Plants of Southern California: Atriplex canescens Varieties

Introduction and Summary
Geographic Distribution of Each Variety

var. laciniatavar. canescens

Fig. 1. The inflorescence (top) and fruit (bottom) of Atriplex canescens var. laciniata (left; from Ella Wash, Anza-Borrego Desert State Park, 12/4/2008) and var. canescens (right; top photo from Upper Coyote Canyon, Anza-Borrego Desert State Park, 10/29/2009; bottom photo from Cottonwood Trailhead near Banning Pass, 1/8/2015).


The first edition of the Jepson Manual (1996) gave two subspecies of Atriplex canescens: var. canescens and var. linearis. Var. macilenta was synonymized with ssp. canescens, and var. laciniata was said to be a form of both of these subspecies. Both of these latter two varieties have been resurrected in the second edition of the Jepson Manual (2012; henceforth JM2)

The JM2 key to distinguish these varieties is:

4. Twigs generally slender; shrub 1025 dm; leaves 1035 mm, 13 mm wide; fruit bracts 48 mm, 46 mm wide ..... var. linearis
4' Twigs generally not slender; shrub 320 dm; leaves 1050 mm, 28 mm wide; fruit bracts 425 mm, 412 mm wide
5. Fruit bracts entire to dentate; widespread ..... var. canescens
5' Fruit bracts irregularly to deeply sharp-dentate; s California
6. Plant generally 1020 dm; fruit bract 612 mm, deeply sharp-dentate ..... var. laciniata
6' Plant generally 315 dm; fruit bract 48 mm, irregularly dentate ..... var. macilenta

The bract length and width is for the entire length and width of the fruit. The JM1 used the width of the wing, which is half the bract width.

The bract of var. linearis have dentate margins, and so are presumably similar to the bract of var. canescens. However, this picture shows some sharp lobes that are fairly deep on a plant that has the linear leaves of var. linearis.

Fig. 1 shows the bracts of var. canescens and var. laciniata.

The bracts of var. macilenta are shown in this picture, from this page that discusses the varieties in some detail.

The Flora of North America, which treats var. linearis as the species of A. linearis, gives some more insight into two of these varieties:

var. macilenta, a high polyploid (x = 9), apparently arose through initial hybridization involving Atriplex canescens and one chromosomal race of A. polycarpa. It shares a similar morphology with the diploid A. linearis, with which it has been confused. That entity is present in the south California vicinity of both varieties macilenta and laciniata. It [A. linearis] differs in its very slender branchlets, less laciniate wings, and more elongate, elliptic leaves.

var. laciniata, also a high polyploid, is supposed to have originated through hybridization of Atriplex canescens with a second chromosomal race of A. polycarpa. It is certainly similar to but has larger and more apparently laciniate bract wings than A. canescens var. macilenta, which arose similarly but purportedly from a different chromosomal race of the same species.

Geographic Distribution of Each Species

Fig. 2 shows the geographic distribution of each of these species.

var. canescens


var. laciniata


var. linearis


var. macilenta

Fig. 2. The geographic distribution from vouchers of the four varieties. Note that some vouchers determined as var. canescens may actually be var. laciniata or macilenta, since those latter two varieties were treated as part of var. canescens in the JM1. Click on the maps for larger versions.

Go to:

Copyright © 2015 by Tom Chester
Permission is freely granted to reproduce any or all of this page as long as credit is given to me at this source:
Comments and feedback: Tom Chester
Last update: 11 January 2015