Plants of Southern California: Rhamnus ilicifolia and R. pilosa
When I saw my first Rhamnus ilicifolia shrubs growing in the shade at Heller's Bend in northern San Diego County on 8 October 2002, I didn't even recognize them. These bushes were so different from the ones just 11 miles north at the Santa Rosa Plateau! The bushes at the Santa Rosa Plateau were tall, with erect to ascending stiff twigs, and large revolute leaves, and typically growing in the sun. In contrast, the bushes at Heller's Bend had mostly spreading branches, some of which were descending down a bank, that were weak and flexible. Further, they had much smaller leaves than the ones at the Santa Rosa Plateau.
These bushes were so different that I was initially quite confused about their identification. Here is the evolution of my identification of those bushes during about a half-mile stretch of trail:
- I first wrote down "R. crocea?", primarily due to the size of the leaf.
- I then crossed off "R. crocea?", and wrote down "unknown" after seeing a few more, since their habit was distinctly not like R. crocea.
- Finally, out of the heavily-shaded area I saw some relatively "normal" R. ilicifolia, and so wrote down "R. ilicifolia?" for the identification of the first bushes.
I did not know about the existence of R. pilosa until I got home and used the JM Rhamnus key to try to figure out these bushes, and hence was quite excited to check out the hairiness of the twigs to confirm the id. However, when i returned on 13 January 2003, I found that the twigs were glabrous, making them clear R. ilicifolia.
The implications of finding R. ilicifolia with flexible spreading twigs did not dawn on me until I saw plants with a similar habit at Blue Sky Ecological Reserve on 13 December 2003. There I found numerous plants with flexible spreading twigs, with some plants having densely hairy twigs and other plants having glabrous twigs, growing within a half mile of each other. Since this is the area in which R. pilosa is supposed to be found, with the type specimen from Ramona, just a few miles away, it was clear that further investigation was needed.
My analysis here is only preliminary; I reach no conclusions yet about whether R. pilosa is actually a separate species. Although I suspect R. pilosa is not a distinct species, I reserve judgment until I collect considerably more data.
The Discriminants Between These Species In The Floras
You know you've got trouble when one flora flatly contradicts the discrimination between the species used in another flora.
The Munz key:4. lvs glabrous to pubescent, rarely slightly revolute; widespread ... R. ilicifolia
4'. lvs pilose on both surfaces, revolute; San Diego County ..................... R. pilosa
His description says they are similar except R. pilosa has "young branchlets, lvs, floral parts densely pilose". He locates it only near 1500 feet in San Diego County.
The JM throws away all the Munz differences:
The JM key:5. branches stiff, twigs glabrous to densely hairy; CA-FP, DMtns ... R. ilicifolia
5'. branches flexible, twigs densely hairy; PR ... R. pilosa
The key also states in 3' that the leaf for R. ilicifolia and R. pilosa are both concave on their lower surfaces, which is equivalent to being revolute. The description for R. ilicifolia says both surfaces of the leaves can be hairy.
Hence the distinguishing characteristics in the Munz key have become characteristics in common between the two species in the JM. Thus according to the JM, any R. pilosa keyed out by the Munz key and description has not been keyed properly and cannot be trusted!
In my experience, R. ilicifolia leaves are indeed mostly revolute, often very strong so, so I can attest to the validity of that change in the JM.
Now let's look at the JM key. Specimens with glabrous twigs are always R. ilicifolia. Specimens with densely hairy twigs can be either species, with only a single discriminant: branches stiff or flexible.
The only other discriminants in the descriptions are:
R. ilicifolia: lf blade 20-40 mm; petals 0; fr 8 mm
R. pilosa: lf blade 15-20 mm; petals 0 or 4; fr 6 mm
In my judgment, this all smells somewhat fishy. Having this kind of contradiction between Munz and the JM is symptomatic of taxa that are at minimum not clearly distinguished, and possibly not separate.
The JM differences seem subtle at best, and possibly artificial. The key difference, whether the twigs are flexible or stiff, is a horrible criterion. There is no quantitative way to distinguish flexible twigs from stiff twigs, and I know from experience that different observers often classify the same twigs in different ways.
As for the leaf blade length, I'm always suspicious to see what looks like an arbitrary division in a continuum, giving one end of the range to one species and the other end to another species. However, this situation does happen infrequently, so further investigation here is needed. It is interesting that Stuart and Sawyer, in Trees and Shrubs of California, mention that R. ilicifolia can have leaves as short as 13 mm. If this is true, the difference in leaf size between the two species is not significant.
The fruit size difference is probably insignificant since floras often underestimate the range of variation in the population due to small number statistics.
As for the final difference, for the petals, if a significant population of R. pilosa in fact had petals, this would be an indication that it truly was a separate species. But if that were true, it probably would have been made a part of the key, to supplement the single criterion used there.
The only other information I could find are vouchers. SMASCH has many online vouchers for R. pilosa, nearly all at Ramona, vouchered from 1928 to 1936. But as pointed out above, these could all be R. ilicifolia, distinguished only by the Munz key before it was recognized that R. ilicifolia could be densely hairy for its twigs and leaves.
Hence I have begun to collect data on all R. ilicifolia and similar specimens to see if I can distinguish any difference for R. pilosa candidates.
Data From Santa Margarita Ecological Reserve On 18 December 2003
The Santa Margarita Ecological Reserve (SMER) has some specimens of R. ilicifolia in sunny open areas, as well as some in heavy shade. I collected data from individual plants, and plotted them below.
I kept track of the hairiness of the twigs and how spreading the twigs were, the size of the largest leaves, and how shaded each plant was. I only gathered data on the largest leaves, since a typical R. ilicifolia plant has a huge range in the size of its leaves. Typically, the youngest leaves are very small, and hence the size of the smallest leaves is meaningless.
I measured the flexibility of the twigs by pushing on the end of the twigs with my palm. If the twig did not bend, but instead the entire branch moved by deforming its lower length or along its entire length, I classified it as stiff, and gave it a numerical value of 1.0 in the plots below. If the twig immediately began to bend near its end, I classified it as flexible, and gave it a numerical value of 2.0. I recognized one intermediate value for twigs that bent significantly less than others, and gave it a numerical value of 1.5. Thus at least for these specimens, judging the flexibility was fairly straightforward.
The hairiness of the twigs fell into two distinct categories. Twigs were either fairly densely hairy, or were essentially glabrous. I gave glabrous twigs a numerical value of 1.0 for the plots below; hairy twigs received a numerical value of 2.0.
As seen from the plots below, the twigs ranged from glabrous to hairy, erect to spreading, and flexible to stiff. All of these plants would be classified as R. ilicifolia from the JM key, except for a single plant, which would be classified as R. pilosa from the key. Although it is interesting that this plant also has the smallest "maximum leaf size", I don't think anyone would call it R. pilosa, for three reasons. First, the leaves are nearly glabrous. Second, its largest leaves were 41 x 29, 37 x 26, and 25 x 22 mm, much larger than the 10-15 mm range given in the JM. Third, it is not very likely that there would be a single specimen of R. pilosa surrounded by numerous specimens of R. ilicifolia. It is probably more significant that this plant was in the heaviest shade of all of the specimens I examined.
I have plotted this specimen, which keyed to R. pilosa, in a separate color in the plots below. This does not imply that it is R. pilosa; it was done simply to easily show where this specimen falls in the context of the other specimens.
Another specimen had the identical flexible spreading branches as this R. pilosa candidate, but its branches were glabrous. This specimen was the third most-shaded plant, and had the second largest "maximum leaf size".
This analysis clearly shows that R. ilicifolia can have spreading flexible branches, and a continuum of properties that could easily encompass those plants that have been called R. pilosa.
I therefore am deeply suspicious about the existence of R. pilosa as a species separate from R. ilicifolia. However, I defer judgment until I collect considerably more data, especially from plants near Ramona, including data from the flowers.
Size of Largest Leaf on Individual Plants
This plot contains data from a single leaf per plant, the largest one I could find on each plant. The R. ilicifolia specimen that keyed to R. pilosa is plotted in pink; the other R. ilicifolia specimens are in blue.
Also shown for each species is the range of leaf blade length given in the JM, which is undoubtedly closer to the average leaf size for each species. Hence I would expect that the maximum leaf blade size would plot to larger values than the range in the JM. However, I would not expect that a plant whose typical leaf blades were 15-20 mm in length would have a maximum leaf blade length of 41 mm.
The following plot includes the second and third largest leaf for many of the specimens:
The R. ilicifolia specimen that keyed to R. pilosa does indeed have smaller maximum leaf sizes in general than the other plants, but I suspect the leaf size is related to the shadiness of its habitat. Also, it is clear that the three largest leaves on this plant are not terribly different from the three largest leaves on other specimens.
The plot above shows that another specimen shares the same flexible spreading stem properties as the R. ilicifolia specimen that keyed to R. pilosa. (The value for its stem flexibility was dithered slightly so that its plotted point could be seen clearly.) However, the plot below shows that specimen has glabrous twigs. The values for pure R. pilosa plants are indicated by the pink rectangle.
I intend to collect more data to definitively establish how separate these two species actually are, including, of course, data from plants that are consistent with being R. pilosa.
Copyright © 2003 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: 20 December 2003