Plant Species of San Jacinto Mountain:
Keckiella rothrockii var. jacintensis, San Jacinto Mts. keckiella

Fig. 1. Photographs of a large plant of Keckiella rothrockii var. jacintensis in full bloom, taken on 11 July 2011 by Tom Chester, of a plant just south of Skunk Cabbage Meadow. Note the lower legs of the person for scale in the upper left picture, and fingers for scale in the bottom pictures. The plant width was about four feet (1.2 m), and its height was about one foot (30 cm). The flowers are about a half inch (15 mm) long and a quarter inch (7 mm) wide.
Click on the pictures for larger versions.

Introduction and Summary as of 17 September 2016
Taxonomic History and Comparison to var. rothrockii
Detailed Census Results of each survey (on a separate webpage)

Introduction and Summary as of 17 September 2016

Keckiella rothrockii var. jacintensis, San Jacinto Mts. keckiella, is endemic to the high elevations of San Jacinto Mountain. Its name honors the botanist David D. Keck, who studied this and related species; Joseph Trimble Rothrock, the first collector of the species (of var. rothrockii); and San Jacinto Mountain.

Plants now in the genus Keckiella were all in the genus Penstemon until 1967 when Richard M. Straw split them out, giving them the Keckiella name. They were originally named Keckia in 1966, but that name turned out to be illegitimate, so in 1967 they were renamed Keckiella. DNA studies have confirmed that Keckiella is distinct in its phylogeny from Penstemon, and that K. rothrockii is the most basal Keckiella species (i.e., the closest of all the Keckiella species to the Penstemon clade).

As shown in Fig. 1, this taxon is a sprawling low shrub which can reach about four feet wide, with opposite nearly-glabrous green leaves and flowering stems about one foot high. The sprawling stems often root when they are in contact in the ground. Many people could easily pass a specimen in full flower and not notice it was in bloom since the flowers are small and an undescript yellow color.

This taxon primarily lives in four habitats, all of which are fairly infrequent at San Jacinto Mountain:

See Fig. 2 for a preliminary incomplete map of all GPS positions we had for Keckiella rothrockii var. jacintensis as of 4 September 2016, with the western escarpment and the valleys labeled, and Fig. 3 for a Google Earth view of those positions plus those from later surveys. The map is incomplete since not all of our GPS records have been digitized so far, and we had done few targeted surveys for this taxon at that time. As of that date, we had 32 digitized GPS points from 17 separate full-day surveys for all species, and 175 digitized GPS points from surveys dedicated to GPS'ing Keckiella. Those GPS points range from 7500 to 9845 feet elevation, which as far as we know is the full range of elevation for this species.

Very few plants of this taxon are found outside of these habitats. Some of the few such plants are one small populuation found in the west-facing uppermost Powder Spring Drainage along the Devils Slide Trail just west of Saddle Junction, and one small population found on a northwest-facing slope in southeastern Tahquitz Valley. Those locations are plotted in Fig. 2.

Fig. 2. A preliminary incomplete map of all our digitized GPS positions (blue diamonds) for Keckiella rothrockii var. jacintensis as of 4 September 2016, with the western escarpment and the valleys labeled, and north up. The GPS points in Tamarack, Round and Long Valleys are primarily from a single GPS point per two to four mile survey, whereas the points along Fuller Ridge and south of Saddle Junction come from a census of the population. There will be many more GPS locations in Tamarack, Round and Long Valleys after a census is done there.

A map of vouchers from the Consortium of California Herbaria shows a number of vouchers plotted at locations outside the areas shown in Fig. 2, but there are no actual Keckiella plants at those locations (see voucher map with erroneous locations marked with red circles). Some of those vouchers circled in red had to be georeferenced with a guess as to where the voucher was taken, since the voucher localities were extremely vague, such as San Jacinto Mts, from vouchers taken in 1880 and 1895, or somewhat vague, such as Tahquitz Valley for vouchers taken in 1901 and 1923. A few of the vouchers were georeferenced incorrectly, such as one from Long Valley, near Cornell Peak georeferenced far to the north of Cornell Peak and Long Valley.

All known locations of Keckiella rothrockii var. jacintensis are within a distance of 6.6 miles from northwest to southeast, and 3.0 miles from southwest to northeast. This taxon occupies only a small part of the rectangle with those dimensions. A rough estimate is that the vast majority of the population lives in about five square miles (12 square km), with a very small number of plants found outside that area.

A Google Earth "airplane view" showing all the GPS locations as of 17 September 2016 is shown in Fig. 3. This map includes the results of three censuses not shown in Fig. 2.

Fig. 3. A Google Earth "airplane view", looking to the west from a vantage point high above San Jacinto Mountain, showing all our GPS locations as of 17 September 2016. Three surveys were done along and just below the western escarpment, with faint blue lines showing the path of the surveys. Survey GPS locations with over 50 plants are plotted as larger pink balls. In the survey areas, the small pink balls represent survey GPS locations with fewer than 50 plants. Other small pink balls are GPS points with no information given in this map about the number of plants at those locations.
Click on the map for a larger version.

Interestingly, this taxon is a rare plant that is not listed as a rare plant by anyone! For example, Calochortus palmeri var. munzii is almost surely more abundant in the number of plants compared to Keckiella rothrockii var. jacintensis, yet it has a CNPS rare plant rank of 1B.2, Rare, threatened, or endangered in California. In fact, every other endemic species of San Jacinto Mountain has that same rare plant rank of 1B; only Keckiella rothrockii var. jacintensis is not considered rare.

We can only speculate on why this rare plant taxon is not listed as a rare plant. Perhaps it simply has not had a champion, someone to bring it to the attention of those that compile rare plant lists. It is in a genus that has no other listed rare plants, and it doesn't have showy flowers like those of Calochortus. Perhaps no one realized what a limited distribution this taxon had, because it is fairly abundant along several major hiking trails at San Jacinto Mountain. Perhaps it is because Hall wrote in his 1902 Flora of the Pine Belt of San Jacinto Mountain that this taxon was quite frequent in the open forests from Tahquitz Valley to Fullers Ridge.

Tom and Bruce got interested in censusing the population of this species in late 2016, to see how abundant it actually was. This census is a work in progress that probably won't be finished until sometime in 2017. As of 17 September 2016, we have censused the population in three areas, finding a total of 7,118 plants:

See a separate webpage giving Detailed Census Results of each survey.

It seems likely that the final count for the total number of plants of this taxon will be somewhere between 10,000 and 20,000 plants. For comparison, Bruce counted 38,002 plants of Calochortus palmeri var. munzii in Garner Valley in spring 2016.

Taxonomic History and Comparison to var. rothrockii

Penstemon rothrockii was first described by Asa Gray in his Synoptical Flora of North America published in 1878, from a specimen collected by Joseph Trimble Rothrock in 1875 in Tulare County on Little Olanche Mountain at an elevation of 10,400 feet.

The first voucher from the San Jacinto Mountains was taken by Anstruther Davidson in 1876, and was initially taken to be the same species as the 1875 voucher. Hall called it P. rothrockii in his 1902 Flora of the Pine Belt of San Jacinto Mountain.

Le Roy Abrams, in 1906, defined our plants as a new species, Penstemon jacintensis, differentiating from P. rothrockii on its conspicuously though sparsely pubescent corolla, which he wrote was dull yellow veined with purple and 15 mm long, compared to the glabrous reddish 9 mm corolla of P. rothrockii. Other differences stated were very sparsely and minutely puberulent leaves and branches, compared to minutely and rather densly hispidulous leaves and branches; and lanceolate 6 mm sepals, compared to ovate-lanceolate 3 to 4 mm long sepals. But see below; the corolla color difference, and the sepal shape, apparently was later found not to be significantly different between the species. The type specimen was Hall's number 704, from Tamarack Valley, San Jacinto Mountain at 9000 feet elevation.

In 1924, Philip A. Munz and Ivan M. Johnston, in their monograph on The Penstemons of Southern California, made P. jacintensis a variety of P. rothrockii without discussion.

In 1936, David D. Keck, in his Studies in Penstemon, changed it from a variety to a subspecies, saying that P. rothrockii has two natural geographic subspecies. He commented that ssp. jacintensis was justified by the 1 to 3 mm increase in corolla length, 13-16 mm long compared to 11 to 13 mm long. He also mentioned that the shape of the corolla was often different as well, often ampliate at the throat so as gradually to arcuate the upper side and increase the width to 5 mm. Subspecies typica has straight corollas 11 to 13 mm long, which are not noticeably ampliate nor much over 4 mm in breadth. He stated that ssp. jacintensis was not justified as being taken as a separate species because the other differences, such as differences in flower color and sepal shape, appear to have no foundation. Keck also cited one collection of P. rothrockii near its type locality in Tulare County that has corollas up to 14 mm long, and leaves with hairs like those of ssp. jacintensis.

In 1967, Richard M. Straw transferred Penstemon rothrockii to the genus Keckiella, named in honor of David D. Keck, and kept jacintensis as a subspecies of Keckiella rothrockii.

In 1992, N.H. Holmgren changed it to a variety, which is where it stands today in the latest floras.

A map of the complete distribution of the two taxa is shown in Fig. 4. Variety jacintensis is disjunct from the nearest population of var. rothrockii by 150 miles.

Fig. 4. Map of the complete voucher distribution for the two varieties of Keckiella rothrockii, from SEINet and the Consortium of California Herbaria on 18 September 2016. K. rothrockii lives only in California and a small strip of adjacent Nevada.

The Jepson eflora key to distinguish these varieties is:

7. Corolla 1315 mm, sparsely long-hairy; leaves green, becoming glabrous ..... var. jacintensis
7' Corolla 1012 mm, becoming glabrous; leaves canescent ..... var. rothrockii

Fig. 5 shows photographs of the two varieties side by side.

One difference apparent in most of the Calphotos pictures of the two varieties is that the inflorescence is much more robust in var. rothrockii, which often has two flowers per node, with larger bracts (leaves in the inflorescence). In comparison, the inflorescence of var. jacintensis usually has just a single flower per node, and the bracts are much smaller. Since no flora has ever commented on this difference, it might be environmentally caused, due to the general higher rainfall amounts in the Sierra Nevada where most of the var. rothrockii lives. On the other hand, this might be at least a general difference between the two varieties as well.

Fig. 5. Photographs of Keckiella rothrockii taken by Keir Morse. Left: var. jacintensis, from San Jacinto Mountains. Right: var. rothrockii, from Piute Peak in Kern County.
The leaves of the two varieties are distinctly different in color and hairiness.
The pictures of the corollas have been scaled such that the displayed size of the corolla of var. jacintensis is 14/11 times as long as that of var. rothrockii, using the average corolla length of 14 mm for var. jacintensis and 11 mm for var. rothrockii, and hence have been made to display 1.27 times larger.
The curvature of the upper lip of the corolla is noticeably more in var. jacintensis, and in these pix its corolla is a deeper yellow. Its lower lip either is without the darker-color veins of var. rothrockii, as shown here, or with fainter veins (see Fig. 1).
The back of the upper lip is a bit hairier in the corolla for var. jacintensis, but is not strikingly different in these pictures, which were not optimized to show those hairs. See Fig. 1 for a better view of the hairs for var. jacintensis.
Click on the pictures for larger versions.

Voucher data provided by the participants of the Consortium of California Herbaria (

Go to:

Copyright © 2016 by Tom Chester and Dave Stith.
Commercial rights reserved. Permission is granted to reproduce any or all of this page for individual or non-profit institutional internal use as long as credit is given to us at this source:
Comments and feedback: Tom Chester
Updated 19 September 2016.