Rubus ursinus macropetalus

Short version

Am I missing something, or does Rubus ursinus macropetalus not occur wild in central or southern California?

We may need to reverse a bunch of IDs, so as to avoid "the algorithm" reinforcing a fallacy - as it is wont to do sometimes.

Longer version

I've noticed bunch of "Rubus ursinus macropetalus" IDs popping up in our neck of the woods on the Central Coast of California. I wondered what that taxon might be and how I might identify one. It turns out that it's not that easy to know how to ID it, and in fact, my current theory is that it may not exist at all in central or southern California. Read on, and let me know if you have any thoughts on the matter.

As at May 23rd 2022, there are 257 observations of macropetalus (199 Research Grade), most of them in the Bay Area of California, and quite a few south to the Central Coast and Southern California. There's also a few in Oregon, Washington, & British Columbia.

Here's what some floras and keys have to say about macropetalus. Very few recognize it. For example:

  • Munz & Keck (1959) - Rubus macropetalus recognized as a species, distinguished by having "pinhead glands" on pedicels and calyx, distribued from Lake and Butte counties (northern California, well north of the Bay Area) north to British Columbia, with some forms in cultivation.
  • Plants of the World Online (a database, not a key) accepts macropetalus as a species, and cites the subspecies as not being recognized.

...and many floras don't recognize it at the species or varietal level (* other than as a synonym for R. ursinus), these including:

  • Flora of North America (online) (*)
  • Jepson (1923, 1925, 1959)
  • The Jepson Manual 1st ed. (1993) (*)
  • The Jepson Manual 2nd ed. (2012) (*)
  • Thomson (1961) - Santa Cruz County
  • Hooker (1970) - San Luis Obispo County
  • Matthews (1997) - Monterey County
  • Matthews & Mitchell 2nd ed. (2015) - Monterey County
  • Beidleman & Kozloff (2003) - San Francisco Bay Region

The history of "macropetalus" is inter-twined with the ambitious history of genus Rubus (see Weber 1996), which at one point boasted perhaps an order of magnitude more "species" than it does today. The farthest I've looked back so far is Rydberg (1915) who included macropetalus it as a species in a treatment of the Rosaceae. Bailey's (1933) monograph circumscribed 100s of Rubus species; so unsurprisingly, macropetalus was one of them - distinguished by the "pronounced glands on pedicels and calyx" among other things, distributed from BC down to Lake County, California, and also the primary cultivated form among related taxa. Darrow & Longley (1933) echoed Bailey, noting R. ursinus as the "coastal representative of R. macropetalus". Brown (1943) reduced species macropetalus to a variety of ursinus, but retained geography described by previous authors. Taylor & MacBryde (1977) then elevated macropetalus it up to be a subspecies of ursinus, noting that the varieties are geographically distinct, thus meeting the criterion for recognition as a subspecies. They were writing about British Columbia but their recognition of the subspecies level appears to be the first and lasting one. An abstract by Anderson & Finn (1996) retained the subspecies designation, and strengthened the evidence for it with field experiments.

As iNaturalists at the end of the day, we're left with macropetalus being an ID available to us with a long historic tradition of being glandular and northern. However, the current statewide floras exclude it altogether. The only statewide flora that included it restricted it to the north, and it is absent in floras around the Bay Area and in counties south.

Munz & Keck (1959) would perhaps be the most recent reference for ID'ing it (unless there are recent northern California floras that cover it?).

It would then seem that the ID "macropetalus" should not be used in central or southern California unless the glandularity is clearly evident, ideally accompanied by other characters detailed in Munz & Keck (1959).

Posted on May 23, 2022 09:53 PM by fredwatson fredwatson

Comments

Before the most recent computer vision update Hesperomecon linearis would always be called Platystemon californicus by the iNat AI. I thought it was just because H. linearis is not nearly so common as P. californicus and they look a bit alike. But one evening I was looking through a bunch of the P. californicus on iNat and noticed that many of the "research grade" P. californicus were actually H. linearis as you have pointed out here with macropetalus. So I spent about three or four hours going through all the California Platystemon and found a dozen or so incorrect research grade ID's and "voting" for Hesperomecon. In nearly all cases the "authors" and their "reviewers" agreed with me. Now, after the most recent computer vision update the AI gets the identifications correct. I'm not sure it was due to me :-) but I like to think so. The computer vision is only as good as the image set it is trained on. I think you have a good case for doing the same here. I think it helped that I copied and pasted a little blurb each time explaining why I thought the ID was Hesperomecon linearis and not Platystemon californicus.

Posted by hkibak almost 2 years ago

Thanks. Yeah. That's one of my intents here. Test the waters a bit to see if I haven't missed something. Anyone? Please chime in :) .....And then if the waters are clear, suggest IDs on the obs that look incorrect, with a link to this post.

Posted by fredwatson almost 2 years ago

Nice writeup. It does seem moreso like a classification glitch in this system than a proper misidentification based on any morphology, if this distinction has any cogency... Computer vision often will only take these images as far a Rubus (a welcome precaution), so I suspect the error may have as much or more to do with the indicated common name for ssp. macropetalus 'California blackberry', which is just as often applied to R. ursinus.

You have me now interested to distinguish these taxa, and I wonder if you can point to or embed any preferred diagnostic photos you've found, showing this glandularity? Is below maybe what we're looking for in macropetalus, as opposed to the typical prickles and hairs still found on central coast ursinus? https://www.calflora.org/entry/occdetail.html?seq_num=mu19667

Posted by andym almost 2 years ago

Thanks @andym . Yes, that CalFlora obs you linked shows glandular hairs (black dots on stalks.) on the calyx, and to a lesser extent on the pedicel. So this would appear to qualify as subspecies macropetalus. The northern California location is also more consistent with the recognized literature that references macropetalus. The glandularity in that CalFlora obs is not particularly pronounced, compared to the glandular hairs on various other unrelated taxa. So, I'm curious what the most glandular macropetalus might look like.

Posted by fredwatson almost 2 years ago

@fredwatson I think I did a bunch of IDs for this in the past by accident when I was putting "California blackberry" in and not noticing that name applied to a subspecies and not just plain R ursinus.

It seems that neither POWO or Jepson recognize this subspecies as valid, so it should probably be folded in with R ursinus. I'll add a flag to the taxon and see if anyone wants to change it.

Posted by graysquirrel over 1 year ago

Thanks @graysquirrel. If it were up to me, then before folding the taxon I'd look at some other floras closer to the core of the taxon's geographic range e.g. Hitchcock et al. Flora of PNW. I don't own that one, yet...

Posted by fredwatson over 1 year ago

Add a Comment

Sign In or Sign Up to add comments

Gracias al apoyo de:

¿Quiere apoyarnos? Pregúntenos cómo escribiendo a snib.guatemala@gmail.com