Zeltnera texensis and Centaurium species in Texas


Three similar looking species in the Gentianaceae family, Zeltnera texensis, Centaurium pulchellum, and Centaurium teniuflorum likely come in contact in various parts of central and eastern Texas and can be difficult to differentiate due to similar corollas. Previous to 2004, these three species were all in the genus Centaurium and were differentiated with characters like pedicel length and form of the inflorescence, among others (Diggs et al 1999). In 2004 (Mansion 2004), the new world species of Centaurium were segregated into Zeltnera based on the different shape and separation of stigma lobes compared to those of the old world. Unfortunately, these details are rarely captured by observers on iNaturalist, which makes rigorous identification based on these characters unlikely. A less rigorous identification may be done using the characters previously used, based roughly on the key in Shinners & Mahler's Illustrated Flora of North Central Texas (FNCT), substituing C. texense with Z. texensis and C. floribundum with C. tenuiflorum. Features have been collected in the following table from various sources (Holmes and Wivagg 1996, Mansion 2004, Pringle 2010, and Weakley 2022 (based on Struwe and Pringle 2018)). It is incomplete due to the difficulty of accessing more comprehensive works which are typically unpublished, so I have attempted to fill in some gaps in the data with data gleaned from an informal examination of images of herbaria specimens and my own personal experience.

Useful features for identification of Zeltnera texensis and Centaurium species in Texas
Feature Z. texensis C. pulchellum C. tenuiflorum
Stigmaa fan shaped reniform or shoe shaped reniform or shoe shaped
Stylea not divided slightly bifid slightly bifid
Pedicel usually greater than 1/2 as long as calyces, occasionally shorter, 4-14mmb shorter than calycesc (1-)3-5(11)mm d sessile or subsessiled-f, no greater than 2mme*
Inflorescence monochasial helicoid cymea open-spreading compound dichasiumf: diffuse not corymboide dense flat topped umbellate cymef, dense corymboide
Leaves stem leaves linear to lanceolate, 1-3(4)mm widec stem leaves lance-ovate to lanceolate, 2-7mm widec TBD
Height 7-25cm b (5)10-18(29)cmf (19)27-45(55)cmf
Branching typically just below midstem, occassionally near base or upper stemb midstem or belowf upper 1/3 - 1/4 of stemf
a Mansion 2004; b informal examination of 10 specimens from Texas and Missouri from SEInet and personal experience; c Diggs et al 1999 d Weakley 2022; e Pringle 2010, *pedicel length up to 2mm maybe overridden by later treatment in Weakley; f Holmes & Wivagg 1996

The FNCT keyed the three following small flowered species of Centaurium: Centaurium texense, Centaurium pulchellum, and Centaurium floribundum, with a mention of C. tenuiflorum in the text of C. puchellum. Since its publication, Centaurium texense has changed to Zeltnera texensis (Mansion 2004, Pringle 2011) (though with some nomenclatural hiccups) and the specimens treated as Centaurium floribundum from Texas are now treated as Centaurium teniuflorum (Pringle 2010). With the understanding of the name changes, the easiest distinction to be made appears to be between Centaurium teniuflorum and the other species, due to its sessile or subsessile pedicels and tendency to form a dense flat topped inflorescence. Differentiating Z. texensis and C. pulchellum would likely be more difficult from photographs due to the subtle differences in pedicel length and leaf shape and dimensions. However, it is unclear if C. pulchellum and Z. texensis are found in the same areas at this point. Specimens that cannot be clearly identified will likely need to be assigned to subtribe Chironiinae

Historical distributions

The historical distribution of Z. texensis in Texas is unclear. The current BONAP map has a fairly wide distribution of Z. texensis, ranging from the eastern hill country northward through north central Texas almost to the Oklahoma border and also eastward to the Houston area. Additional occurrences seem to be found in parts of Oklahoma, Louisiana, Arkansas and Missouri. The reference for Texas is listed as "TX: 1967. Catholic University of America Biological Series", which is most likely an unpublished dissertation by Dunn dating from 1967 (A revision of the genus Centaurium of continental United States), but I suspect data from other sources may have been included. In contrast, Turner (1993), using samples mostly from Tex/LL, showed a much narrower range concentrated around the Balcones Escarpment and eastern hill country, extending rather weakly to almost disjunct occurrences in Tarrant and Dallas counties. He did not address occurrences outside of Texas. Given the occurrence of Z. texensis mostly on eroding limestone slopes and soils, I suspect Turner's distribution may be more likely., though examination of later collections has added more specimens in north central Texas in the counties around DFW and even along the Red River, though not so much towards the Houston area. Holmes and Wivagg (1996) indicated that C. pulchellum was often misidentified as Z. texensis, so that may account for the wider distribution on BONAP.

Historical distributions of Centaurium pulchellum and Centaurium tenuiflorum in Texas are based mostly on Holmes and Wivagg(1996). In their work, they identified larger, more densely flowering plants as C. muhlenbergii(a species of the west coast United States), indicating that they had previously been misidentified as either C. pulchellum or C. texense. These Texas specimens were for a brief time treated as C. floribundum (Diggs et al 1999) and now have been identified as C. tenuiflorum (Pringle 2010). At the time, C. pulchellum appeared mostly in the southeast portion of the state (Brazos, Galveston, Hardin, Harris, Jasper, Liberty, and Orange counties). C. tenuiflorum (there treated as C. muhlenbergii) was found mostly in a number of counties south of Dallas/Fort Worth and northeast of Austin (Ellis, Falls, Freestone, Hill, Johnson, Leon, Limestone, McLennan, Milam). I have personally seen it in Williamson county, just to the southwest of its reported range and based on soon to be corrected iNaturalist observations in the DFW area, it has also spread northward (see later discussion on survey of iNaturalist observations). Both species have apparently been in Texas since at least the early 1970s (Holmes and Wivagg 1996).

Based on the distributions found in Turner and Holmes and Wivagg, it seems most likely that the ranges of Z. texensis and C. tenuiflorum might overlap. It is less clear if C. pulchellum may have been likely to spread from southeast Texas into the narrow historic range of Z. texensis described by Turner. If so, problematic specimens in that area may have to be assigned to a higher taxon.

Differentiation using the stigmas

Photographing the stigmas of these species is difficult given their small size. Line drawings showing stigma shapes of similar species of Zeltnera and Centaurium can be found in Mansion (2004) and Pringle (2010). Stigmas of Zeltnera species are described as being "fan-shaped" while those of Centaurium are described as being more "shoe" or "iron" shaped.

Z. texensis

C. tenuiflorum

Differentiation based on pedicel length and inflorescence

Differentiation based on pedicel length is most likely on the extreme ends. While the FNCT key indicates that the pedicels are usually nearly as long as the calyces on Z. texensis, the actual length seems to vary, likely from about 1/3 the length of the calyx to being slightly longer than the calyx. It is likely however that at least some and possibly most will be near the length of the calyx. Specimens with shorter pedicels may be difficult to differentiate.

Z. texensis

C. tenuiflorum

Small survey of Z. texensis and Centaurium observations

Out of curiosity of what distinguishing features might be shown in iNaturalist observations of these species, I examined research grade observations for Z. texensis andCentaurium species from Tarrant and Travis counties. As of May 29, 2022, Tarrant county, which is thought to be on the north part of the range in Texas, had a total of 164Z. texensis observations, 115 of those being research grade. Of those, none of the observations included sufficient pictures of the stigma lobes for identification. Only two specimens clearly had a pedicel length consistent with Z. texensis, while numerous ones appeared to have sessile flowers indicative of either C. teniuflorum or C. pulchellum, so most observations identified as Z. texensis appear to have been misidentified. There were only 18 observations of Centaurium species, of which none were research grade, so I did not examine them. Travis county, which is roughly in the heart of the range for Z. texensis, had 146 Z. texensis observations with only 53 marked as research grade. Only one observation (one of mine) had pictures of the stigma sufficient enough for genus ID. In contrast to Tarrant county, most had pedicel lengths consistent with Z. texensis while a few had pedicel lengths consistent with Centaurium species.

I have not yet done a close examination of these species in the southeast part of the state. It is not clear if Z. texensis can currently be differentiated from C. pulchellum using currently available photographic evidence. Intermediate specimens or specimens that don't show the stigmas or pedicels will probably have to be assigned to a higher taxon. Some specimens are being identified as Z. texensis there though. As of May 29, 2022, Harris county had 58 observations of Z. texensis, but the vast majority of similar species were identified as Centaurium (178).

Observations used in this post

C. tenuiflorum : 119653908
Z. texensis : 118706199


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.
Holmes, W. C. and D. C. Wivagg (1996). Identification and distribution of Centaurium muhlenbergii (Griseb.) Piper and C. pulchellum (Sw.) Druce (Gentianaceae) in Louisiana, Mississippi and Texas. Phytologia 80: 23-29.
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)].
Mansion, G. (2004). A new classification of the polyphyletic genus Centaurium Hill (Chironiinae, Gentianaceae): description of the New World endemic Zeltnera, and reinstatement of Gyrandra Griseb and Schenkia Griseb. Taxon 53: 719-740.
Pringle, J. S. (2010). The identity and nomenclature of the Pacific North American species Zeltnera muhlenbergii (Gentianaceae) and its distinction from Centaurium tenuiflorum with which it has been confused. Madrono 57: 184-202.
Pringle, J. S. (2011). Validation of the name Zeltnera texensis (Gentianaceae). Rhodora 113:514-515.
Turner, B. L. (1993). The Texas species of Centaurium. Phytologia 75: 259-275.
Weakley, A. S. and Southeastern Flora Team (2022). Flora of Southeastern United States. University of North Carolina Herbarium, North Carolina Botanical Gardern.


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


0.1 - June 3, 2022 - Draft.
Posted by rymcdaniel rymcdaniel, June 03, 2022 19:10


I appreciate your work on this subject and sharing with iNaturalist!

Posted by suz 4 months ago (Flag)

@suz Thanks. I'll be updating the DFW area ones first because that seems to be where most of the misidentifications are, but I am just going to do a few at a time to ease people into it. So I probably won't link to this for every change. Also, a lot might get left at genus Centaurium because I think there is still some disagreement about whether C. teniuflorum and C. pulchellum are actually separate species. Still looking into that.

Posted by rymcdaniel 4 months ago (Flag)

@rymcdaniel I've been linking it on the observations in which I have participated. I didn't ask permission so I hope that's okay with you.

Posted by suz 4 months ago (Flag)

@suz No worries. I am just trying not to be too heavy handed and repeating it too much. Often the same people are involved in these observations. So far the changes fall mostly into two categories: short pedicels (usually looking sessile) indicate Centaurium rather than Z. texensis or one can't even see the pedicels so it gets raised to subtribe Chironiinae.

Posted by rymcdaniel 4 months ago (Flag)

Thank you so much for putting this journal post together. It is certainly going to be helpful.

Posted by sbdplantgal 4 months ago (Flag)

Ryan, this is yet another magnificent journal post!!! Big time thanks for tackling this.

Posted by sambiology 4 months ago (Flag)

Thanks @sbdplantgal and @sambiology. Hopefully, it will be useful. I am still a bit tentative about identifying Centaurium specimens to species, so it may be a while until we can do that reliably.

Posted by rymcdaniel 4 months ago (Flag)

A couple of interesting qualitative points from looking at this stuff the last couple weeks.

Zeltnera texensis in the DFW area is actually quite rare. I'll probably publish some stats in the coming weeks. Looking at research grade observations only in about 12 counties in the area, all counties except one had from 0 to 2 observations that I could say were Z. texensis. Ellis county was the standout with about 7.

Plant height measurements above for Centaurium tenuiflorum perhaps don't account for extreme drought years. One area at Lake Georgetown where I have seen C. tenuiflorum off and on for about the last 10+ years had plants yesterday that were about half the height I would normally see, this year from about 8-22cm.

Posted by rymcdaniel 4 months ago (Flag)

Thanks for your research and for sharing this journal post. My apologies for contributing to so many mis-IDs in the NTx area.

Posted by postoak 4 months ago (Flag)

Thanks for this! I know you've come by some things I've ID'd wrongly recently so I appreciate the effort to fix this and help us fix it for others!

Posted by oceanicwilderness 4 months ago (Flag)

@postoak No worries. You are in good company. Everyone one up there was getting these wrong. The corollas are so similar it is understandable, and who knew the Centauriums would become so much more common.

Posted by rymcdaniel 4 months ago (Flag)

@rymcdaniel Check this one: https://www.inaturalist.org/observations/123218262. Thanks for the journal post!

Posted by taylorgarrison 3 months ago (Flag)

Add a Comment

Sign In or Sign Up to add comments