Astragalus alpinus var. brunetianus

Astragalus labradoricus fma. albirtus (albino, a white-flowered form) Rouss. in Contrib. Lab. Bot. Univ. Montreal 24: 26. 1933.—"Berthier-en-bas, comte Montmagny, 13 août 1925 (Jacques Rousseau, 20849). Type dans l’herbier de l’Universite de Montreal)."—Holotypus not seen; described as a white-petaled form.

The var. Brunetianus is extremely close to the form of var. alpinus prevalent in the southern Rocky Mountains, and it is almost impossible in the absence of labels to distinguish some flowering specimens from Vermont and Colorado. Fernald, in the original description of the variety, and Robinson & Fernald (1908, p. 516) included the Rocky Mountains in the range of var. Brunetianus. The small calyx of the eastern variety can be matched exactly in many Colorado specimens of var. alpinus, and while the range of petal-length is not exactly the same in the two areas, for the flowers of var. Brunetianus are on the average a little longer, there is a wide overlap in all measurements. Thus it seems that the length of the hairs on the pod is the one reliable differential character distinguishing the two. Up to 0.4 mm. long and almost always appressed in var. Brunetianus, the hairs are at least slightly longer and nearly always loosely ascending in the southern Cordilleran populations of var. alpinus with similarly small calyx. Plants of var. Brunetianus with spreading pod-hairs up to 0.4 mm. long have been collected on Murray Bay in the St. Lawrence estuary, and a fruit detached from one of these would pass anywhere for genuine var. alpinus. It is impossible therefore to admit A. labradoricus to specific rank.

In his Revision of Astragalus in Quebec, Rousseau (1933, passim) separated the riparian and estuarine forms of A. alpinus as distinct species, A. labradoricus of Newfoundland, Lake St. John, and the St. Lawrence estuary, and A. Brunetianus confined to Gaspe, adjoining New Brunswick, and northern Maine. The differential characters stressed in Rousseau’s key are, for the first, wing-petals relatively broad (2-2.5 mm.) and pod straight, oblanceolate, and inflated; and for the latter, narrower wings (1.5 mm.) and an incurved, somewhat flattened pod. It has been found, however, that the wings vary from 1.5 to 2.5 mm. wide in plants from the lower St. Lawrence and reach a width of at least 2.2 mm. in the area of A. Brunetianus. In the centennial edition of Gray’s Manual, Fernald (1950, p. 910) maintained the two entities as varieties to A. alpinus, and reduced the supposed differences to outline and curvature of the fruits. These differences are extremely tenuous. It is true that in Quebec the ordinarily collected pod is straight or nearly so and broadest above the middle (cylindro-oblanceolate); in New England it varies from lunately incurved to nearly straight and tends to be broadest near the base (lance-cylindric). Along the Connecticut River the pod is often nearly straight and exactly linear-oblong in profile, with the sides precisely parallel. Variation of this sort must be expected from one population to another when these occupy, as is the case with var. Brunetianus, mutually isolated ecological niches. Moreover the variation is no greater than that found in some single populations of var. alpinus in the Rocky Mountains. I must add that the pod of Michaux’s A. secundus preserved in the Torrey herbarium (NY) is distinctly curved.

Along the rivers of New England var. Brunetianus is found on rock ledges and stony banks just above the normal high-water line, and sometimes on gravel bars inundated during flood. It is associated in northern Maine and New Brunswick with A. eucosmus, and along the Connecticut River with A. Robbinsii var. Jesupi, both species of boreal and Cordilleran affinity. The remarkable ecology of var. Brunetianus in the intertidal zone of the St. Lawrence estuary has been described at length by Rousseau (1933, p. 26, sequ.).