far southern part, but a perfect solution is not really possible using the tools our current species concept. Lumping all of R. hudsonianum / petiolare into a single name just seems like the only practical and reasonable choice at the moment.
Of course, I again have to emphasize that I don't have much experience with this taxon, having only seen it in habitat in northern Utah and having not looked at any herbarium specimens, and none of this is set in stone. It may turn out that a more detailed and comprehensive examination of all the traits of R. hudsonianum, over its entire range, may reveal some more cryptic distinctions within the species and cause us to realize that a split was justified after all. For now, though, I think FNA got it right by lumping the species under one name.
One more note: as stated above, the most recent FNA does not recognize any varieties for R. hudsonianum, listing this as their argument why:
"Plants of Ribes hudsonianum with leaf blades that are pubescent abaxially and mostly lack sessile glands, and have ovaries with sessile glands, have been recognized as var. hudsonianum; those with leaf blades that are shaggy-hairy abaxially and sessile-glandular, and have ovaries lacking such glands, have been named var. petiolare. Variety hudsonianum has a more northern distribution; var. petiolare is western. Where their ranges overlap, for instance in Saskatoon, pubescence density varies continuously and does not correlate with presence or absence of glands (V. L. Harms, pers. comm.)."
If we take the citation at its word, this continuous variance in pubescence is yet more evidence that a split is unwarranted. Furthermore, the stated idea that var. petiolare should have ovaries lacking sessile glands does not hold up. As I have seen in person, the populations from the far southern limit visibly have tons of sessile glands on their ovaries, despite being allegedly referrable to var. petiolare in every other way (massive flower per raceme count, far southern/western distribution). Again, this does not necessarily prove that a split is unreasonable, but it should be clear by now that most of the characters traditionally used to justify a split don't hold up under scrutiny.
Comentarios
Agregar un comentario