A short guide to Callirhoe in Texas


Introduction

The genus Callirhoe is a genus of mostly reddish and sometimes white flowering plants in the Malvaceae family native to North America. The most recent study of the genus was by Dorr in A Revision of the North American Genus Callirhoe (Malvaceae) published in 1990 and revised for the Flora of North America (FNA) in 2015. Of the nine species in the genus, six species are found in Texas, five of which have observations on iNaturalist with the sixth being a rare endemic found in only a few counties of the southern Rolling Plains. A seventh species occurs in Oklahoma along the Texas border in southeast Oklahoma.

The reason for providing this short guide is to educate iNaturalist observers and identifiers about the ways to identify the different species in Texas. In the spring of 2021, I began checking various research grade observations and noticed a significant number of incorrect observations. At that point I decided to perform a more rigorous examination of research grade(RG) observations, starting with RG observations from Texas for the year 2019 which I had not already identified as the initial data set. Preliminary results indicate that as high as 19.6% (168 of 856) of those observations were misidentified. While I plan to explore those results and others in a future post, it became clear that an article illustrating some of the concepts used in identification might be helpful for the iNaturalist community.

Species in Texas

A mentioned in the introduction, six species of Callirhoe occur in Texas. Relatively accurate distribution maps for these species can be found on the Biota of North America Program (BONAP) website, though the distribution of one species, C. leiocarpa maybe somewhat erroneous and currently seems to be in flux. The six species as ordered by their prevalence on iNaturalist are (vegetation areas as found in Shinners & Mahler's Illustrated Flora of North Central Texas (1999) (FNCT) :

  • C. involucrata: The most widespread of all Callirhoe species, it occurs throughout most of Texas, except for part of the Trans-Pecos in west Texas and parts of northeast Texas where it gives way to C. papaver.
  • C. pedata: Occurs in central and north central Texas from the Edwards Plateau north to the Oklahoma border in the Cross Timbers and Prairies and Blackland Prairies
  • C. leiocarpa: Historically occurred in central and south central Texas from the Edwards Plateau south to the Gulf Coast in the South Texas Plains and Gulf Prairies and Marshes. Recent iNaturalist observations have placed it in additional areas, typically around the major metropolitan areas of Houston, Dallas, and Fort Worth, where it was not found before
  • C. alcaeoides: Restricted to the northern sections of the Cross Timbers and Prairies and Blackland Prairies, starting just south of the DFW area and running northward into Oklahoma
  • C. papaver: In east and southeast Texas, primarily in the Piney Woods and Gulf Coast Prairies and marshes possibly into Post Oak Savannah and Blackland Prairies
  • C. scabriscula: A rare endemic known from only a few counties along the Colorado River in the Rolling Plains. It has not been observed on iNaturalist as of this time and I will not discuss it in detail.

A seventh species, C. bushii, occurs in southeast Oklahoma along the border with Texas. It has not been found in Texas, though it seems plausible that it might occur along the Red River in northeast Texas. One notable exception to this list is C. digitata, which is no longer treated as being found in Texas. Earlier generic treatments apparently included the species now known as C. pedata within the species concept of C. digitata, but that is no longer the case. C. digitata is now only known to be found outside Texas in states like Arkansas, Oklahoma and Missouri, though BONAP maps do show them close to the Oklahoma Texas border. The last species, C. triangulata, is found outside of Texas, primarily in Illinois and Wisconsin but also sporadically in the southeast.

Identification

Identification of the different species of Callirhoe in Texas is not difficult if the appropriate features are photographed. Of course, as with many species on iNaturalist, this is the problem. A significant percentage of Callirhoe observations consist of one or maybe two photos of the flower, which generally is not sufficient for a high confidence identification, at least not one based on the morphological features.

Keys to the genus can be found in both the Flora of North America online (efloras) and also in the Flora of North Central Texas (omitting C. papaver and C. scabriscula), which is also available online at the BRIT website. The reader can reference those keys; I will try to illustrate some of the important differentiating features found in them below. While corolla color can be helpful, some other important features are the buds, the involucel, the stipules, the fruit, the inflorescence and the sepal hairs.


Useful features for identification of common Callirhoe species in Texas
Species Petal
Color
Involucel Buds
Valvate
Stipules
Auriculate
Fruits
Large Beaked
Sepal
Hairs
C. involucrata reddish purple basally white, white, white with reddish vertical stripes yes no - no long simple
C. papaver reddish purple usually yes - no long simple
C. leiocarpa reddish purple basally white no yes yes yes glabrous
C. pedata reddish purple to purple basally white or lightened, white, rarely pink no yes no no mostly glabrous
C. alcaeoides white to light pink no yes - yes short simple

Petal color

Callirhoe in Texas are predominately reddish purple in color, though white species and forms also occur as well as more rare pink ones. The reddish purple ones range from a reddish purple to purple, often within each species. All species in Texas have a red purple form except for C. alcaeoides, though it can appear light pink. Almost all red species of Callirhoe in Texas can have some lightening of the color near the base of the petals, which can sometimes be completely white. This is sometimes referred to as a basal spot. Typically C. involucrata has a basal spot, often larger than in other species. C. leiocarpa generally has a smaller basal spot, while C. pedata and C. papaver often have a mild lightening of color, rarely completely white. Given the variation of this feature, it cannot be used as a diagnostic feature on its own, but is sometimes helpful in conjunction with other features. The literature indicates that C. pedata does not have a basal spot, but it seems specimens on iNaturalist apparently do (TBD).

Red forms of Callirhoe: C. involucrata, C. leiocarpa, C. pedata (C. papaver not pictured)


Three species of Callirhoe have white or light colored forms. C. alcaeoides is predominately white but sometimes light pink. Two of the predominately reddish species, C. pedata and C. involucrata have light forms, which are often initially identified incorrectly as C. alcaeoides. Like C. alcaeoides, the mostly reddish purple C. pedata can also be white and rarely light pink, with all three colors sometimes occurring in the same population. Light forms of C. involucrata also occur, chiefly in areas of central Texas. In Williamson county, a white form with variable large reddish vertical stripes is the predominant form, and it is probably the only one which can be identified on the basis of the corolla color alone. White color forms also occur on parts of the Edwards Plateau such as Kimble and Menard counties (Enquist 1987).

Light forms of Callirhoe: C. alcaeoides, C. pedata, C. involucrata, C. involucrata var. lineariloba (see discussion on varieties)

Involucel

The Flora of North America(FNA) key for Callirhoe starts with the presence or absence of the involucel, which for Callirhoe consists of a whorl of bracts (usually three, sometimes one) subtending the sepals. This feature separates the species into those with an involucel (C. involucrata, C. papaver, C. scabriscula, and C. bushii) and those without (C. pedata, C. leiocarpa, C. alcaeoides). The involucel is most easily visible on the buds, but can also been seen, sometimes with greater difficulty due to proximity of the sepals, on open flowers and fruiting heads. In many parts of Texas, C. involucrata is the only species with an involucel and a somewhat confident identification can be made based on the presence of that feature. Where C. involucrata comes in contact with other species that have an involucel (in Texas mostly C. papaver in east and southeast Texas but also potentially northeast Texas close to C. bushii) other features need to be used for identification, such as the bud (see next section).

C. involucrata with an involucel (other species not pictured)


In treatments previous to Dorr, the spacing between the involucellar bracts and the base of the calyx was sometimes used to differentiate C. involucrata and C. papaver, with that space being larger in C. papaver than C. involucrata. However, according to Dorr (1990), though this is generally true, it is apparently not always consistent or clear, so he uses the buds to differentiate the two species.

There is some variation in the size and shape of involucel bracts. Outside of Texas, this variation is useful for differentiation of various species such as C. papaver,C. bushii, and C. triangulata. Within Texas, the variation is one of the characters used in differentiation of the varieties of C. involucrata.

Species without an involucel C. leiocarpa, C. pedata, C. alcaeoides

Bud

The bud is a useful feature to capture for identification of Callirhoe in Texas, though it is rarely photographed on purpose. Buds in C. involucrata are unique among Callirhoe in that the tips of the sepals/calyx lobes do not come together to form a point; there is always at least some small amount of visible separation. When the calyx lobes do come together in the bud, the bud is called valvate and all Callirhoe species besides C. involucrata have valvate buds. The FNA key only uses this feature to differentiate C. involucrata from other species with involucels, but it can be used to differentiate C. involucrata from all other Callirhoe species. The presence of a non-valvate bud provides evidence that the specimen is C. involucrata whereas the presence of a valvate bud rules out as a C. involucrata possibility.

An added benefit of capturing the bud is that the presence or absence of an involucel is often more obvious on buds than on flowers.

Non-valvate bud of C. involucrata with sepal tips widely separated


Valvate bud of C. leiocarpa and C. alcaeoides with sepal tips coming together


Stipules

The stipules (leaf like appendages at the base of the leaf petioles) are useful for distinguishing two of the taller species in Texas, C. leiocarpa and C. pedata, which are somewhat similar in their lack of involucel and typically erect or ascending habit. C. leiocarpa has auriculate stipules whereas C. pedata does not. This feature needs to be used with other features for identication such as the absence of an involucel, as C. involucrata can have somewhat auriculate stipules as well. Though this differentiating character was included in Dorr's original treatment, it was not included in the key for Callirhoe in the Shinners & Mahler's Illustrated Flora of North Central Texas, which references Dorr's work. It is possible that is not completely definitive as I have seen some examples of C. pedata with small projections at the stipule base, but it does generally seem to be true and is present in Dorr's FNA key. The auriculate stipules of C. leiocarpa are often noticeable even from a distance and in blurry photographs. This may be because their inner face (adaxial) tends to spread away from the stem and face upward like a sessile leaf or because the auricles often encircle much of the stem.

C. leiocarpa stipules


C. pedata stipule

Fruit

The fruit is primarily helpful in distinguishing C. leiocarpa from other species in Texas due to its distinctive appearance. The fruit (shown in pictures below) is usually seen from above and visually has three distinct regions from this perspective: a whitish central column, a variable size greenish ring (appears to correspond to the beak), and lastly a variable size greenish yellow to white ring on the outside (which can correspond to the seed bearing portion or the collar depending on species). In C. leiocarpa, the green region (corresponding to the beak) is quite large compared to the encircling white ring (in this case a small projection called a collar which subtends the beak). Most other species in Texas, except C. alcaeoides, have a smaller green region in mature fruit (a smaller beak), similar to that shown in the C. pedata example below. Note however, that C. pedata is described as having a larger beak in areas outside of Texas, so the this difference may not apply there. Also note that immature fruit of C. pedata often have a large green area relative to the surrounding white area, so it is important to observe a mature fruit.

Fruit of C. leiocarpa, C. pedata


Inflorescence and Sepals Hairs - C. pedata and C. alcaeoides

Both C. pedata and C. alcaeoides are species without an involucel and both can have white corollas, which can make differentiation difficult in areas where they overlap as Dorr noted(1990). While the inflorescence of C. alcaeoides is typically more compact (Dorr's FNA key describes the racemes as "appearing corymbose or subumbellate"), it can be difficult to ascertain this in photos, especially in younger plants where the first open bloom may be close to unopened buds. The FNCT uses an additional character to differentiate the two species, the vestiture of the calyx. In C. alcaeoides it is described as "hispid-pubescent" and in C. pedata as glabrous or sparsely pubescent. Neither of Dorr's treatments use this as a differentiating character between the two species, though the species descriptions in his earlier work do seem to agree with this difference.

Early compact inflorescence of C. alcaeoides

Pubescent sepals of C. alcaeoides


Glabrous sepals of C. pedata

Varieties of C. involucrata

C. involucrata is described as having three varieties, two of which occur in Texas and which Dorr indicates are weakly separated(2015). The two which occur in Texas are C.i. var. involucrata and C.i. var. lineariloba. These two varieties are differentiated by qualities of the leaves, stipules, involucellar bracts and lastly, only partially sometimes, color. Determination of the variety for the most part requires close examination and measurement of these features, so it is typically not possible to make a variety determination based on photos alone. It is probably the case that most observations of C. involucrata in Texas are in fact C.i. var. lineariloba as attested to by Dorr's range maps(1990) for C. involucrata which show that C.i. var. lineariloba is found throughout much of Texas with C.i. var. involucrata apparently only occurring in a few areas in north central Texas, predominantly close to the Oklahoma border.

Some confusion exists around identification of C.i. var. lineariloba in central Texas where the variety has become associated only with the color form found mostly in Williamson county (sometimes referred to by locals as the "Williamson county winecup"), where the petals typically have a central broad vertical reddish purple region borded by white margins. I don't know if earlier research restricted the variety to this color form, but it is implicitly done in Marshall Enquist's popular Wildflowers of the Texas Hill Country (which predates Dorr's work). Ironically, though C.i. var. lineariloba probably mostly consists of reddish purple specimens, only this different color form is easily identifiable as C.i. var. lineariloba since it does not apparently occur in C.i. var. involucrata (or any other species of Callirhoe for that matter). Thus the misconception is perpetuated, but I don't see an easy way around it.

iNaturalist Observations


Observations used in the guide
Taxon Observation
Number
Features Illustrated
C. alcaeoides 88741669 white color, lack of involucel, valvate bud, sepal pubescence
C. involucrata 80169982 red color
C. involucrata 78509404 white color
C. involucrata 50584755 involucel
C. involucrata 83043800 involucel
C. involucrata 78382364 bud
C. involucrata var. lineariloba 89098715 light color with reddish regions
C. leiocarpa 88464887 red color, lack of involucel, valvate bud, auriculate stipules, fruit
C. pedata 88721978 red color, glabrous sepals
C. pedata 88745685 white color, lack of involucel, fruit
C. pedata 89107758 non-auriculate stipule

Other useful Observations
Taxon Observation
Number
Features Illustrated
C. papaver 34204774 involucel, bud

References

Diggs, G. M., Lipscomb, B. L., O'Kennon, B., Mahler, W. F., & Shinners, L. H. (1999). Shinners & Mahler's Illustrated Flora of North Central Texas. Botanical Research Institute of Texas.
Dorr, L. J. 1990. A Revision of the North American genus Callirhoe (Malvaceae). Mem. New York Bot. Gard. 56: 1–75.
Dorr, L. J. 2015. Callirhoe. In: Flora of North America Editorial Committee, eds. 1993+. Flora of North America North of Mexico. 20+ vols. New York and Oxford. Vol. 6. http://www.efloras.org/florataxon.aspx?flora_id=1&taxon_id=105128
Enquist, Marshall. 1987. Wildflowers of the Texas Hill Country. Lone Star Botanical, Austin, Texas.
Kartesz, J.T., The Biota of North America Program (BONAP). 2015. North American Plant Atlas. (http://bonap.net/napa). Chapel Hill, N.C. [maps generated from Kartesz, J.T. 2015. Floristic Synthesis of North America, Version 1.0. Biota of North America Program (BONAP). (in press)].

Copyright

All photos used in this post are the property of Ryan McDaniel, all rights reserved.

Revisions

1.0 - August 6, 2021 - Original revision.

Posted on August 6, 2021 07:14 PM by rymcdaniel rymcdaniel

Comments

This is really fantastic! Thanks for posting.

Posted by bosqueaaron over 2 years ago

I had been wondering how to easily tell apart C. pedata and C. leiocarpa for several years now: the stipules! Top notch stuff! Many thanks!

Posted by joshua_tx over 2 years ago

@joshua_tx Yes, when I was primarily using the Flora of North Central Texas to try to differentiate them, I basically gave up, but the stipules seem to work for the most part. I have seen some border line ones where I can't make a species ID though based on what is photographed, so I wonder if there might be some overlapping variation. It's something to investigate further in the future. I'd say if you can remember, definitely get a photo of the fruit if present.

Posted by rymcdaniel over 2 years ago

This is GREAT! Thank you for spending the time to create it. I hope that it helps us get more accurate ID's!

Posted by suz over 2 years ago

Absolutely awesome, Ryan! One of the best, if not the best, identification guides I have seen. Love the photos and the layout. And thanks for your patience and explanations on all my mis -IDs of this genus!

Posted by connlindajo over 2 years ago

Ryan, be sure and add a link to this post in the "More Info" list under the "About" tab for every taxon involved here. I'm assuming you have Curator privileges. Or I can put that on my Do List or help you with it.

Posted by gcwarbler over 2 years ago

@gcwarbler Shows how much I know. I don't think I have ever even looked at the "More Info" box. I don't have Curator privileges, but if you would like to add it to that area, feel free.

Posted by rymcdaniel over 2 years ago

Done. You can look at the "More Info" list under the About tab for the five primary taxa covered in this journal post and see the link.

Posted by gcwarbler over 2 years ago

@gcwarbler Thanks!

Posted by rymcdaniel over 2 years ago

This is beautifully made, Ryan, I'm sure countless hours of works must of gone into this! We should probably get @lisa281 to add this to the list of Helpful Identification Guides so we can spread this to more people.

Posted by arnanthescout over 2 years ago

I am happy to hear that Callirhoe papaver is now included in the Computer Vision model as of April 2022. Though it did not have 100 observations when they started the new model back in October, it either had enough when they started to rerun the model or satisfied the condition of having over 100 photos, which was different from what was required from the previous models. Anyway, at least now those in the eastern part of the state should see this as a suggestion now.

Posted by rymcdaniel almost 2 years ago

Add a Comment

Sign In or Sign Up to add comments