Speckled Clarkia ID Guide
There are five speckled Clarkia species with cup shaped petals that are commonly misidentified for one another: C. bottae, C. dudleyana, C. cylindrica, C. lewisii, and C. rostrata. This is especially common in the LA area, where three of these species overlap. This guide aims to outline the key characteristics that can be used to tell these five species apart.
A sixth species C. jolonensis is also very similar in appearance to the species listed here. It shares all of the same characteristics listed here with C. bottae and overlaps in range with C. lewisii in Monterey county. It can be differentiated from C. cylindrica and C. lewisii using the same characteristics as C. bottae (inflorescence shape), but it can be only be consistently differentiated from C. bottae based on seed color (dark gray vs brown or black)1. I haven’t personally worked much with this species, so I won’t be including it in this guide.
Key
Key for the five Clarkia species that are being compared in this guide. This key will not necessarily help to differentiate these species from the other 37 species in the genus.
Species Ranges
C. lewisii and C. jolonensis overlap in Monterey county, but are easily differentiated based on inflorescence shape (see below). There are some reports that C. cylindrica and C. bottae can also be found in Monterey county, but this is outside of the range that has been formally defined for these species1,3. It is unclear to me whether the ranges for C. cylindrica and C. bottae have expanded northward or whether these reports are misidentified C. lewisii and C. jolonensis, given their morphological similarity to C. cylindrica and C. bottae, respectively.
C. dudleyana overlaps C. rostrata, C. cylindrica, and C. bottae at different portions of it’s range, but it can be easily differentiated based on the prominent ribbing of it’s ovary and mature fruit. Although the range of C. dudleyana looks mistakenly split, there are two disjunct regions where this species is found, one in the Sierra Nevadas and one in the mountains above LA, though there are likely some isolated populations connecting through Kern county.
C. cylindrica ssp. cylindrica and C. bottae overlap for much of their range but can be easily differentiated based on inflorescence shape. C. cylindrica ssp. clavicarpa could possibly overlap with C. rostrata in the northern part of it’s range, but when C. rostrata was defined as a separate species from C. cylindrica, all of the northern populations tested were not interfertile with the southern populations, thus leading to it being defined as a new species.
Taxonomic History
Originally only four species were recognized by the Lewis’, who formally described the genus in their 1955 monograph on Clarkia: C. bottae, C. lewisii, C. cylindrica, and C. dudleyana4. Also, quite unfortunately, in the 1970s there was a species name change; what was originally described as C. deflexa was renamed to C. bottae, and what was C. bottae got renamed to C. lewisii. So all older literature that mentions C. bottae is what we currently call C. lewisii.
Subsequently, populations of C. bottae in Monterey County were found to be reproductively isolated from the southern populations, so these Monterey County populations were defined as a new species, C. jolonensis1. In addition, populations originally attributed to C. cylindrica in the northern Sierra Nevadas were found to be a new species as well, C. rostrata, and C. cylindrica was split into two subspecies: populations from the southern Sierra Nevadas became C. cylindrica ssp. clavicarpa while populations from the coastal mountains became C. cylindrica ssp. cylindrica3.
County Distributions
C. jolonensis: Monterey
C. bottae: San Louis Obispo, Santa Barbara, Ventura, Los Angeles, Orange, Riverside
C. dudleyana: Tuolumne, Mariposa, Madera, Fresno, Tulare, Los Angeles, San Bernardino, Riverside
C. lewisii: Monterey, San Benito
C. cylindrica ssp. cylindrica: Fresno, San Louis Obispo, Kern, Santa Barbara, Ventura, Los Angeles
C. cylindrica ssp. clavicarpa: Fresno, Tulare, Kern
C. rostrata: Mariposa, Tuolumne, Stanislaus, Merced
*Note: Individuals of each species are likely found outside of these counties, but these ranges reflect where the bulk of the populations for these species will be found. Due to the ease with which these species are misidentified, it is always unclear whether reports outside of the normal range are true sightings or not.
Inflorescence
The easiest and quickest way to differentiate C. bottae from the other four speckled species is using the shape of the inflorescence when it still has young flower buds on it. While C. rostrata, C. dudleyana, C. lewisii, and C. cylindrica all have a recurved inflorescence when in bud and the terminal bud is facing down, the inflorescence of C. bottae is straight with the terminal bud facing up. This only refers to the stem itself; all of the buds of these species are nodding, but become erect right before the flower opens. Also keep in mind that once all of the buds have opened the stem will no longer be recurved, so this trait cannot be used to identify individuals that are at the end of their flowering period.
Ovary/Capsule
C. dudleyana can be easily distinguished from the other four speckled species by the very deeply 8-ribbed appearance of it’s ovary and capsule. In contrast, all four other species have shallowly four-grooved ovaries and capsules.
C. rostrata can be easily distinguished from the other species based on the long beak length (>5mm, but usually in the range of 7-15mm) of it’s capsule. C. bottae and C. lewisii both have capsules that are fatter and essentially beak-less (<3mm for lewisii, and 0mm for bottae). C. cylindrica has a capsule that also has a long beak (3-5mm), but shorter than that of C. rostrata.
The defining characteristic for differentiating the subspecies of C. cylindrica is the appearance of the capsule. C. cylindrica ssp. cylindrica has a capsule that is equal in width along it’s entire length, while C. cylindrica ssp. clavicarpa has a capsule that is wider at the distal end. I find that the difference in capsule shape can be very subtle, but their ranges are also non-overlapping which makes identifying them much easier.
Flower
There is a lot of variation within species in floral appearance. The background petal color can be light lavender to much darker magenta, and can fade to white at the base of the petal or not. The number of speckles can vary and the basal spot varies in the strength of color, and both of these types of spots can be completely absent. And the petals can vary in length and width. Thus, identifying the species based on floral appearance alone can be challenging.
C. cylindrica can be differentiated from the other four species compared here by the placement of the speckles, when present, and on the depth of color in the spot at the base of the petal4. While the speckles of C. rostrata, C. dudleyana, and C. lewisii are most concentrated at the very base of the petal, the speckles of C. cylindrica are shifted up and are most concentrated in the middle and distal portions of the petal. In addition, the pigmented region at the base of the petal is usually deeper in color, larger, and a redder purple in C. cylindrica. In contrast, this same basal region in C. rostrata, C. dudleyana, and C. lewisii is lighter and skinnier (usually takes on the appearance of a line more so than a spot). However, some occurrences of C. lewisii do appear to also have a deeply pigmented region at the base of their petals which looks very similar to the appearance of C. cylindrica, so in the absence of speckles, capsule traits should be used to differentiate these two species.
C. bottae can be differentiated from the other four species compared here by the lack of a distinct spot at the base of the petal and on the placement of the speckles, when present. When many speckles are present, the speckles of C. bottae are often asymmetric across the petals, with the upper petals usually more speckled than the lower petals (see examples here: https://www.inaturalist.org/observations/48974891 and https://www.inaturalist.org/observations/26094733). This also occurs in C. bottae’s sister species, C. xantiana, and may be due to the speckle pigmentation being light induced in these two species (my pet hypothesis, but it hasn’t been tested formally yet). The speckles of C. bottae are also quite large when compared to the speckles of the other species. In addition, C. bottae also usually lacks a distinct spot at the base of the petal; the base of the petal can be shaded a darker pink, but this shading is not usually bright in contrast to the rest of the petal and is usually almost unnoticeable except during petal development.
The flowers of C. dudleyana, C. rostrata, and C. lewisii are incredibly similar to one another and other characteristics (such as the ovary and capsule traits, see above) should really be used to identify these species. C. dudleyana does sometimes (mostly in the populations in the Sierra Nevadas) display faint vertical white streaks along it’s petals, which is distinct to this species, but otherwise the flowers from these three species are almost identical.
References
1 Parnell, D. R. (1970). Clarkia jolonensis (Onagraceae), a new species from the inner coast ranges of California. Madroño, 20(6), 321-323.
2 Calflora: Information on California plants for education, research and conservation. [web application]. 2017. Berkeley, California: The Calflora Database [a non-profit organization]. Available: www.calflora.org (Accessed: Jan 16, 2022).
3 Davis, W. S. (1970). The systematics of Clarkia bottae, C. cylindrica, and a related new species, C. rostrata. Brittonia, 22(4), 270-284.
4 Lewis, H, and Lewis, M. (1955). The genus Clarkia. Univ. Calif. Pub. Bot., 28, 241-392.
Resources
A really amazing guide on differentiating C. dudleyana and C. bottae:
http://tchester.org/plants/analysis/clarkia/bottae_dudleyana.html
Key for the Clarkia of California:
https://ucjeps.berkeley.edu/eflora/eflora_keys.php?key=9760
Kimmy Stanton 