Flora of North America species comparison

Sometimes Salix interior is treated as a subspecies of S. exigua (R. D. Dorn 1998). Salix exigua and S. interior hybridize and apparently intergrade in the western Great Plains; because the area of overlap is relatively small and distinctiveness of the two taxa is not compromised by hybridization and introgression, it is best to treat them as separate species.

Leaves on sylleptic shoots are usually very densely silky. Salix interior sometimes has shoots that arise from buds on either side of the normal axillary bud. They do not seem to be directly related to the stipules because they are enclosed by the petiole. Catkins with both staminate and pistillate flowers are rare in S. interior, but a Quebec specimen had some catkins predominantly pistillate and others staminate; most were a mixture. The flowers were not teratological, but a mature capsule contained aborted ovules.

Hybrids:

Salix interior forms natural hybrids with S. exigua var. exigua. Controlled pollinations using S. interior (as S. exigua) from southern Ontario (A. Mosseler 1990) successfully produced F1 hybrids with S. bebbiana, S. discolor, S. eriocephala, and S. petiolaris. Seed production was usually relatively low, except in crosses with S. discolor. In general, F1 viability was relatively low in crosses with these members of subg. Vetrix. No seeds were produced in crosses with members of subgenera Protitea or Salix. Morphology of the hybrids usually was intermediate between the two parents, but when S. petiolaris was used as the maternal parent, the F1s more closely resembled that species. J. Salick and E. Pfeffer (1999) extended these findings to show that, although crosses between S. interior (as S. exigua) and S. eriocephala are partially sterile, their clonal growth parameters (sprouting, shoot length, and biomass production) are strong and thus permit these partially sterile hybrids to exist as successful individuals and perhaps to “... make a contribution to interspecific gene flow over time.” Of particular taxonomic interest is that, in this cross, the staminate parent has a significant influence on leaf shape, whereas in the cross S. eriocephala × S. petiolaris it is the pistillate parent that is significant for leaf shape. Relatively few hybrids resembling those produced by Mosseler have been recognized in nature, but it is possible that the unusually broadly leaved plants named S. interior var. exterior and var. wheeleri, from northern Maine, Nebraska, New York, and West Virginia, and probably elsewhere, may be hybrids. Phenological isolation may be strong enough to prevent crosses in nature (A. Mosseler and C. S. Papadopol 1989) with the earlier flowering S. eriocephala and S. petiolaris, a barrier that even an occasional period of overlap cannot breach.

(Discussion copyrighted by Flora of North America; reprinted with permission.)

Varieties 2 (2 in the flora).

Salix rotundifolia is closely related to S. polaris, from which it can be separated by its glabrous ovaries and fewer-flowered catkins. They also differ somewhat in leaf venation: S. rotundifolia typically having three main veins arising from the leaf base, often only one or two pair of secondary veins, and no or indistinct tertiary veins; S. polaris typically having pinnate venation, multiple secondary veins, and distinct tertiary veins. Salix rotundifolia consists of two varieties, the diploid var. dodgeana and the hexaploid var. rotundifolia. In general, var. dodgeana is a high alpine species in the southern cordillera of Wyoming and Montana, the St. Elias Mountains in Alaska and Yukon, the Mackenzie Mountains, Northwest Territories, and the Richardson Mountains, Yukon Territory. A diploid specimen of S. rotundifolia in the Cherski Mountains, Yakutia, Russia (B. A. Jurtzev and P. G. Zhukova 1982), which fits var. dodgeana in its 2–3-flowered catkins, relatively small leaves (3.5 × 3.9 mm), and small stomata (490 µm2), may represent an ancestral population. Variety rotundifolia usually occurs at lower elevations in Alaska and in easternmost Chukotka and Wrangel Island, Russia, but elevation separation is not distinct. There is a general correlation between stomatal size and ploidal level (W. Buechler, pers. comm.), but relatively large stomata in some diploid specimens of S. rotundifolia indicates a need for further cytological study. For the present, it is best to recognize the two cytotypes as varieties.

Hybrids:

Salix rotundifolia forms natural hybrids with S. arctica, S. phlebophylla, and S. polaris.

(Discussion copyrighted by Flora of North America; reprinted with permission.)