Unusual Spigelias in Central Texas

Note: Most Spigelia plants in Central Texas are pretty clearly Spigelia hedyotidea. As far as I can tell the borderline plants are not as common, and occur in different habitat than typical S. hedyotidea. See bolded note on borderline plants.

My interest in these plants was sparked by this observation:
https://www.inaturalist.org/observations/161471597

Spigelia is one of a few genera in the family Loganaceae which occur in Texas. There are three species of Spigelia in Texas:

• Spigelia marilandica A.DC, restricted to East Texas
• Spigelia texana (Torr. & Gray) A.DC., apparently endemic to Texas in the Gulf Coastal Plain
• Spigelia hedyotidea A.DC. , occuring in the Edwards Plateau south to South Texas and Mexico

(Species distributions roughly determined through descriptions in Correll and Johnston's Manual of the Vascular Plants of Texas (1970) (MVPT), maps from the Biota of North America Program maps (BONAP), and data from the Global Biodiversity Information Facility (GBIF). For Spigelia texana, I referenced personal comm. with Bill Carr.)

The last two are of interest as they are both morphologically similar and closely related to each other according to a dissertation by Gould (1997), along with Spigelia loganioides (Torr. & Gray) A.DC. All three taxa used to be placed in the genus Coelostylis and have been shown to form a monophyletic group (Gould 1997). The taxonomic boundaries within this group have been debated. Some authors, such as Hendrickson (1996) and Hurley (1968), propose that S. texana and S. loganioides are conspecific i.e. the same species. However, while S. texana and S. loganioides are most morphologically alike (and in fact nearly identical in appearance), Gould determined using cpDNA and ITS sequence analyses that "S. texana is more closely related S. hedyotidea than to S. loganioides." Based on information from online databases such as GBIF, Integrated Taxonomic Information System (ITIS), and Plants of the World Online (POWO), it appears that the three species are currently considered separate taxonomic entities as of May 2023, which Gould concluded in her paper, and which I will follow with.

Interestingly, Gould (1997) notes that there are populations of S. hedyotidea in Central Texas which appear to share vegetative characteristics closer to S. texana, although this is not elaborated on. Gould does not mention whether those populations were sampled in her analyses, but for each population, morphological characters were measured and recorded, although the data was not shown in the paper. Voucher specimens were deposited at TEX/LL, so I may look at some the S. hedyotidea vouchers from population sampled in the study fall of this year.

A good explanation of the morphological differences between the two can be found Gould's (1997) discussion under both Spigelia texana and Spigelia hedyotidea:

• S. hedyotidea has a shorter "bushy" growth habit (5-15(-19) cm) compared to S. texana ((10-)20-45(-50 cm) [numbers with dashes indicate uncommon extremes]

• S. hedyotidea has leaves which are much smaller than those of S. texana;
  Mid-stem leaves for S. hedyotidea are (1.2-)1.5-3.0(3.5) cm long and 0.3-1.0(-1.3) cm wide & shorter than the internodes;
  Mid-stem leaves for S. texana are 3.0-5.5 cm long & sometimes longer than the internodes.

• S. hedyotidea has leaves that are generally thicker and slightly coriaceous (leathery), often with the presence of minute papillae (think dots on a tongue) on the upper surface to create a scabrous look, although sometimes glabrous as well. S. texana will have thin, membraneous leaves without any papillae/roughness, except occasionally along the margins.
  Note 1: In the shade, S. hedyotidea can exhibit membraneous leaves as well.
  Note 2: In herbarium specimens, S. hedyotidea leaves tend to get wrinkled when dry; those of S. texana dry flat

• S. hedyotidea is "profusely scabrous" on the stems and leaves; S. texana is glabrous except for papillae at the nodes.
• S. texana usually produces a "pseudowhorl" of 4 leaves (4 leaves attached to one node) under the inflorescence. S. hedyotidea can do so, but rarely does.
• The corolla tubes of S. hedyotidea (ca. 4 mm wide at throat) are apparently slightly wider than those of S. texana (1-3 mm wide at throat).

Henderson's paper provides excellent illustrations which detail some of these differences on page 97 of the online version at the Biodiversity Heritage Library (BHL). Note that Henderson considers S. texana under S. loganoides, but the two are so morphologically similar (differing only by corolla length and allopatric distribution) that the illustrations remain useful.
https://www.biodiversitylibrary.org/page/9303086#page/99/mode/1up

S. hedyotidea is the most variable species and thus is the cause of most confusion between the two.

The Spigelia observation which I first mentioned above was located in Travis County along the River Place Trail, within the Balcones Canyonlands region of the Edwards Plateau. The plant exhibits characteristics similar to that of Spigelia texana. Most notably, it contains a whorl (or "pseudowhorl" as called in the literature) of 4 leaves on one stem, underneath an inflorescence. It also contains leaves which are less thick and more membraneous in texture.

"Pseudowhorl" of 4 leaves underneath an inflorescence

The location of this plant seems to fit better with S. texana than with S. hedyotidea. The plant's location is very close to a creek and in a shaded woodland area.
MVPT describes the habitat for S. texana as "wooded slopes and floodplain woods along rivers," and in both Gould and Hendrickson in wooded creeks and riparian forests.
S. hedyotidea occurs "in open gravelly soil and among boulders... about breaks or in prairies" according to MVPT. Gould and Hendrickson describe its habitat similarly: "open, gravelly, sandy clay loams, dark-soiled prairies... limestone bluffs on generally dry soils." However, they also add that it occurs on "shaded woodlands, river banks, and rocky creekbeds," which is completely different from the first set of habitats.

I took a quick hike down to locate the plant or others with similar characteristics at River Place, and made a few findings.
https://www.inaturalist.org/observations/162239090
https://www.inaturalist.org/observations/162239104

I noticed that the plants were all growing on shaded sloping ground around the creek. This is the same habitat which S. texana is described to grow in, although if accounting for Gould and Hendrickson, S. hedyotidea can also occur in similar environments. The leaves seemed notably large; midstem leaves of the two specimens I observed both surpassed 4 cm in length, which is significantly beyond the normal range noted for Gould, Hendrickson, and the MVPT. I admit that I was biased towards more closely observing plants that were larger in aboveground growth, so this may be skewed towards the upper extreme. A better method would have been to measure the mid-stem leaf length on all plants for a better picture, which I did not get to due to both limited daylight and time that day. Other notable features of leaves were that they were indeed noticeably thin and membraneous rather than coriaceous. The plants overall appeared glabrous on the leaves and stems, though this seemed to vary and I didn't check with my loupe on all of them. Not everything was similar S. texana, though; I did not detect the presence of a pseudowhorl of 4 leaves on any plants I found.

It is clear that these plants occupy a different niche than what I will call the "typical" S. hedyotidea, which tends to occur in areas with more sunlight (ranging from out in the open areas to partial shade) and tolerate drier soils. The question, for me at least, seems to be whether or not this variation falls under S. hedyotidea.

There are a few hypotheses here:

• It could be attributed to variation under Spigelia hedyotidea. S. hedyotidea is quite variable in appearance. Henderson notes that plants have "longer internodes, less coriaceous... and often larger leaves" in shady areas, as well as appearing "smooth and membraneous" like S. texana. Gould also cites this variation as well. According to both authors, these plants would probably fall under S. hedyotidea.
• It could possibly represent S. texana. If keyed out through MVPT, these plants would likely be considered S. texana, although that information is from 1970 and is thus more outdated than the more recent literature.
• It could be something else, maybe an intermediate between the two.

I think studying morphology under S. hedyotidea specimens analyzed Gould's paper would likely clear this up best, as those specimens are tied to genetic analyses which supported the recognition of the three former Coelostylis species as they are. Considering Gould's paper, which appears to be the most recent treatment, these plants would fall under variation of S. hedyotidea, so unless there is a newer update in the taxonomy, S. hedyotidea would be a safe bet. For more cautious individuals, it may suffice to leave the plants at genus.

I have created a set of observations using the observation field "Similar observation set" under the value "81436857_ct_spigelia" so that Spigelia plants exhibiting similar leanings toward S. texana can be easily found and studied.
https://www.inaturalist.org/observations?verifiable=any&place_id=any&field:Similar%20observation%20set=81436857_ct_spigelia

Borderline Spigelia plants will likely occur in Central Texas, and will likely (1) occur in shaded areas with ample moisture, along creeks and waterways and in bottomland areas (2) exhibit leaves that tend to exceed 3 cm in length and 1 cm in width (3) have thinner leaves that appear smooth and membraneous rather than rough and coriaceous and (4) contain a pseudowhorl of 4 leaves under the inflorescence.

To be frank, the identification of these plants is not so important, whether they are ID'd as S. hedyotidea or remain at genus level; rather it is important to recognize this variation exists so that it can be further studied. I may dive some more into these specimens, but I would encourage anyone interested to provide useful observations or do further study on these plants, for example:

• Morphological measurements of leaves, surfaces, other characteristics from these plants
• Tracking and adding borderline individuals to the observation set

REFERENCES (to be updated)
Gould 1997 https://www.proquest.com/docview/304373809?fromopenview=true&pq-origsite=gscholar
Hendrickson 1996 https://www.biodiversitylibrary.org/page/9303086#page/98/mode/1up
Correll and Johnston 1970 MVPT

Posted on May 16, 2023 06:11 AM by arnanthescout