Astragalus Robbinsii var. minor

Stream banks, mountain meadows, damp thickets, moraines, often with willows and shrubby cinquefoil, in sun or shade, in moist or cool humus, or alluvial soil overlying granites or limestones, widespread and common but of interrupted range in the southern and central Rocky Mountains from ± latitude 53° N. in Alberta and British Columbia south to northwestern Wyoming, and again in the high mountains of central and southwestern Colorado, west to the Marble Mountains in British Columbia, the Okanogan Plateau in eastern Washington, and the high mountains of central Idaho, in this area mostly in the timber belt at ±4000-10,500 feet, in Colorado (and in northern Montana) ascending to dry but cold, rocky tundra and up to 11,900 (possibly "12,500") feet, in Alberta extending rarely along streams out into the prairie zone; also disjunctly in the Yukon Valley, east- central Alaska, and at or near sea level (in a form approaching var. Harringtonii) on the shores of Glacier Bay, coastal southern Alaska; also remote from the Cordilleran range on moist talus and ledges of shaded or north-facing cliffs, very local in the mountains of northern and central Vermont (Mt. Mansfield; Willoughby Mt. and vicinity; Pond Mt. Wells, Rutland County) and on the St. John River in extreme northern Maine; again isolated on sea bluffs of Cape d’Or, Cumberland County, Nova Scotia.—Map No. 5.—June to August.

Astragalus Robbinsii var. minor (Hook.), comb, nov., based on Phaca elegans (3 minor Hook., Fl. Bor.-Amer. 1: 144. 1831.—"...on the higher summits of the mountains. Drummond."—Holotypus, collected in 1825-7, mounted on the same sheet with the holotypus of Phaca elegans, but distinguished by the legend "ß, Hook.," K!—A. minor (Hook.) Jones, Contrib. West. Bot. 10: 64. 1902, nom. provis. A. eucosmus fma. minor (Hook.) Rouss. in Contrib. Lab. Bot. Univ. Montreal 24: 43. 1933, quoad nom.

Astragalus Blakei (Rev. Joseph Blake, 1814—1888)) Egglest. in Bot. Gaz. 20: 271. 1895. —"Type stations: Willoughby Mt., Smuggler’s and Nebraska Notches, Vt., Willoughby Mt., Westmore, Vt. Rev. Joseph Blake (previous to 1856), etc. Smuggler’s Notch, Mt. Mansfield, Stowe, Vt., C. G. Pringle, etc. Nebraska Notch, Mt. Mansfield, Underhill, Vt., C. G. Pringle and F. H. Horsford... "—Spm. authent., Blake (from Willoughby Mt.) in June, 1861 (NY, PH), and June, 1864 (PH)! Eggleston’s collections of 1894 cited above were distributed under the nomen nudum of A. Robbinsii var. borealis (one sheet at NY named A. Blakei) and are doubtless paratypi.—A. Robbinsii var. borealis (northern) Egglest., l.e., in syn. Atelophragma Blakei (Egglest.) Rydb. in Bull. Torr. Club 55: 125. 1928. Astragalus Robbinsii var. Blakei (Egglest.) Barneby ex Gleas., New I11. Fl. 2: 419. 1952.

Astragalus Macounii (John Macoun, 1832-1920, pioneer Canadian botanist) Rydb. in Mem. N. Y. Bot. Gard. 1 (Fl. Mont.): 243. 1900.—"British Columbia: Deer Park, Lower Arrow Lake, [June 5] 1890, John Macoun 25 ... "—Holotypus, NY! isotypus, US!—Atelophragma Macounii (Rydb.) Rydb. in Bull. Torr. Club 32: 660. 1905. Astragalus labradoricus var. occidentalis sensu Jones, Rev. Astrag. 134, pro max. parte, exclus. basonym. 1923.

Atelophragma Collieri (the collector) Rydb. in Bull. Torr. Club 55: 128. 1928.—"Type collected at Eagle, Alaska, June 29, 1902. A. C. Collier 50... ’’—Holotypus, US!—Astragalus Collieri (Rydb.) Porsild in Sargentia 4: 50. 1943.

From the points of view of individual numbers and extensive (though interrupted) dispersal, the var. minor is by far the most important form of the Robbins milk-vetch. There seems little doubt that it represents the archetype of the species and that the remaining varieties are not much more than minor variants which have been able to maintain a certain individuality through accidents of physical isolation from the main gene-stream. In the Rocky Mountains var. minor (often passing under the names A. Macounii or A. occidentalis) is variable in several directions, notably in stature, pubescence, and length of the pod’s stipe and body. It is ordinarily an erect plant of streamsides and moist thickets in the timber belt, but it climbs occasionally to open crests in the tundra zone, there becoming dwarf and diffuse (cf. Weber & Barclay 9188, COLO, from central Colorado; Calder & Kukkonen 27,204, NY, from southeastern British Columbia; Breitung 16,435, NY, from southern Alberta). Commonly the leaves are so thinly pubescent that they appear glabrous to the casual glance, but the lower surface is sometimes cincereous when young and forms with genuinely glabrate foliage do occur (cf. Hitchcock & Muhlick 14,104, NY). The inflorescence and pod are, as a rule, black- or fuscous-hairy, but the vesture is sometimes largely or even entirely white. The variants mentioned so far are quite minor and of sporadic occurrence, and no correlation seems to exist between the length of the pod-body and a significant pattern of dispersal. It is otherwise with the length of the stipe which, usually quite short in Colorado, tends to become (in a series only rarely interrupted or irregular) longer as one traces var. minor northward along the Cordillera, reaching a length of 3.5—4.5 mm. or exceptionally 5-6.5 mm. (cf. Standley 15,692a, from Glacier Park, US) in Alberta and northwest Montana. Some individual plants from this area (cf. Macoun, from Kicking Horse Lake, in 1885, NY, significantly identified by Sheldon as "A. Jesupi" ) are technically inseparable from the plant of the New England mountains upon which var. Blakei is based. For many years I have tried to maintain a separate category for the Cordilleran A. Macounii (and have freely used an unpublished trinomial in annotations), but this course no longer seems logically defensible. At the time Eggleston was studying the eastern forms of the Robbins milk-vetch, he must have suspected that A. Blakei was conspecific with at least some of its western relatives, for he applied his manuscript var. borealis, as well as the binomial A. Blakei, in the herbarium to a few plants of both areas. The Cordilleran populations possess collectively a greater fund of variation than would be possible in the four or five restricted stations in New England and Nova Scotia. The eastern plants in the interior have relatively large flowers and relatively dense pubescence on the pods, but since these features can be matched in fine detail by individual Cordilleran plants, no taxonomic division is possible. The population on Cape d’Or in Nova Scotia matches some Rocky Mountain material very closely.

The var. minor is apparently unknown in the Rocky Mountains northward from about the latitude of Banff, but it reappears in the Yukon Valley just west of the Yukon-Alaska line. The Alaskan plant from this area was described by Rydberg as Atelophragma Collieri, supposed to differ from A. Macounii (our Cordilleran var. minor) in its larger flower and from the coastal Alaskan A. Harringtonii in its longer stipe and thin pubescence. I agree with Hultén (1943, p. 1088) that the plant of interior Alaska is essentially identical with A. Macounii. In the original description of A. Collieri Rydberg cited a peculiar plant, Anderson 1264 (NY) from Glacier Bay, that is from the Alaskan coast where var. Harringtonii is the common phase of A. Robbinsii. Evidently representing the same type are three other collections from the Bay, Coville & Funston 744 from Port Gustavus (US), W. S. Cooper 175 (US) and Anderson 1261 from Willoughby Island (US). These plants have been referred to var. minor with misgiving, for they are unusual in their long calyx-teeth (up to 2.9-3.3 mm. long), long stipes (4—5 mm.) and broad septum (1—1.5 mm. wide). The calyx-teeth and septum are suggestive of var. Harringtonii, which also occurs in typical form on Glacier Bay, but the pubescence is that of var. minor. Hultén (op. cit., p. 1087) listed Anderson 1261 as A. Harringtonii, but it does not fit comfortably into my concept of that entity. Possibly more material will show that another very local variety like those described below is isolated at this point on the coast. It should be noted that Hultén’s record (op. cit., p. 1088) of A. Macounii from the Behring Strait district in far western Alaska is based on a specimen (Palmer 658, NY) of A. alpinus.