Slenderleaf saltbush

Atriplex canescens (Pursh) Nuttall subsp. linearis (S. Watson) H. M. Hall & Clements; A. canescens var. linearis (S. Watson) Munz

Shrubs dioecious, erect, mainly 10-25 dm; branchlets slender, terete. Leaves sessile; blade narrowly linear-elliptic, 10-50 × 2-3 mm, firm, revolute, often acute apically. Staminate flowers in glomerules borne in slender interrupted mostly paniculate spikes. Pistillate flowers paniculate or in few-flowered axillary glomerules. Fruiting bracteoles sessile or subsessile, lanceolate to ovate, 4-6 mm, about as wide, each bract with a pair of thin wings 3 mm broad or less, irregularly dentate or laciniate, free tips of bracts much exceeding the wings. 2n = 18.

Flowering spring-fall. Saline deserts, with shadscale, Canotia, Yucca, Opuntia, Rhus, and Eriogonum; 0-800 m; Ariz., Calif.; nw Mexico (Baja California, Sonora).

Specimens of Atriplex canescens var. macilenta resemble A. linearis. The taxa have been placed together by some previous workers. Nevertheless, the stems of A. linearis are consistently more slender, the leaves proportionally narrower, and the bracts, though smaller, more closely simulate those of A. canescens. Its diploid nature signals a different evolutionary pathway than that for most of A. canescens, considered broadly. Narrow leaves occur within A. canescens, in the broad sense, sometimes with geographic correlation, sometimes not.

Excluded species:

Atriplex barclayana subsp. dilatata H. M. Hall & Clements, Publ. Carnegie Inst. Wash. 326: 315. 1923

A new species, most related to Atriplex argentea and A. expansa, judging from the habit; but the fruits are flat and their sides not appendaged or muricate." The plant was described from San Benito Island, Baja California.

Atriplex nitens Schkuhr, Bot. Handb. 3: 541. 1803

P. C. Standley (1916) reported this species in New York, New Jersey, South Dakota, and Oregon.

P. Aellen (1964) wrote: "Like 4 [i.e., A. hortensis] but the leaves white beneath; bracteoles oblong-cordate, conspicuously reticulate-veined." Specimens having at least some of the above noted characteristics occur here and there among the North American materials of A. hortensis, and the name nitens is here in considered in synonymy of that species. It belongs to sect. Atriplex.

This name was apparently not taken up, even in synonymy, by either L. Abrams and R. S. Ferris (1923-1960, vol. 2) or P. A. Munz (1959), but was cited as a synonym of Atriplex argentea by P. C. Standley (1916). H. M. Hall and F. E. Clements (1923) cited this as a synonym of "A. argentea expansa," based on "An insect stung monstrosity, according to Jepson (Fl. Calif. 436, 1914)."

Atriplex nodosa Greene, Pittonia 1: 40. 1887

Atriplex sibirica Linnaeus, Sp. Pl. ed. 2, 2: 1493. 1763

This was supposedly collected in northeastern United States, but no specimens have been seen from the flora area. It belongs to sect. Obione.

Atriplex vesicaria Heward ex Bentham, Fl. Austral. 5: 172. 1870

The section Dialysex, to which Atriplex vesicaria belongs, is treated in the Australian flora with sixteen species; A. vesicaria with eight subspecies; its records in the United States, if any, are not known to me. It would not, in any case, be treated with the herbaceous species. H. M. Hall and F. E. Clements (1923) noted that this species, and the related A. holocarpa, have both been grown in American gardens with the thought of using both of them as forage plants, but neither of them found suitable for general planting.