When exploring the diversity of a genus I’m familiar with (Allium) in a place I’m not (California) it is, of course, reasonable to start with the low-hanging fruit. And so it is that I saw a couple of observations labelled Allium abramsii and thought to myself, “Well, that looks like a species I can’t get wrong, let’s go through the iNaturalist observations identified as Allium abramsii, as well as the other Allium from the same area, and see if we can correct some observations.” Not long after, I encountered some observations like this one and this one. Obviously not Allium abramsii, so what species is it? I ended up at Allium fimbriatum, but was not very confident. Over the next couple of days, I gradually realized that I had made an error—these plants are not Allium fimbriatum, but I couldn’t find any other plausible candidate. Eventually, only one plausible hypothesis remained: in the existing treatments of Allium in California, both of these species are called “Allium abramsii”.

Some further investigation revealed that the type of Allium abramsii belongs to the second species I encountered. My “Allium fimbriatum”, then, is the true Allium abramsii. The other plant is either an undescribed species or a species that had been described but then (mistakenly, I think) obscured in synonymy. I checked the Consortium of California Herbaria and SEINet for type specimens that might belong to this species. Then I checked the Flora of North America and Jepson’s A Flora of California for names that might apply to it. I didn’t turn up anything, though my search is not yet as thorough as it should be. Pending further developments, I believe this is an undescribed species. We’ll call it “Allium cf. abramsii” for the moment. Here are the observations on iNaturalist that I’ve found for Allium abramsii and Allium cf. abramsii and a couplet that distinguishes the two:

Margins of the inner tepals ragged-denticulate in the distal (widely spreading) half; flowers red-purple, usually ± concolorous throughout, sometimes paler toward the center of the flower when viewed from the front and toward the base of the erect, ± tubular portion of the perianth when viewed from the side; plants of the southern Sierra Nevada north of Granite Gorge to near Mammoth Lakes and Wawona . . . . Allium abramsii
Margins of the inner tepals crispate in much or all of the distal (widely spreading) half; flowers blue-purple to lavender, usually much paler toward the center when viewed from the front and toward the base of the erect, ± tubular portion of the perianth when viewed from the side (rarely, in plants with darker flowers, more or less concolorous throughout); plants of the southern Sierra Nevada, south of Granite Gorge to the vicinity of Bald Mountain . . . . Allium cf. abramsii

There may be other characters separating the two but the difference in color—unfortunately, always problematic given the inadequacies of our spectral vocabulary—and the crispate margins of the inner tepals are the obvious and eye-catching features in photographs of the plants. The crispate margins of the inner tepals also serve to easily separate Allium cf. abramsii from Allium denticulatum, to the south.

This is a case where both the advantages and disadvantages of working from digital images of live plants rather than from herbarium specimens become apparent. There are plenty of limitations when working from photographs. The inability to measure parts, to look at them under magnification, and even to see them if they don’t happen to be within the frame are probably the most obvious and universal limitations. In Allium, features of the bulb coat have played a central role in prior taxonomies, and this is entirely invisible except on the rare occasions when someone has photographed the bulb coat under a microscope. There are also advantages. In photographs, I find Allium cf. abramsii to be one of the most distinctive and easily recognized Allium in the western United States. So long as a photograph is clear enough to tell if the inner tepal margins are crispate, it is impossible to mistake it for any other taxon. In herbarium specimens, the differences between the two can be seen but are subtle. That it has not been named is evidence enough of this! In the Flora of North America treatment, McNeal mentions that the margins of the inner tepals are sometimes crispate but apparently attaches no particular importance to variation in this feature. The ability to look at a set of observations at once and in geographical context is also very helpful. Were I looking at the same set of plants as herbarium specimens, it probably would have taken me a while to work out that the two forms are allopatric, if it came to my attention at all. On iNaturalist, all it takes is applying some label to sort observations into the two groups and the geographic pattern is obvious.

I think I will likely publish a name for Allium cf. abramsii, though I don’t know what name I might give it. I would feel silly naming a plant I have not seen in person, so I hope to make a trip to see it in the field in 2023—and I will need to describe the bulb coat, of course! I’ve made an initial pass through herbarium specimens that have online images, and arrive at the following list for Allium abramsii: Fresno County: Hall & Chandler 201, 15-25 June 1900 (TYPE! CAS, MO, NY, UC, US); Cole 30, 17 June 1972 (CSLA); Holland s.n., 16 May 1965 (OBI); McNeal 3147, 26 May 1986 (BRY, UT); McNeal 3707, 29 May 1986 (BRY, UT); Morton 2815, 4 July 1941 (US); Stebbins s.n., 15 May 1985 (FSC). Madera County: Constance 2389, 1 July 1938 (NY). And for Allium cf. abramsii: Kings County: Clifton 13928, 28 March 1986 (PUA); Gill 80, 4 April 1959 (UCSB). Tulare County: Baker 7334, 30 May 1974 (LOB); Ganley 721, 18 June 1971 (MACF); Haultain s.n., 13 July 1995 (THRI); Keil 18835, 15-16 June 1985 (OBI).