Plants of Southern California: Lycium species of the San Diego County Desert

Lycium species are difficult to identify in some cases for at least four different reasons:

  1. The key to determine specimens uses both flowers and fruit, and often all one has are flowers.

  2. One major couplet in the 1932 Hitchcock key used in Munz and the Jepson Manual, using the ratio of the calyx lobes to the calyx tube and the calyx lobe length, is often indeterminate, with some specimens satisfying both branches.

  3. The key uses the hairiness of the calyx and leaves, and some specimens do not appear to have read the key and might be easily keyed under the "wrong" hairiness.

  4. The measurements given in the floras of the leaf lengths and widths, and the calyx tube and lobes, often do not give the full range for all species.

  5. The flowers quickly fade and shrink in voucher specimens, making it difficult to determine colors and corolla lobe lengths.

Fortunately, we have just five Lycium species (see note below) in the desert portion of San Diego County (according to the Checklist of the Vascular Plants of San Diego County, Jon P. Rebman and Michaal G. Simpson, 4th edition, 2006), and they each seem to have one or more unique characteristics that immediately determine them, except for two species that require the combination of two characteristics to distinguish them.

The following table gives the unique characteristics, along with a picture showing those characteristics. Most of the botanical terms are illustrated in the picture, and all are explicitly defined below the table. The non-plant parts in some of the pictures are human fingers.

Caveat: I have limited experience with seeing Lycium flowers, and have seen only four of these five species in bloom. Therefore I am relying fairly heavily on keys, floral descriptions, and online pictures in the following. I plan on accumulating measurements on all these species in the future to see how well these "unique characteristics" work. If any reader comes across flowers that appear to contradict the following, please let me know. (Note added 1/24/12: several of the ratios given below have been extended from what I originally deduced from the floras, using measurements from plants in the field. But the numbers are still based on a small number of samples, and so the ranges might be extended farther in the future.)

SpeciesUnique CharacteristicsPicture
L. andersoniiCorolla tube = (4-7) x calyx tube and calyx glabrous (without hairs). All other species except L. fremontii and L. parishii have a ratio of less than 4. Those two species have ratios of 2-5, but they have glandular-pubescent calyces.

In the picture on the immediate right, on my screen, I measure the corolla tube as 42 mm and the calyx tube as 8 mm, for a ratio of 5.2. You should get the same ratio, even if your screen size is different.

L. brevipes var. brevipesCorolla lobes = (0.33 - 1) x corolla tube; the other species usually have a ratio of 0.15-0.33.
# of calyx lobes and corolla lobes 4-5; the other species seem to almost always have 5 calyx and corolla lobes (However, I've now observed that L. fremontii also sometimes has 4 lobes, so this may not be a good discriminant).

In the picture on the immediate right, I measure the corolla lobes to be 20 mm and the corolla tube to be 40 mm, for a ratio of 0.50.

L. cooperiCorolla greenish-white; the other species are a pure white to pink to purple

Fruit greenish-white, constricted at top; the other species have red fruit without any constrictions

L. fremontiiPlant densely glandular pubescent AND calyx lobes short, not appearing leaf-like. The other species are without hairs, to moderately hairy, usually without glands, except for L. parishii. L. parishii has leafy calyx lobes that are approximately as long as the calyx tube.

The dense stalked glands are visible as the "little balls of goo" on tiny stalks, and often glistening in the light. The white lines point to only a small number of the glands. The glands are sticky, and often have attached objects, such as the dandelion fluff in this picture. In the other picture, the glands are visible mostly as white dots.
The ratio of the corolla tube to the calyx tube is 1.9-2.4 in the pictures, but I have observed flowers with a ratio of 5.4 in Fossil Canyon in Imperial County.

Picture copyright © 2004 by Michelle Cloud-Hughes

Picture copyright © 2008 by John Marquis
L. parishiiPlant slightly- to densely-glandular pubescent AND calyx lobes leafy, about as long as the calyx tube. The other species are without hairs, to moderately hairy, usually without glands, except for L. fremontii. L. fremontii has short calyx lobes that do not appear leaf-like.
The ratio of the corolla tube to the calyx tube is 2.9-3.5 in the pictures here; this picture has a ratio of 4.4 and shows the glands more clearly.

Picture copyright © by J.S. Miller

Picture copyright © by J.S. Miller

Definitions of the botanical terms used above:

The corolla is what most people refer to as the flower, the showy white to purple part. The corolla consists of a tubular lower portion, the corolla tube and the 4-5 lobes at the tip. Botanists consider the lobes to be the free (not connected) portion of the petals, which are fused into the tube below the lobes.

The calyx consists of the green bracts at the base of the corolla, surrounding it there. It also consists of a tube and lobes.

The fruit is like a little tomato, a berry that is either green or red.

Pubescent refers to hairs that can only be seen by a magnifying glass, although often when botanists refer to a plant having hairs it can mean either hairs visible to the eye or smaller ones. Glands are hairs with balls of goo at their tips.


Distribution information is given in the following table. The species name is linked to distribution maps from the Consortium of California Herbaria, using data provided by the participants of the Consortium of California Herbaria (

Species linked to distribution map# Vouchers
Borrego Desert / San Diego County
Representative San Diego County Desert Distribution
L. andersonii3 / 73Borrego Valley; San Felipe Valley; Vallecito; Jacumba; Mountain Springs
L. brevipes4 / 13Borrego Valley; Clark Valley; Borrego Badlands; northeast corner of County
L. cooperi0 / 33San Felipe Valley; Box Canyon
L. fremontii1 / 22Borrego Valley; Split Mountain; Canebrake Canyon; Fossil Canyon
L. parishii0 / 8Mountain Palm Springs; Vallecito; Fossil Canyon


Note on Lycium torreyi

There are four online vouchers of L. torreyi in the San Diego County desert, but it is not accepted as being in the County by Rebman and Simpson. L. torreyi is very similar to L. andersonii, and those vouchers may be misdetermined L. andersonii.

L. torreyi is supposed to have woolly-ciliate corolla lobe margins, whereas L. andersonii has glabrous to finely straight-ciliate margins. There is also a large difference in the ratio of the corolla lobes to corolla tube. On 4/6/11 I measured the ratio of a number of flowers from plants in the Indian Canyons west of Palm Springs. All were consistent only with L. andersonii, despite some having ciliate corolla lobe margins.

There is also supposed to be a difference in the ratio of the calyx lobes to calyx tube, but this measurement is fraught with peril because exactly what is lobe and what is tube depends on the orientation of the calyx. In many Lycium flowers, there is one deep division in most calyces, with all other divisions shorter, so you'd get very different values for the lobes and tube if that deep division was used.

Although I found no plants that could be considered L. torreyi from the Indian Canyons west of Palm Springs, there are two vouchers of L. torreyi from there, and six vouchers of L. andersonii. Hitchcock, who defined the species of L. torreyi, stated

Among the North American species, it would appear that L. Andersoni and L. Torreyi may occasionally hybridize where the ranges of the two species overlap-in the Coachella Valley of southern California and in Arizona and southern Utah. Through-out its range L. Torreyi has white-fringed corolla-lobes, and a more thick, but proportionately shorter, corolla-tube than L. Andersonii, the latter species having but sparsely ciliate corolla-lobes. However, where the two species occur together, forms have been collected having the slender corolla-tube and leaves of L. Andersonii but with the lobes of the corolla being lanate as in L. Torreyi, the habit of the plant being more like that of L. Torreyi. Similarly, at least one instance has been noted where L. Richii [TJC: =L. brevipes] has apparently hybridized with L. Torreyi; this plant was collected in the Coachella Valley, the only region where the two species overlap in range.

It may be that Hitchcock's "L. torreyi" from the Coachella Valley may be due to a looser interpretation of how woolly-ciliate the corolla lobe margins need to be to qualify a specimen as L. torreyi. It would be good to re-examine those vouchers using the current determination of woolly vs. straight-ciliate. Many "hybrids" of the past have gone away once the full range of variation within a species has been recognized.

Go to:

Copyright © 2008-2012 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: 24 January 2012