Plants of Southern California: Clarkia bottae and C. dudleyana

Text by Tom Chester; (Most) Photographs by Michael Charters

C. bottae

C. dudleyana

Fig. 1. Left: Clarkia bottae, from Monrovia Canyon Park on 21 May 2014.
Note that the young flower bud is nodding, but the tip of the stem above the single nodding bud is not nodding (see text for links to two other photographs). The flower bud becomes upright just before it blooms, as seen in the photograph.
Right: C. dudleyana, from Chantry Flats on 19 May 2014. Note that the entire tip of the stem, with two flower buds, is nodding; it will eventually become erect as it matures.
Also note that the ovary of C. bottae is long and slender, with less-apparent ridges, compared to the overy of C. dudleyana.
Click on the pictures for larger versions without labels.

Table of Contents

Introduction and Distinguishing Characteristics
Geographic Distribution of These Taxa
Taxonomic History

Introduction and Distinguishing Characteristics

Clarkia bottae (=C. deflexa in Munz 1974) and C. dudleyana are similar species that are often confused even by good field botanists. The Jepson Manual descriptions of these two species are essentially identical. Their only distinguishing characteristic in the Jepson Manual, whether the stem tip is nodding or not when there are at least two buds remaining to open, is given only in the key, and is fairly subtle.

Both of these species have a long and tortured history, detailed in the Taxonomic History below.

As far as we know, these two species differ primarily in only three subtle, but important, characteristics:

  1. Both species have flower buds that nod when young, and become erect just before blooming. The tip of the stem in bloom, as long as at least two flower buds are present, is erect for C. bottae and nodding for C. dudleyana; see Fig. 1. This is the key characteristic to distinguish these two species in Munz 1974 and the Jepson Manual, both editions. Since this characteristic is somewhat subtle for C. bottae, it may help to see the following additional photographs showing its erect young upper stem: Photo by Don Quintana (see the erect stem behind and to the left of the photographed flower); Photo by Tony Valois, showing an almost identical erect stem which is seen more clearly in this photograph; Photo by CSU Channel Islands Native Plant Team (main page)

    This characteristic can only be observed when there are at least two remaining flower buds, since if only one bud remains, one cannot tell whether the stem is nodding, or if just the bud is nodding. If no flower buds remain, this characteristic is lost.

  2. The shape of the stigma was mentioned as the single distinguishing morphological characteristic in Jepson 1925, but oddly is not mentioned in Munz 1974 or the Jepson Manual, First or Second edition. This is the only characteristic that can be used to distinguish these two species on photographs of the flower from the front, or in mature plants that no longer have buds.

    C. bottae has stigmas that are fused at their base, forming a disk in the fused portion when the stigmas open, whereas C. dudleyana has stigmas free at their base. As a result, the stigmas for the two species look different, both before the stigmas open and when they are open. The unopened stigma of C. bottae is cut only halfway to the base, since the lower portion is fused, and quickly becomes wider than long as it opens. The unopened stigma of C. dudleyana is cut to the base, since the lower portions are not fused, and is longer than wide until it opens fully.

    Figs. 2 and 3 show the difference.

  3. C. bottae has a longer and more slender ovary and fruit, with fewer and less-apparent ridges, than the ovary of C. dudleyana. Fig. 4 shows the difference of the ovary in flower and in fruit.

Due to the confusion between these species, and the subtle characteristics that separate them, it appears that many seed companies sell plants of C. dudleyana labeled as C. bottae; see the nodding stem tip here and here; the shape of the stigma here from this page; the shape of the stigma here; etc.! Some seed companies possibly even sell a completely different species as C. bottae: note the shape of the stigma and the very wrong leaves here.

Fig. 2 shows the flowers of both species, with the stigmas circled.

C. bottae

C. dudleyana

Fig. 2. Pictures of the flowers of both species, with the unopened stigmas circled.
Left: C. bottae has an unopened stigma wider than long (except for very young unopened stigmas; see Fig. 3).
Right: C. dudleyana has an unopened stigma longer than wide.
See Fig. 3 for close-up photographs showing the evolution of the shape as the stigmas open.
Pix of Clarkia bottae from Big Sycamore Canyon in the Santa Monica Mountains. Pix of C. dudleyana from Kings Canyon in the Sierra Nevada.
The red flecks can be present or absent in flowers for both species, so are not a distinguishing characteristic. Click on the pictures for larger versions without the circles.

The unopened stigmas of C. bottae are cut only halfway to the base, and are initially longer than wide. They quickly become wider than long as the stigmas begin to open, since they open only from the halfway point. A central depression is visible until the stigmas become fully mature and the stigma becomes head-like.

The unopened stigmas of C. dudleyana are cut nearly to the base, and are always longer than wide until the stigma lobes open completely. When the stigmas open, they are joined only in their lowermost part, and do not form a central disk. The stigma never becomes head-like at any stage.

Fig. 3 shows the variation in the shape of the stigma as it opens for each species.

C. bottae


C. dudleyana

(intentionally blank)  
Photo by Aaron Schusteff

Photo by Aaron Schusteff
Fig. 3. Examples of the evolution of the stigmas for both species as they open and mature. Left: Clarkia bottae, from Monrovia Canyon Park on 21 May 2014. Some of the photographs show pollen attached to the stigmatic hairs. Right: C. dudleyana.
Click on the pictures for larger versions without labels.

C. bottae has a longer and more slender ovary than does C. dudleyana. This difference is easily seen in Fig. 1. The ovary and fruit of C. bottae are also ridged differently, with fewer and less-apparent ridges. Fig. 4 shows the difference in appearance of the ovary in flower and in fruit.

C. bottae


C. dudleyana

Photo by Tom Chester
(no photograph yet)  
Fig. 4. Ovaries of C. bottae (left) and C. dudleyana (right) in flower (top) and in fruit (bottom). Click on the pictures for larger versions.

Geographic Distribution of These Taxa

Fig. 5 shows the voucher distribution of these two taxa, from a search of the Consortium of California Herbaria on 21 May 2014. North of the Transverse Range, their distributions are disjunct, with C. bottae found in the coastal range, and C. dudleyana in the Sierra Nevada.

In southern California, the separation into "coastal" and "inland" holds in part, but the species overlap in "in-between areas" of the Los Angeles basin and adjoining portions of the San Gabriel Mountains.

C. bottae

C. dudleyana

Fig. 5. Geographic distribution of C. bottae (left) and C. dudleyana (right). The top maps show the entire distribution in California. The bottom maps show just the southern California distribution where the range of the species overlap, with the range for each species shown on the voucher map of the other species for easy comparison of the area where the two species overlap. The single C. dudleyana voucher in the Santa Monica Mountains is possibly misdetermined, since that species is not otherwise known from that area. Click on the maps for larger versions.

Taxonomic History

This section gives the time-order taxonomic history of C. bottae, C. deflexa, and C. dudleyana, with brief mentions of two species from Monterey County, C. lewisii and C. jolonensis that are close to these species.

Raven and Parnell (1977) synonymized C. deflexa with C. bottae, under the name of C. bottae, probably because those two species were previously separated only by some pretty insignificant characters. Prior to that paper those species were treated as separate taxa. This synonymy was followed in the Jepson Manual First and Second editions.

As detailed above, these species are so close that it is not surprising that things were confused between these three species until (presumably, since I don't have their paper) Lewis and Lewis recognized the importance of the "nodding or not" inflorescence stem tip and buds as separating characteristics in 1953.

Each reference has been separated with a divider, with the date as a header, to make the time history clearer.

These species were in the genus Godetia until 1953, separated from Clarkia as follows:

- petals distinctly clawed, often much lobed .. Clarkia
- petals sessile, not lobed (except in one species); stamens 8 ... Godetia


C. bottae was the first of these species to be recognized, in 1835, as Godetia bottae Spach, in Nouv. Ann. Mus. Par. iv. (1835) 393. The specimens were collected by Paolo Emilio Botta, who collected along the California coast in 1827 and 1828. The only mention of the type locality is Spach's paper is "in California australiori legit cl. botta <=== (V. s. sp.)". I have no idea what this means other than "southern regions of California".

This paper does not mention whether the inflorescence or buds are nodding. Spach had only 11 species in his key.

There are two specimens from Botta's trip in the Museum National d'Histoire Naturelle in Paris. One of these, the type specimen, is annotated as "Godetia bottae, nob. (Spach, 1839), California, M. Botta". The second is annotated "Godetia bottae, Monterey, M.P.E. Botta, 1829". (Raven and Parnell 1977)

Raven and Parnell state that this second inscription had to be made in error, since these two specimens are actually of C. deflexa (Jeps.) Lewis & Lewis 1955 (determination corroborated by Harlan Lewis), which "does not occur at or near Monterey", and claim that these two specimens were actually collected either at Santa Barbara or at San Pedro.

However, Raven and Parnell synonymize C. deflexa with C. bottae (see below), and there are 66 online Consortium vouchers of C. bottae from Monterey County (search done 25 May 2014), including 13 vouchers from Lewis and Lewis, and one from Raven. One therefore has to wonder about whether how reliable this deduction was that Botta did not collect his specimen from Monterey.


C. dudleyana was defined in 1904 as Godetia dudleyana Abrams, in Abrams' Flora of Los Angeles, 1904, p. 267

Abrams had both G. bottae and G. dudleyana as having flowers drooping in bud, with no mention of whether the stem tip was drooping or not.

Abrams had G. bottae as

common in the Santa Monica Mountains and in the foothills above Los Angeles. G. pulcherrima Greene is apparently the same, Dr. Greene having evidently confused this species with the next (G. dudleyana).

Abrams had G. dudleyana as

frequent in the upper portions of the chaparral belt of the San Gabriel Mountains. The type is the author's number 2625, collected in Little Santa Anita Canyon at 2500 feet altitude. This species has been confused with G. bottae, but it is much nearer G. hispudula Wats.


C. deflexa was defined in 1907 as Godetia deflexa Jepson in Univ. Calif. Publ. Bot. ii. 332 (1907). Unfortunately, I can't find this article online.


Things are very different in Jepson's Manual of the Flowering Plants of California in 1925!

Jepson 1925 had all three species, with:

Jepson 1925 separated them as follows:

- Sierra Nevada species; stigmas distinct ==> G. dudleyana
- coast range species; stigmas partly united so as to form a saucer-shaped or shallowly cup-shaped base
     - ovary often curved before anthesis; infl usually very loose; capsule typically with flat sides ... G. bottae
     - ovary straight and deflexed before anthesis; flowers in a closer spike ... G. deflexa

From this key, apparently C. deflexa was named for the pendent buds, even though all three species have pendent buds. It is odd that this key implies that C. bottae did not have deflexed buds, instead calling the ovary merely "often curved before anthesis".

The difference between G. bottae and G. deflexa is pretty subtle in this key, so one might have predicted they were actually the same species.

By the way, note that Jepson apparently considered SnGb and SnBr as part of the Sierra Nevada, a "non-coast-range"!


In 1953, these Godetia species were transferred to Clarkia species by F.H.Lewis & M.E.Lewis in Madroņo 12: 33.


Munz, in his 1959 Flora of California (red Munz) delineated these three species differently, in the modern way, but not apparently in the modern sense for each species! I presume that Lewis and Lewis were the ones to recognize the importance of whether the inflorescence step tip and/or the flower buds were deflexed or not, since it first appears in this treatment.

The Munz key was:

H. rachis of infl straight, only the buds deflexed. Orange co. to Monterey Co. .. C. deflexa
HH. rachis of infl reflexed at tip, becoming straight as the buds mature.
    I. immature capsule 4-grooved or the grooves obscure. Monterey County... C. bottae
    II. immature capsule 8-grooved, -ribbed or -striate
        J. petals mostly streaked with white. Tuolumne Co. to Riverside Co. ... C. dudleyana
        JJ. petals not streaked with white. Eldorado Co. to Butte Co. ... C. biloba brandegeae

The ranges of C. bottae and C. deflexa are completely different than in Jepson 1925; the key is different; and a fourth taxon appears in the key.

C. bottae went from being a widespread species in SnGb, Santa Monica Mtns to Monterey County, to being restricted to Monterey County, with the range of C. deflexa being expanded from just "sandy plains of Los Angeles" to "Orange County to Monterey County". But remember, these two taxa are now considered to be identical, and so it is not surprising that delineations of their ranges might change if they were based on features not significant enough to separate plants into these two species.

Note that according to the modern delineation of the species in the Jepson Manual, this key is incorrect for C. bottae, since it has its axis of infl in bud straight, not reflexed at tip.

C. biloba brandegeae is separated from C. dudleyana now by its 2 lobed petals, compared to the entire or occasionally notched at tip petals of C. dudleyana.


Parnell defines Clarkia jolonensis (Madrono, 20: 322, 1970), from Monterey County as distinct from C. deflexa. I don't have the reference, so I don't know how these two species were distinguished in this paper.


Munz, in his 1974 Flora of Southern California, has just two species, C. deflexa and C. dudleyana, separated in the modern sense by whether the tip of the inflorescence stem is initially reflexed or not. Munz doesn't have C. bottae since he apparently still considered that to be only in Monterey county, and his flora was just for southern California.

Munz gives the following geographic ranges for each species:


Raven and Parnell claim that the type specimen of C. bottae is actually of C. deflexa, and define a new species C. lewisii from Monterey County based on a voucher from "Point Lobos, along the trail to China Cove from the end of the road, 26 June 1947, H. and M. Lewis 498 (LA)". They equivalence this as C. bottae in the sense of the Lewis & Lewis publication in 1955, not the C. bottae Spach.

They synonymize C. deflexa with C. bottae, under the name of C. bottae.

This is the basis for the wonderful statement, that can only happen in botany, that what we now call bottae was called deflexa, and what we now call lewisii was bottae (Brian LeNeve, 2014, personal communication). (:-)

The story gets yet more curious.

The type specimen for C. lewisii, housed at LA, is now determined as C. bottae! Furthermore, the online version of the duplicate of that specimen at UC is determined as C. jolonensis, determined by Parnell in 1997! However, the physical image of the voucher, linked from the online record, shows it only determined as C. bottae.

Fortunately, since we're only concerned with distinguishing C. bottae and C. dudleyana, we are not going to even attempt to figure out what is going on in Monterey County.

We just note that the Jepson Manual second edition retains both C. lewisii and C. jolonensis. It distinguishes C. jolonensis from C. bottae by the color of the seeds, gray for C. jolonensis and brown for C. bottae. It distinguishes C. lewisii from C. dudleyana (not from C. bottae!) by ovary and immature fruit 4-grooved for C. lewisii and 8-grooved for C. dudleyana. The image of the type specimen duplicate for C. lewisii shows a plant mostly in fruit, so whether the inflorescence is nodding or not cannot be deduced from that voucher, perhaps accounting for Raven and Parnell thinking it was C. bottae in the sense of Lewis and Lewis.

1993 to 2012

The Jepson Manual first edition 1993 combined C. bottae and C. deflexa, under the first-published name of C. bottae, following the synonymy of Raven and Parnell 1970. The range of C. bottae was SCoRO, SCo, WTR, nw PR, as shown above in more detail from voucher locations. The range of C. dudleyana was c&s SNF, Teh, TR, as described above in more detail from voucher locations.

The treatment of these species remained the same in the Jepson Manual second edition 2012.

Text by Tom Chester; photographs by Michael Charters except where credited to others.

We thank:

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

Go to:

Copyright © 2014 by Tom Chester and Michael Charters
Permission is freely granted to reproduce any or all of this page as long as credit is given to us at this source:
Comments and feedback: Tom Chester
Last update: 26 May 2014