rymcdaniel's Journal · iNaturalist

May 16, 2025

Differentiating Zeltnera muehlenbergii from Centaurium and other species in western North America

TLDR (Abstract)
The species Zeltnera muehlenbergii (Griseb.) G. Mansion has long been confused with Centaurium tenuiflorum (Hoffmannsegg & Link) Fritsch, Mitt. (Pringle 2010). This confusion has continued on iNaturalist where most observations identified as Z. muehlenbergii are likely C. tenuiflorum. A review of observations on iNaturalist has revealed characteristics of the corolla which may be used to differentiate the species and raised questions about characteristics and taxonomy of the species.

Introduction

Zeltnera muehlenbergii (Griseb.) G. Mansion is a small flowered member of the Gentianaceae family that is native to the northwest of North America, including California, Oregon, Washington, Idaho, and British Columbia (Pringle 2023b). Due to similarities in flower size and pedicel length, it has historically been confused taxonomically with non-native Centaurium species such as Centaurium tenuiflorum (Hoffmannsegg & Link) Fritsch, Mitt. until Zeltnera was separated from Centaurium by Mansion(2004). The tangled history of several small flowered Centaurium and Zeltnera in north America was explored by Pringle(2010) with a close comparison of Z. muehlenbergii, Z. davyi, and C. tenuiflorum.

In 2022, both C. tenuiflorum and C. pulchellum were being misidentified on iNaturalist as the native Z. texensis in Texas (McDaniel 2022). Though the genera are formally differentiated by aspects of the stigma (Mansion 2004, Pringle 2010, 2023a, 2023b), differences in the form of the inflorescence and pedicel length could often be used to correct those identifications. Given the historical confusion of Z. muehlenbergii and C. tenuiflorum documented by Pringle(2010), I began an initial examination of observations of Z. muehlenbergii on iNaturalist. As there appeared to be no definitive photograph of Z. muehlenbergii at the time, it was unclear how similar live specimens actually were to C. tenuiflorum. Since the literature indicated a much closer morphological similarity with C. tenuiflorum than was the case with Z. texensis, it appeared that photos of the stigma would be necessary for identification of the two species. Given that clear photos of the stigma shape were uncommon and quite difficult to obtain given the small size and perspective needed, it seemed unlikely to be able to sort out the situation with the observations available at the time.

In the summer of 2024, in preparation for an aborted trip to California to search for and photograph Z. muehlenbergii, I began a review of the available online herbarium evidence of Z. muehlenbergii and iNaturalist observations of similar Gentianceae in the known range of Z. muehlenbergii, looking primarily for locations to search but also for any possible differences that may not have been documented in the literature. The herbarium data was problematic, for the most part only revealing a few scant details not already covered in Pringle(2010). The review of iNaturalist observations uncovered a small set of around one hundred observations of small flowered Gentianaceae which appeared consistent with descriptions of Z. muehlenbergii but which had corollas that were visually distinct from C. tenuiflorium, C. erythraea, and Z. davyi. Within this small group, there may be two different corolla color patterns. Further research will be needed to determine if they simply represent variation within Z. muehlenbergii or if one of them might warrant recognition as a separate taxon.

This post is intended primarily as a guide for users of iNaturalist to identify Z. muehlenbergii in order to correct the large number of misidentifications present at this time (May 2025). For the general reader the two sections "A Brief History and Differentiating Characteristics" and "Guide to differentiating species" are probably the most useful. Portions of the section "Methodology and General Results" focus on the herbarium record and so may be of less interest, though there are also subsections specific to iNaturalist. A list of likely Z. muehlenbergii observations is provided towards the end.

A Brief History and Differentiating Characteristics
Methodology and General Results
Guide to differentiating species
List of likely Z. muehlenbergii observations
Moving Forward on iNaturalist
Copyright
References

A Brief History and Differentiating Characteristics

The species now included in the genus Zeltnera, which are native to the new world, were part of Centaurium until they were separated out by Mansion (2004). Early descriptions of these species were typically found in the genus Erythraea until it was later replaced by Centaurium, so many early descriptions of Zeltnera were originally described in one of those two genera. Species now included in the genus Centaurium are mostly native to the old world, though many have been introduced in other areas. Pringle (2010) unravels the historical confusion between Z. muehlenbergii and C. tenuiflorum and addresses to various extents other similar species such as Z. davyi, C. erythraea, and C. pulchellum.

The holotype of Z. muehlenbergii (Douglas s.n., K, barcode K000195655) was collected by David Douglas in California, in the early 1830s, possibly near his base of Monterey but other locations have been suggested (Pringle 2010). It was described by Grisebach as Erythraea muehlenbergii (1839, some sources say 1838). Additional species that were later synonimized with Z. muehlenbergii were collected in the Sacremento Valley of California and also Oregon and Washington throughout the 1800s and into the early 1900s (Pringle 2010). Its documented range extends from Monterey County, California in the south to British Columbia in the north and eastward into Idaho (Pringle 2010, 2023b), though recent reports appear to extend the range southward to San Luis Obispo county, California (Excoff 2025). The holotype of Z. davyi (Davy 396, UCJEPS) was collected in Alameda county, California, in 1893, though earlier samples likely exist. It was originally treated as a variety of C. exaltatum by Jepson(1925) and later elevated to species by Abrams (1951). Regarding Centaurium, the known North American introductions are C. pulchellum, C. erythraea, and C. tenuiflorum (Mansion et al 2005; Pringle 2010,2023a). According to Pringle (2010), the earliest specimens of C. tenuiflorum he encountered come from the Humboldt county area of California in the 1890s and early 1900s, with other early specimens from Butte and Yuba counties in California and Douglas county, Oregon, while specimens from the Bay Area start in the 1940s. It was also introduced to the southern U.S., with earliest samples from that region apparently from the 1960s (Pringle 2010). However, C. tenuiflorum in north America and introduced in much of the world is thought to be of hybrid origin, possibly derived from both C. erythraea and C. tenuiflorum and it is sometimes called C. xtenuiflorum (Mansion et al 2005). It is not known if Pringle's (2010,2023a) descriptions of C. tenuiflorum were based only on North American specimens and thus are limited to the hybrid entity. It is unclear if there are any significant morphological differences between it and C. tenuiflorum in the narrow sense found in parts of the old world.

As carefully explained by Pringle (2010), the root of the historical confusion between Z. muehlenbergii and C. tenuiflorum appears to be an early misapplication of the name C. floribundum to the entity referred to here as C. tenuiflorum. C. floribundum was a synonym of Erythraea floribunda which was described based on a sample collected by Hartweg in the Sacramento Valley around 1846 or 1847(Bentham 1839-). Apparently, throughout much of the 1900s, C. tenuiflorum in the U.S. was incorrectly treated as C. floribundum while Z. muehlenbergii was treated as C. muehlenbergii. However, C. floribundum was eventually synonymized with C. muehlenbergii in the 1980s by N.H. Holmgren, and though the type of C. floribundum may have actually been C. muehlenbergii, many of the specimens which had been identified as such were not. Pringle theorized that this may have led to a situation where descriptions of C. muehlenbergii may have been based more on the increasing number of specimens of C. tenuiflorum than on C. muehlenbergii, as may have been the case with Hickman's descriptions (1993). During this period however, again as noted by Pringle, C. Rose Broome appeared to be the first to make the connection between the specimens previously noted as C. floribundum and C. tenuiflorum and began annotating specimens as the latter from the late 1970s to the early 1990s. However, despite some indications from these annotations that she intended to publish a monograph on Centaurium, it does not appear to have occurred, and thus adoption of C. tenuiflorum in the U.S. was slow to occur. Only with Mansion's work, ultimately in Mansion and Zeltner (2004), was this finally fully recognized.

The main morphilogical differences between Centaurium and Zeltnera as presented by Mansion (2004) are aspects of the stigma lobe, style division, and capsule shape. Pringle (2010) contrasts additional aspects of Z. muehlenbergii, Z. davyi and C. tenuiflorum, which helps with the practical difficulties of using the stigma, style, and capsule. Later treatments (Pringle 2012,2023b) mostly reflect the earlier descriptions, but sometimes differ in subtle yet, important ways. The following tables summarize the morphilogical differences.

Morphological differences between Zeltnera and Centaurium in Mansion (2004)

Feature	Zeltnera	Centaurium
Stigma	flabelliform (fan shaped); no or very weak fleshy surface	reniform to shoe shaped; distinctly thickened to fleshy
Style division	not divided to subcapitate, absent or very inconspicuous	slightly bifid, distinct (2-5mm)
Capsule shape	elliptic to oval	linear

Characteristics for Z. muehlenbergii, Z. davyi and C. tenuiflorum from Pringle (2010,2012,2023a,2023b)

Feature	Z. muehlenbergii	Z. davyi	C. tenuiflorum
Stigma	flabelliform	flabelliform	elliptic, ovate
Style	entire, cleft < 0.5mm	entire, cleft < 0.5mm	cleft 0.5-1mm
Pedicel	proximal flower in center of cyme division, sessile or to 12mm; distal flowers sessile or to 4mm	(2–)4–25(–55) mm	sessile or subsessile to 2mm
Calyx keels	not keeled or proximal only &lt 0.25mm wide	keeled most of their length, proximally 0.3-0.6mm wide	no description
Corolla lobe width	less than 2mm wide	2-3mm wide	no description
Branching	stems usually 1, occasionally 2–4, simple or variously branching	stems 1–10, simple (small plants) or few-branched ± throughout	stems usually 1, simple or branching above or near middle or occasionally with few, slender branches from base
Inflorescence	distally or completely monochasial cymes;open, few flowered, not corymboid	completely monochasial or occasionally proximally dichasial	dense, +- coymboid, dichasial cymes
Calyx lobes	linear-oblong, usually 0.5-0.7mm wide at midlength and nearly parallel sided most of their length,acute to short acuminate at the apex		acicular, tapering gradually from the base, 0.3-0.5mm wide at midlength, and long acuminate apically
Corolla throat	tube flares gradually between summit of ovary and base of lobes	tube flares at ca. 40 deg. almost immediately above summit of ovary	slender neck abrubtly differentiated from limb
Leaves	broader leaves usually limited to lower 1/3	elliptic to ovate leaves over 5mm well into inflorescence

The above table focuses on differentiating aspects. The actual treatments include other details such as size and shape of some features, which are not included here for brevity and clarity.

The stigma shape and associated drawings(Mansion 2004, Pringle 2010) appear to apply to the abaxial view. Pringle (2012) mistakenly includes a drawing of the shape of both Z. muehlenbergii and C. tenuiflorum in the line drawing for Z. muehlenbergii, perhaps left over from an earlier draft before Pringle became involved. Pringle (2010) however does contain the correct drawings.

It is also worth noting that Pringle(2010) went to great care to determine that the holotype of Z. muehlenbergii (Douglas s.n., K, barcode K000195655) was not Z. davyi, indicating that some samples of the two may bear great similarity. He ultimately relied mostly on two features, the size of the calyx lobe keel and the size of the corollas to make that determination. Z. muehlenbergii may apparently contain a smaller, proximal keel, and has smaller, mostly narrower, corolla lobes than does Z. davyi. The presence of the keel in Z. muehlenbergii was omitted in Pringle (2023b), though it is not known if was in error or a purposeful revision.

Muehlenbergii or Muhlenbergii?

According to Pringle(2010), Grisebach (1839) originally named Erythraea muehlenbergii after Gotthilf Heinrich Ernst Mühlenberg, an American botanist of German descent, also known as Henry Ernest Muhlenberg. It is unclear to me which spelling should take precendence. It is apparently customary to write 'ü' as 'ue' when unable to use the umlaut, and since Latin does not contain diacritics, the 'ue' spelling appears appropriate for plant names (Eckel 2023). Many plants named after Muhlenberg, such as the plant genus Muhlenbergia, appear to have dropped the 'e' however. Some references to his name include an umlaut but others do not (Historical Society of Pennsylvania, n.d.). Most synonyms of Z. muehlenbergii seem to have retained the 'e' until Mansion(2004) and Pringle(2010,2012), though numerous annotations prior to then also seem to have dropped it. Pringle (2023b) reverted to having the 'ue' spelling. Not being an expert on the matter, I do not know which is correct. However, as a practical side effect, herbarium specimens have often been annotated in either way, and not all variants of the spelling end up being synonomized in many database systems.

Methodology and General results

This effort was an informal, exploratory one to determine possible locations to look for Z. muehlenbergii and possible differences from C. tenuiflorum and other similar species such as Z. davyi. Research initially focused on online images of herbarium specimens, but encountered significant issues. The bulk of the research eventually was done mostly using iNaturalist observations(iNaturalist 2024, 2025) of Gentianaceae in western North America. Initially, I focused on locations documented by distribution maps in Pringle(2010) as containing Z. muehlenbergii, which focused on California and southern Oregon. I used both the Explore and Identify interfaces of iNaturalist to scan through observation photos, looking for observations which appeared dissimilar from the other observations. After the initial review, I expanded the scope to include all Gentianceae observations in those areas and included locations further north such as Oregon, Washington, and British Columbia. After the initial research provided some possible characteristics useful for identification, I reviewed a small sample of Research Grade Z. muehlenbergii observations to assess their correctness.

The state of online herbarium images for Z. muehlenbergii and similar species

The goal of looking for herbarium images was twofold: to find specific locations where Z. muehlenbergii was known to occur to aid in searching both on iNaturalist and possibly in the field and to find images that might help elucidate the differences between the various species. Two key problems were encountered in this effort: finding correctly identified specimens and insufficient resolution in most images to discern important features, which was not unexpected. Ultimately, these two challenges led me to deprioritize and discontinue the effort, though I am documenting the limited findings here for future reference.

A safe set of specimens
Given the historical confusion between Z. muehlenbergii, C. tenuiflorum and also Z. davyi outlined in Pringle(2010), it seemed important to ensure that any herbarium specimens that were to be used be properly identified, but this became problematic for two reasons. First, neither Mansion nor Pringle provided a list of representative specimens in most of their published works (Mansion 2004, Mansion and Zeltner 2004, Pringle 2010, 2012, 2023a, 2023b), with the only exception being a list of Centaurium specimens collected by Mansion (Mansion et al 2005), which apparently were mostly deposited at NEU and do not appear to be online. The only other specimens they did name mostly came from the lists of types and synonyms, and also a few additional collections Pringle mentioned directly (Mansion 2004, Pringle 2010). The types and synonyms list mostly applies to Z. muehlenbergii, which has a number of historical synonyms, while Z. davyi has only the type(Davy 396, UCJEPS - image not online) and C. tenuiflorum apparently does not have a type(Pringle 2010). Pringle (2010) directly referenced several specimens from UCJEPS, but UCJEPS does not appear to have images or determination history online yet. It seems likely that quite a few of the specimens he used to create his distribution maps may have come from there.

Second, specimens annotated by them also appear either uncommon or unavailable. I have yet to find anything annotated by Mansion. Annotations by Pringle were only found at CHSC, 6 on Z. muehlenbergii, 1 on Z. davyi and 38 on C. tenuiflorum. While he noted using specimens from a number of different herbaria, none of those with a significant number of specimens online like GH, K, and NY have Zeltnera or Centaurium specimens annotated by him online, though it is unclear if they may have been imaged or databased before he worked with them. The resulting number of collections(duplicates were sometimes found) was thus quite small for Z. muehlenbergii(12) and Z. davyi(1) but more moderate for C. tenuiflorum(38).

Types and Synonyms for Z. muehlenbergii based on Pringle (2010)

Name	Collector Info	Location	Herb	Image found	Notes
E. muehlenbergii	Douglas s.n.	California	K	K	Barcodes K000195655 and K000195658 per Pringle(2010) but currently only found under Centaurium genus at K. Likely collected in 1831/1832 somewhere between San Luis Obispo and Sonoma (see below)
E. floribunda	Hartweg 405 (Aka 1832)	Sacramento Valley, California	K,GH,LD,NY	K,GH,LD,NY(2),P	An annotation on one of the NY samples by R. Barneby speculates that it was probably near Marysville, probably based on knowledge of Hartweg's travels
E. tenella	Hall 425, 1871	Oregon	GH,ILL,K,MO,NY	GH,NY(2?),BRU
E. curvistaminea	Suksdorf s.n., 30 July 1885	Falcon Valley, Klickitat Co., Washington	S,BM,GB,GH, H,K,LD,MANCH, MPU,WRSL	S,LD,VT	Originally mislocated by Mansion(2004) and Pringle(2010) as Spokane and Lincoln counties (see below)
E. minima	unknown s.n., June 1883	near Hillsboro, Oregon	ORE	OSC	name misattributed to J.T. Howell by Pringle(2010) but should have been T. or T.J. Howell (Howell 1903)
C. muehlenbergii var. albiflorum	Suksdorf 8903, 20 July 1916	Spokane Co., Washington	WS,BM,CAS,GH, ILL,K,MO,NY, PH,US	CAS,GH,NY

Locations from the safe set of specimens
The number of locations for Z. muehlenbergii resulting from this search was low. Three of the types could only be narrowed to state regions based on collection notes or historical information about the collectors. Two of the types could be traced to counties(Klickitat and Spokane counties in Washington) and one had a municipality (Hillsboro in Oregon). The 6 collections from CHSC were limited to three counties in California: Butte, Colusa, and Sutter. Though some had good locality information, they all appeared to be on private land. One of those, Schlising 3434 (CHSC), was from a site which apparently has since been converted to agriculture (Richvale vernal pools) according to satellite imagery.

Of the locations published for the types and synonyms, the location for the type of E. curvistaminea, Suksdorf s.n. 30 July 1885(S) was incorrect in both Mansion(2004) and Pringle(2010). V. B. Wittrock apparently published the name as part of an exsiccata named Erythraeae exsiccata, and the portion including E. curvistaminea seems to have been announced at the December 15, 1885 meeting of the Botanical Society of Stockholm(Botaniska Sällskapet i Stockholm). A summary of the contents of this exsiccata, including most of the description of E. curvistaminea, was later published in the Swedish Botaniska Notiser and the German Botanisches CentralBlatt in 1886 (Wittrock 1886a, 1886b). The summary description did not include the date or specific locality of collection, only that it was collected by Suksdorf in Washington. However, the original treatment provided in the exsiccata seems to have included more information, noting that it was collected by Suksdorf in Falcon Valley, Washington on July 30, 1885 (Suksdorf s.n. 30 July 1885, S, LD, VT). I have not seen the original exsiccata, but the pages for E. curvistaminea appear to be scanned at S, LD, and VT. It is possible that the original collection information was not widely known due limited availability of the actual exsiccata. Howell (1903) noted that E. curvistaminea occurred in Lincoln county, Washington. Some of the confusion may come from an earlier sample by Suksdorf, which was thought to be collected in Lincoln county. A sample at OSC collected on June 13, 1884, has a hand written label indicating E. curvistaminea and being collected near Duck Lake, Lincoln county. An apparent duplicate, Suksdorf 388 (WS), is annotated as E. douglasii, a synonym of Z. exaltata. Much later, Dunn designated the specimen from S as the lectotype by annotation, though she did not explicitly mention this designation in her dissertation(1967). This may have been why it was missed by Mansion(2004) and Pringle(2010) as neither of them indicated seeing that particular sample, though both indicated that the holotype was at S. Mansion(2004) seems to have conflated the date and location of Suksdorf s.n. 30 July 1885 with the county and collection number of Suksdorf 8903, which was collected in 1916 in Spokane County. Pringle(2010) actually formally designated Suksdorf s.n. 30 July 1885(S) as the lectotype, despite apparently only seeing specimens from BM and GH. He mistakenly placed Falcon Valley in Lincoln county, Washington, perhaps based on Howell(1903). The actual location of Falcon Valley may also have been a source of confusion as it likely was not a widely used name. However, it was apparently located near Suksdorf's home of Bingen in Klickitat county, Washington (Cooke and Shaw 1952). In a bit of botanical irony, the collection label on Suksdorf s.n. 30 July 1885(S) appears to have orginally been annotated by Suksdorf as E. muhlenbergii, but apparently Wittrock changed it first to E. washingtoniana and then finally E. curvistaminea.

The collection location of the holotypes(Douglas s.n., K, barcodes K000195655 and K000195658) was long thought to be near Monterey since Douglas was based there during his time in Calfornia from late 1829 to late 1832 (McKelvey 1955). Pringle(2010) speculated that a more likely location might be near Sonoma, based on his interpretation of the range of Z. muehlenbergii typically being further north, its apparent bloom time (unstated by Pringle), and accounts of Douglas's travels in McKelvey(1956). Several aspects of this interpretation are suspect. First, while Monterey was thought to be at the southern end of the range of Z. muehlenbergii, Pringle did note that collections by Elmer which seemed to suggest intergradation with Z. davyi occurred in the area, and the type is quite similar to Z. davyi. Also, Z. muehlenbergii has been noted more recently as occurring as far south as San Luis Obispo in the recent update to Vascular Plants of San Luis Obispo County by David J. Keil and Robert F. Hoover(Excoff 2025), which seems to be corroborated by recent iNaturalist observations. Second, it is unclear what Pringle's understanding of the bloom time was. Douglas was only known to be near Sonoma in late July and possibly part of August of 1831. While it is difficult to assess flowering time based only on a few verifiable samples that are from a different part of California(specimens in the northern Sacramento Valley annotated by Pringle at CHSC), it seems to range from mid May to mid July with most being in June. In general, iNaturalist observations also seem to be mainly in June. In June of 1831, Douglas was returning from the Santa Barbara area to Monterey while from late June to July he was known to travel north. This bloom time and the expanded geographic range seem to expand the possible range for collection of the type all the way from San Luis Obispo to Sonoma. Lastly, Douglas's travels are only well known in 1831, at least according to McKelvey(1955). He was based in Monterey for much of 1832 until his departure in August, and apparently travelled and worked during that time, though attempting to stay closer to Monterey while waiting for a ship to take him northward.

Identifying specimens from herbarium images
As is common with herbarium images, they can have limited usefulness when visibility of smaller features is necessary for study. Since images are of the whole sheet, smaller features are not usually discernible. The degree to which some features are visible can depend on such imaging factors as resolution, focus, lighting and also on specimen preparation. The image quality on these specimens was variable. There were noticeable differences in resolution, possibly stemming from how early they were scanned. The lowest resolution images were at OSC, 682x1000 pixels, and K, 1150x1600(the holotypes), while those from GH and CAS were the best at roughly 6800x9800 pixels. Images from other herbaria (CHSC,LD,NY,S,VT) were in between these ranges. The low resolution of the Douglas holotypes from K unfortunately obscured any useful details that might have been present, making it difficult to confirm anything other than superficial qualities of the plants such as size and the occasional pedicel length. As for specimen preparation, specimens were usually pressed with corolla lobes closed, obscuring the stigma, and leaves in an upward direction, often obscuring pedicels.

The small data set also introduced a couple of difficulties. First, there was only one specimen of Z. davyi, Triano 204 (CHSC), so it was difficult to generalize about aspects of Z. davyi. Therefore, this data set only allowed comparison of Z. muehlenbergii and C. tenuiflorum. Second, the lack of known good samples from large swaths of the range of Z. muehlenbergii in California made it difficult to assess geographic trends.

The few conclusions to be drawn from this small data set mostly relate to the possibility of identifying the species based on herbarium images. I was unable to confidently identify many of the specimens that Pringle annotated, but that is likely due to the inability to discern features critical for identification from the images. Seeing the specimens in person would probably help.

Larger plants of C. tenuiflorum often identifiable by size and form of inflorescence

C. tenuiflorum

Z. muehlenbergii

Hartweg 405

Ahart 3437

C. tenuiflorum

Smaller plants of C. tenuiflorum often not discernible from Z. muehlenbergii

C. tenuiflorum

Z. muehlenbergii

Pedicels of Z. muehlenbergii sometimes longer than expected

E. tenella

Hall 425

E. curvistaminea

Suksdorf s.n.

Ahart 1775

Ahart 3437

Ahart 3613

Z. muehlenbergii

C. tenuiflorum

Z. davyi

Schlising 3434

Stigmas not visible or discernible
Corolla lobe width difficult to ascertain

Z. muehlenbergii

Z. davyi

Calyx lobes often not discernible
Calyx keels not discernible on the Z. davyi sample

Since the initial investigation yielded few locations and implied that identification of specimens from images might be more time consuming and difficult than expected, I initially did not investigate the herbarium record much further.

Beyond the safe set of specimens
While I mainly focused on specimens that were guaranteed to be correctly identified, I did take a limited look at the state of the herbarium record for Z. muehlenbergii, Z. davyi and C. tenuiflorum for various reasons, including trying to establish whether the eye of Z. davyi was visible in preserved specimens (McDaniel 2025) and also to get a general sense of whether specimens of these taxa could be reliably identified using herbarium images. I used several data portals such as GBIF, SEInet, CCH2, and CPNWH and sometimes individual herbarium portals such as the one for K. While I did attempt to get a sense of the total number of records, I did focus mostly on records with images. Typically UCJEPS had the most records for Z. muehlenbergii and Z. davyi at least, but currently without images, while CAS had the most for C. tenuiflorum in North America, likely due to numerous annotations by C. Rose Broome as discussed below. Synonomies were often mixed and incomplete at the different portals, which made finding all applicable records tricky. GBIF seemed to be the best at aggregating records, but even its synonymy of Z. muehlenbergii appeared incomplete(C. muhlenbergii sometimes matched and sometimes did not). As of early April 2025, GBIF indicated the following number of preserved specimen records without and with images: Z. muehlenbergii (1202,708), Z. davyi (378,200), C. tenuiflorum (356,264). The figures for Z. muehlenbergii are likely an undercount as a significant number of specimens identfied as C. muhlenbergii, though not all, were not getting counted.

As noted earlier, the authors of the most recent works affecting these specimens, Mansion and Pringle, either did not appear to annotate specimens or appear to have done relatively few, and so specimens have generally not been updated to reflect their work. The most obvious attempts at annotation were previously done by two researchers who previously wrote dissertations on Centaurium, Rosalie Dunn (1967) and C. Rose Broome (1973). While their annotations show up at a variety of herbaria, Dunn's annotations show up primarily at GH and Broome's at CAS, the latter on quite a large number of specimens from the 1970s to early 1990s. Both appeared to reference specimens from UC (UCJEPS) extensively, so it is possible they have annotated specimens there as well. From some of Broome's later annotations, it appears she may have planned a monograph on Centaurium, but if so, it does not seem to have been published.

Due to the large number of annotations at CAS, Broome's annotations shape a significant part of the record in two ways. First, as noted by Pringle(2010), she appears to be the first to start annotating north American specimens as C. tenuiflorum. Some specimens from 1978 she annotated as C. floribundum while also comparing to C. tenuiflorum on the annotation (ex. True 40, CAS). Later she annotated specimens as C. tenuiflorum without C. floribundum. Thus she is likely responsible for many of the specimens currently identified as C. tenuiflorum from the U.S. Second, as noted by Pringle(2010), her concept of Z. muehlenbergii appears to have been closer to Pringle's concept of Z. davyi. This may have been a result of viewing the type specimen at K in 1978, which even Pringle(2010) noted was quite similar to Z. davyi and distinguished from it ultimately based on smaller calyx keels and smaller flowers. As a result, numerous specimens of Z. davyi were likely annotated as C. muhlenbergii by her. This seems to be reflected in the relatively low number of specimens identified as Z. davyi at CAS. Unfortunately, though Broome published a number of papers on Centaurium, she does not seem to have published treatments on any of these taxa, so it is not possible to know what her criteria were. Despite these trends, her annotations are sometimes inconsistent based on Pringle's current treatments. Some specimens of C. tenuiflorum she continued to annotate as C. floribundum, and some specimens of Z. muehlenbergii she actually annotated as C. muehlenbergii, in contrast to the many she annotated as Z. davyi. Perhaps a closer examination would reveal an unseen pattern.

Overall, it appeared that there are significant number of specimens annotated as Z. muehlenbergii or a synonym that are likely either C. tenuiflorum or Z. davyi. I picked a pseudorandom set of specimens(92 total) from herbaria with the most images of Z. muehlenbergii and attempted to identify them, with the results as follows.

Taxon of specimens identified as Z. muehlenbergii or synonym

Taxon	Number
Z. muehlenbergii	29
C. tenuiflorum	27
Zeltnera sp.	16
Z. davyi	9
Chironiinae	5
Centaurium sp.	3
Mixed samples	3
Total	92

There were a few additional observations that came from looking at additional specimens of Z. davyi and also additional specimen images with higher resolution and possibly better lighting and focus, mostly from CAS.

Flattened corolla lobes seemed uncommon on specimens of Z. davyi

Z. muehlenbergii

Z. davyi

The prominent corolla eye on Z. davyi is not well preserved in herbarium specimens
Better images sometimes helped identify specimens to genus that might otherwise be identified to Chironiinae

C. tenuiflorum

Z. davyi

Z. muehlenbergii

C. tenuiflorum

Calyx keels still not discernible on better images

Overall, the herbarium record likely needs significant update to be useful for finding locations of Z. muehlenbergii. Most specimens can likely only be tentatively identified to species from images due to lack of visibility of subtle differentiating features such as the stigma shape, calyces, corolla dimensions and pedicels. A significant number of specimens may possibly only be identifiable in person where some of the details can be examined.

iNaturalist Observations of Z. muehlenbergii

After my initial, brief examination of Z. muehlenbergii observations on iNaturalist in 2022, I knew it was unlikely that many observations would show the shape of the stigma lobe, which might be needed to differentiate it from C. tenuiflorum. Additionally, other characteristics noted by Pringle were also unlikely to be shown or at least be difficult to apply. The only possibility was that some undocumented or unexpected difference would appear, though I was prepared for the possibility that seeing the stigma lobes might be the only reliable way to differentiate the species. To ease the process of sifting through thousands of observations, which numbered over 2000 for Z. muehlenbergii alone, I used the gallery style view in the iNaturalist Explore and Identify interfaces to single out specimens of interest for further review instead of closely examining every observation identified as Z. muehlenbergii. I started with the range map for Z. muehlenbergii in Pringle(2010), which was limited to California and southern Oregon, and began looking through observations of subtribe Chironiinae (which includes Zeltnera and Centaurium) in counties in California and southern Oregon where Z. muehlenbergii is known to occur. Later I broadened the search taxonomically to Gentianaceae and geographically further north into Washington, British Columbia, and Idaho. Separately, I did examine all identified observations of Z. davyi to try to understand what identifying features might be visible (McDaniel 2025).

Eventually, I isolated a small set of approximately 100 observations which appeared different from C. tenuiflorum but also seemed to fit the description of Z. muehlenbergii. They ranged geographically from San Luis Obispo County, California in the south to Vancouver Island, British Columbia in the north. In general, specimens from this set can be differentiated from C. tenuiflorum by the corolla color pattern, calyx lobes, and sometimes the appearance of the corolla orifice. They can be differentiated from Z. davyi by the lack of a prominent corolla eye, the presence of which appears to be an undocumented characteristic of Z. davyi (McDaniel 2025).

Though I am confident that most of these observations probably fit within the current description of Z. muehlenbergii, that still needs to be more rigorously confirmed. There is some supporting evidence that at least some of these specimens may be Z. muehlenbergii, but it is often incomplete or circumstantial. For example, none of the observations clearly showed the shape of the stigma lobe, though at least a few observations did attempt to do so. One(F. Watson, iNat ob. 178907160) has a photo and a description consistent with Zeltnera, but does not contain measurements of the corolla lobes, which is important in an area mostly populated by Z. davyi. Some observations from California do seem to clearly correspond in habit with a specimen that Pringle(2010) showed in a figure as being representative of the species (Oswald & Ahart 9267A, CHSC). In particular (H. Hancock, iNat ob. 170202355) from Alameda county shows this and also appears to indicate a corolla lobe width consistent with Z. muehlenbergii. Lastly, the relatively large number of observations from the Uplands Park area on Vancouver Island appear to document one of the populations studied in Canada in furtherance of conservation of the species in British Columbia (Garry Oak Ecosystems Recovery Team 2024; Miles 2002). Those studies present one of the most complete pictures of one of the habitats that this species occupies, so those specimens are most likely correct.

A second issue is that there appear to be aspects of the corolla that don't necessarily appear consistent within the set, raising the possibility that there may be more than one taxon present. Specimens from areas somewhat inland of the California coast and southern Oregon(Type A for the purposes of this post) appear to have a generally different eye pattern and corolla orifice than specimens from British Columbia, Oregon and the California coast(Type B). It is somewhat difficult to reconcile the visual difference, and until I learned that the Uplands Park population on Vancouver Island (consistent with Type B) had been extensively studied I thought they might be a separate species. Their apparently diminutive size, sometimes with longer proximal pedicels, and varying patterns of corolla eye do still seem in stark contrast to some California specimens, particularly some in Marin and Santa Clara counties. However, it is certainly possible that they simply exist at each end of a spectrum for which intermediates have not been observed or recognized. Specimens from coastal California, which appear more aligned to Type B, also occur in proximity to Z. davyi (McDaniel 2025), which raises the possibility of a close relationship to that species. All of these possibilities would require further study.

Z. muehlenbergii Type A with more jagged stark eye, Santa Clara county, California

© Joe MDO (CC BY-NC) (iNat ob. 226058090)

Z. muehlenbergii Type B showing faint or streaked eye, Capital District, British Columbia

© Dan Tucker (CC BY-NC) (iNat ob. 50331884)

Corolla eye and orifice characteristics differentiating Z. muehlenbergii, Z. davyi and Centaurium sp. (illustrated further in Guide section):

Z. muehlenbergii (A) - eye small, white, irregularly protruding, but distinct, into pigmented portion of corolla lobe; orifice usually open and angular
Z. muehlenbergii (B) - eye variable, usually lightened or white laterally, sometimes dark centrally; orifice sometimes open but often more rounded than Type A
Z. davyi - eye prominent, white; corolla orifice appearing variable
Centaurium spp. - eye small white, distinct, often with three nebulous or sharp areas protruding into pigmented portion of corolla lobe; orifice constricted and rounded in C. tenuiflorum

Revisiting Research Grade observations

An examination of research grade specimens after I had identified the set of possible Z. muehlenbergii seems to indicate that, outside of British Columbia, most observations of Z. muehlenbergii were likely in fact Centaurium species, and most likely C. tenuiflorum. In July of 2024 there were approximately 530 research grade observations of Z. muehlenbergii (iNaturalist community 2024), though that number has decreased to 453 as of April 2025 (iNaturalist community 2025) as some identifiers have withdrawn their identifications. A smaller set from 2025 is also available from GBIF (GBIF 2025). I looked at a random sample of 100 of the original 530. Only 3 could be at least tentatively identified as Z. muehlenbergii based on my concept of the species at the time. The remaining 97 were likely C. tenuiflorum, but most would probably be identified to Centaurium due to uncertainty about differentiating C. tenuiflorum and C. erythraea. A small percentage would only be identified to subtribe since distinguishing features may not be visible. In another comparison, of the approximately 100 possible Z. muehlenbergii I found, only 27 were research grade as Z. muehlenbergii, and most of those were on Vancouver Island. It appeared for the most part that people had been misidentifying species of Centaurium (likely mostly C. tenuiflorum) as the apparently uncommon Z. muehlenbergii. Basically, it seemed that, outside of Vancouver Island, when people think of Z. muehlenbergii, they are probably picturing Centaurium. In fact, the default photo for the taxon (A. Kern, photo 14358094, iNat Ob. 10352116) is actually C. tenuiflorum, though that will likely change in the near future.

Guide to differentiating species

The following section provides a guide for differentiating Z. muehlenbergii, Z. davyi, C. tenuiflorum, and to a lesser extent C. erythraea. As the open flowers are most prevalent in photos on iNaturalist, that is the primary focus. However, the calyx lobes (including keels), stigmas and pedicels are also discussed as they can also be helpful in some cases.

Corolla color pattern and orifice

As introduced earlier, two aspects of the corolla can be helpful to differentiate live specimens, the corolla color pattern and corolla orifice. Neither of these aspects are well documented, likely because they are not preserved in herbarium specimens (see McDaniel 2025 for discussion of the corolla eye in Zeltnera). The differences in the color pattern typically concern the base of the corolla lobes. Sometimes there is a clearly white area, extending from the throat, and this differently colored region at the center of the flower is called an eye, which may vary in size or outline. Sometimes there are other patterns. These differences, though often somewhat irregular, can be consistent enough to be useful for differentiation of some species. The color of Zeltnera corolla lobes has been described as ranging from magenta to rose pink (Mansion 2004) or pink to rose-violet (Pringle 2023b), and may also include all white lobes in many species. Centaurium are often similarly colored. Z. muehlenbergii appears to have two different general patterns, which I have split into types A and B as discussed earlier and below. Centaurium in North America appear to have small, variously shaped white eyes while Z. davyi has a prominent white eye. These are described in detail in the table and subsections below.

The second difference in the appearance of the corollas is the appearance of the corolla orifice, which can differ likely due to the structure of the corolla tube and how it transitions to the lobes, which was noted by Pringle (2010) mostly in relation to Z. muehlenbergii and Z. davyi. In Z. muehlenbergii there is only a very gradual widening from the narrow point in the corolla tube to the lobes, so that the orifice occurs at the abrupt transition from the distal part of the tube to the lobes. As there are five lobes, the orifice often appears pentagonal, or square when there are only four, though visibility to the this phenomenon may be obscured with smaller specimens. However, in C. tenuiflorum the distal part of the tube flares open so that the constricted, circular part of the tube is visible, so that the orifice does indeed appear more constricted and circular. In comparison, the orifice of Z. muehlenbergii often appears more spacious. The tube of Z. davyi also flares distally but the appearance, rounded versus angular, seems more variable, perhaps affected by the larger eye. Though types A and B of Z. muehlenbergii seem to have slight differences in the appearance of the orifice, there may ultimately be no correlation.

Corolla differences

Taxon	Lobe Color Pattern	Orifice
Z. muehlenbergii(A)	Colored lobes with small white eye either barely protruding to the lobes or an oblong projection centrally at the base of the lobe	appearing spacious, outline pentagonal or square
Z. muehlenbergii(B)	Colored lobes with basal area laterally white or lightened, intersected by color and sometimes darkened at base, sometimes just diffusely white	appearing somewhat open, outline rounded or sometimes angular
Centaurium spp.	Colored lobes with small white eye consisting of some form of 3 prongs, either sharp or nebulous	appearing tight and rounded in C. tenuiflorum, less so in C. erythraea
Z. davyi	Colored lobes with prominent white eye, rounded pentagonal or star shaped	NA

Z. muehlenbergii type A

Type A appears to have an irregular small white eye, but the border between white in the throat and base of the lobes is distinct compared with the colored distal part of the lobes. The orifice also appears spacious and angular, forming either a pentagon or square. Most observations of this type appear around the Bay area, in the south in Santa Clara county and in the north in Marin county around Mount Tamalpais. I am not knowledgeable about the local conditions, but they seem to appear in what may be at least slightly drier areas and also sometimes areas noted for serpentine (near trails called Serpentine trail or also hidden in populations of Z. trichantha further north).

There are unfortunately few photos currently available that illustrate this feature without zooming in. One can do this by accessing the original photos on iNaturalist.

Z. muehlenbergii with very small white eye with distinct jagged border and spacious orifice, Alameda county, California

© Helen Hancock (CC BY-NC) (iNat Ob. 170202355)

Z. muehlenbergii with small white eye projecting into colored lobes (see large image here), Santa Clara county, California

Z. muehlenbergii type B

Type B seems to have more variation in the pigmentation pattern than Type A. The only pattern appears to be that the base of the corolla lobes is lighter laterally (or white) than centrally. The central part can vary from almost white to the same color as the distal parts of the lobe to noticeably darker at the base. This all seems to follow a pattern of being lighter laterally and darker centrally at the base of the lobes. The orifice in these specimens, though more spacious than in Centaurium species, appears a bit more rounded than in Type A. Most observations of this type currently are in the Capital District of British Columbia with sporadic occurrences in Oregon and mostly coastal California. They seem to occupy areas either known for being wet or even sandy areas. Pedicels, especially proximal ones, often appear quite long, though apparently consistent with early collections in Oregon and Washington often documented as being wet.

Z. muehlenbergii lobes lightened laterally at base, Capital district, British Columbia

© Dan Tucker (CC BY-NC) (iNat Ob. 50331884)

Z. muehlenbergii lobes lightened laterally and darkened centrally at base, Marin county, California

C. tenuiflorum and C. erythraea

Centaurium specimens in North America reliably seem to have a small white eye. On each corolla lobe, the white pattern is usually some variation of being three pronged. In the case of C. tenuiflorum, the prongs tend to be more nebulous and irregular, often producing the appearance of a cog when viewed overall, but in some cases the prongs are more well defined and the appearance may be more pentagonal or star like. In Texas, some specimens have larger lateral prongs, which produces a different pattern. In C. erythraea and C. pulchellum, the prongs seem to be more defined, but this area needs more research. There may not be a reliable difference between the eyes of these Centaurium species to differentiate them on that basis.

Differentiating C. tenuiflorum and C. eyrthraea from each other unfortunately relies on the presence or size of a basal rosette and the size of the corolla lobes (Pringle 2023a), two features which are rarely documented in iNaturalist observations. In my personal observations of C. tenuiflorum in Texas, I have sometimes found dried basal leaves that might be interpreted as a rosette, so it is unclear how to evaluate the sufficiency of the basal leaves even when they are present. The corollas of C. erythraea often do appear larger in photographs, with the lobes overlapping to some degree, but without measurements it is again difficult to determine where the border is. There may yet be more definitive visual cues to help differentiate these species, but I have not yet encountered them. On iNaturalist, it is likely the C. erythraea is restricted to more northerly areas such as the far north of California, Oregon, Washington, and British Columbia. C. tenuiflorum is likely more common in California around the Bay Area and Sacramento Valley, but many Centaurium specimens from Oregon often seem to appear quite similar and probably cannot be identified to species without closer examination.

Specimens of C. tenuiflorum also typically have a more narrow, rounded orifice than in Z. muehlenbergii. This is probably accentuated in most photos since that part of the tube is typically in shadow. This seems to less often be the case in C. erythraea perhaps due to differences in the openness of throat, but the eye is still present. In C. erythraea, the shape of the throat is often interrupted by visibility of the stamen filaments attaching to the corolla tube.

C. tenuiflorum with small, rounded, orifice and nebulous 3 pronged eye

© Ron Vanderhoff (CC BY-NC) (iNat Ob. 226338560)

C. tenuiflorum with small, rounded, orifice and nebulous 3 pronged eye

Z. davyi

As I discussed recently (McDaniel 2025), Z. davyi likely contains a prominent white eye, which appears to have gone undocumented and can be used to distinguish it from Z. muehlenbergii. Though this still needs to be proven with live specimens, I am confident that this feature is present most of the time. The appearance of the orifice seems to vary somewhat in Z. davyi, but it is less important given the presence of the eye.

Z. davyi corolla with prominent eye, Monterey county, California

by Irene (aparrot1) (iNat Ob. 118970955)

Z. davyi corollas with prominent eye, San Luis Obispo county, California

by Alex Heyman (iNat Ob. 24891586)

Calyx Lobes and Keels

In photographs, the calyx lobes can be helpful for distinguishing Z. muehlenbergii from Centaurium, but not as useful for distinguishing Z. muehlenbergii from Z. davyi. While Pringle (2010) does provide a description of the differences of the calyx lobes between Z. muehlenbergii and C. tenuiflorum, some aspects of the descriptions seem self contradictory and others seem to contradict the drawings that were provided, so it is unclear what exactly is trying to be expressed. For example, he describes the calyx lobes of C. tenuiflorum both as acicular and tapering gradually from the base, which seems contradictory unless acicular is being used in fashion I would not expect. The drawing seems to be more acicular (needle-shaped), whereas photos I have taken (see below) seem to be consistent with tapering from the base.

Regardless of the exact meaning of Pringle's descriptions, the calyx lobes do appear different in photos, if only subtlely so. In Z. muehlenbergii, the calyx lobes appear conspicuous, often covering much of the corolla tube, probably ridged or triangular in cross section or terete for really small specimens. In C. tenuiflorum, the calyx lobes are usually fairly inconspicuous, often difficult to separate from the tube, and likely more flat in cross section, though I have seen exceptions. It is possible that these differences are not preserved well in pressed specimens.

Z. muehlenbergii calyx lobes, Contra Costa county, California

© Ryan McDaniel (All Rights Reserved) (iNat Ob. 238274255)

The calyx lobes are likely not as useful for distinguishing Z. muehlenbergii from Z. davyi, at least in photos, for several reasons. The first issue is that there are some discrepancies concerning keeled calyx lobes in accounts of both species. The most significant discrepancy is whether the calyx lobes of Z. muehlenbergii may actually be keeled or not. Pringle (2010) indicates that Z. muehlenbergii may sometimes have weakly developed keels, confined to the proximal part of the lobe, no greater than 0.25mm in width. However, Pringle (2023b), ommits this detail, and it is unknown if Pringle changed his opinion on Z. muehlenbergii or if it was left out for some other reason. Mansion (2004), while mentioning the keeled calyx lobes of Z. davyi, described those of Z. muehlenbergii as ridged, and it may be that there is just a very subtle difference when the keels are not as prominent. There were additional details and subtle discrepancies in Pringle's descriptions of Z. davyi. Pringle (2010) describes the keels as generally occurring for much of the lobe length and being 0.3-0.6mm wide, but that this may only apply to the outer calyx lobes. Pringle (2023b) slightly changes the description of the calyx lobes for Z. davyi, indicating they are keeled "proximally 0.3–0.6 mm wide", and ommits any differences between the inner and outer calyx lobes. Overall, these discrepancies leave the impression that Z. muehlenbergii may sometimes be more keeled than previously thought and Z. davyi may sometimes be less keeled, and that the line between the two may not be so distinct in borderline cases, especially when dealing sub millimeter differences. In practice, photos of the calyx lobes of Z. muehlenbergii sometimes do appear weakly keeled as in the photo above.

The second issue is that photographing the keels, even when present, is likely quite difficult, requiring the proper angle and lighting to highlight the change in width from the adaxial part of the lobe to the abaxial part. Though most photos of Z. davyi do show prominent calyx lobes, it is very difficult to discern whether the lobes are keeled or sharply ridged. The photo below, which is rare in that it actually shows a keeled calyx lobe, illustrates these difficulties. While one can see the keel on the central calyx lobe (running horizontally) due to shading, the feature is not necessarily discernible on other lobes of the same plant.

Z. davyi keeled calyx, San Mateo county, California

Stigmas

One of the primary morphilogical differences between Zeltnera and Centaurium species is the stigma shape, as outlined in Mansion (2004). Though the drawings there do not include Z. muehlenbergii, Z. davyi or C. tenuiflorum, they do include similar species(e.g. C. erythraea) and general trends. Pringle (2010) provides line drawings of the stigma shape of Z. muehlenbergii and C. tenuiflorum, though Pringle (2012) mistakenly includes the drawings for both species under the drawing for Z. muehlenbergii (C. tenuiflorum on the left, Z. muehlenbergii on the right), perhaps a remnant from an earlier draft when the two species were considered conspecific. The drawings are of the abaxial side of the stigma lobe, when the stigma lobes are typically in a more erect, ascending position. The stigma of C. tenuiflorum is described as elliptic (Pringle 2023a) and Z. muehlenbergii fan shaped (Pringle 2023b).

Unfortunately, photographing the stigma shape can be difficult because of the small size and specific perspective required to show the abaxial view. The small size usually requires equipment with macro capability or magnification. I use a 100mm macro lens on an APS-C format DSLR, but other equipment, including some more recent smart phone cameras, may be able to take photos sufficient for identification. Getting the right perspective usually requires handling and orienting the flower at a minimum. An additional complication is that on some specimens of Z. muehlenbergii the stigma lobes may barely exceed the corolla tube. While I have seen some photos of the stigma shape of C. tenuiflorum from California, captured either on accident or on purpose, I have yet to find a clear abaxial view of the one from Z. muehlenbergii. Therefore, I am providing a photo of one from Z. texensis, which appears similar to the line drawing found in Pringle(2010).

Z. texensis stigma lobe (abaxial view) perhaps similar to Z. muehlenbergii

© Dan Tucker (CC BY-NC) (iNat Ob. 50331884)

Another possible difference with the stigma lobes is that they may appear different in profile, though this is not documented. From some of the photos I have seen, it seems likely that the stigma lobes of both Z. davyi and Z. muehlenbergii may appear club shaped in profile, with the stigmatic surface oriented mostly upward on the dorsal surface, similarly to that of Z. calycosa shown below. In contrast, the stigmatic surface in C. tenuiflorum drapes down around the edges, giving the impression of being more like an arrow head.

Z. calycosa stigma lobe (profile view) perhaps similar to Z. davyi and Z. muehlenbergii

© Ryan McDaniel (All Rights Reserved) (iNat Ob. 168399955)

Z. muehlenbergii stigma from an angle, suggesting club shape

© Ryan McDaniel (All Rights Reserved) (iNat Ob. 238275714)

Though both Mansion (2004) and Pringle (2010) note differences in the degree of division of the styles as a possible differentiating feature, the above photos of C. tenuiflorum show the possible difficulties with using that chacteristic. The stigma lobes are often appressed and in this state it is difficult to discern if the styles are divided or not.

Pedicels

Pedicel length can sometimes be a helpful feature if it can be determined reliably. One issue with determining pedicel length is that sometimes small bracts subtending the calyx can be mistaken for calyx lobes, and when this happens branches can be misinterpreted as longer pedicels. This is especially problematic with C. tenuiflorum, which often has only a singular bract that blends in with the calyx due to proximity and similar size. Additionally, sometimes, as there may be only one, they may be obscured by the stem itself. However, if one can safely determine that there are longer pedicels (greater than 2mm) on the plant, then C. tenuiflorum and C. erythraea can be ruled out as possibilities.

Using the pedicels to differentiate Z. muehlenbergii and Z. davyi can be difficult as usually one needs to establish the general trend rather than simply looking at one pedicel as both species can have relatively short pedicels and somewhat longer ones. Most observations rarely show more than a single one clearly.

List of likely Zeltnera muehlenbergii observations

The following is a list of around 100 likely Z. muehlenbergii observations as February 2025. Some may not have enough information for a species level identification.

iNat ID	State	County/District
47317284	British Columbia	Capital
50331884	British Columbia	Capital
50479832	British Columbia	Capital
51522796	British Columbia	Capital
51522815	British Columbia	Capital
51702230	British Columbia	Capital
51702235	British Columbia	Capital
51763760	British Columbia	Capital
52532798	British Columbia	Capital
53011427	British Columbia	Capital
125735529	British Columbia	Capital
125735530	British Columbia	Capital
125914770	British Columbia	Capital
165438746	British Columbia	Capital
169468753	British Columbia	Capital
191019535	British Columbia	Capital
224948059	British Columbia	Capital
224948066	British Columbia	Capital
225097880	British Columbia	Capital
225097959	British Columbia	Capital
225112111	British Columbia	Capital
226035262	British Columbia	Capital
232078053	British Columbia	Capital
81333390	British Columbia	Nanaimo
81333393	British Columbia	Nanaimo
226595322	British Columbia	Nanaimo
170202355	California	Alameda
172236450	California	Colusa
49905958	California	Contra Costa
220969848	California	Contra Costa
26447706	California	Lake
169619161	California	Lake
1428758	California	Marin
3689410	California	Marin
6952045	California	Marin
28851716	California	Marin
29692736	California	Marin
48853399	California	Marin
51979621	California	Marin
80696130	California	Marin
81808246	California	Marin
83908760	California	Marin
85202915	California	Marin
112734244	California	Marin
112843231	California	Marin
118283065	California	Marin
167483440	California	Marin
167484511	California	Marin
168699150	California	Marin
171329527	California	Marin
173116824	California	Marin
173368345	California	Marin
180398634	California	Marin
222138030	California	Marin
230018760	California	Marin
225212437	California	Mendocino
228410903	California	Modoc
178907160	California	Monterey
167640967	California	Napa
158984218	California	Sacramento
120104459	California	San Benito
27477275	California	San Luis Obispo
29678896	California	San Luis Obispo
48923562	California	San Luis Obispo
225589971	California	San Luis Obispo
6412484	California	San Mateo
51636780	California	San Mateo
164358220	California	San Mateo
168085147	California	San Mateo
220765251	California	San Mateo
223270062	California	San Mateo
27195395	California	Santa Clara
27294255	California	Santa Clara
163928042	California	Santa Clara
166117585	California	Santa Clara
168210143	California	Santa Clara
171274683	California	Santa Clara
178955957	California	Santa Clara
221029742	California	Santa Clara
221202496	California	Santa Clara
222680887	California	Santa Clara
226058090	California	Santa Clara
226066237	California	Santa Clara
3150046	California	Santa Cruz
3293061	California	Santa Cruz
6791568	California	Santa Cruz
12941949	California	Santa Cruz
47234072	California	Santa Cruz
47358361	California	Santa Cruz
50300534	California	Santa Cruz
119487187	California	Santa Cruz
169528011	California	Santa Cruz
221658101	California	Santa Cruz
221751755	California	Santa Cruz
12700723	California	Sonoma
47548217	California	Sonoma
165235896	California	Sonoma
219667177	California	Sonoma
216216116	California	Tehama
234047614	Idaho	Owyhee
222566425	Oregon	Columbia
1640104	Oregon	Coos
224086462	Oregon	Josephine
225779623	Oregon	Josephine
227139720	Oregon	Josephine
124135889	Oregon	Lane
126752846	Oregon	Lane
225835132	Oregon	Linn

Moving Foward on iNaturalist

As of this publication (May 2025), I have not yet started changing identifications of most research grade specimens of Z. muehlenbergii, though I plan to start doing so soon after posting this. The default taxon photos of Z. muehlenbergii, which currently are actually C. tenuiflorum, also will need to be changed. As most current observations of Z. muehlenbergii are actually Centaurium, this may result in removal of Z. muehlenbergii from the iNaturalist computer vision model at some point in the near future until enough observations are accumulated to add it back in, which is at least partly why I have included the above list of likely observations so that others my also participate in identification.

It is also my hope that this post will enable observers to better document these taxa in the future, hopefully providing further evidence as to whether the observations treated here as Z. muehlenbergii are in fact that taxon or whether they may represent more than one taxon. The west coast of North America is unfortunately a long way from Texas, so I don't know if I will be able to document these species or not myself.

This post may be revised in the future as needed.

Copyright

Unless otherwise noted, text and photos used in this post are the property of Ryan McDaniel, all rights reserved. Thank you to all those who have observed these taxa and whose photos I have used in the post.

iNaturalist Photo	iNaturalist Observation	Description	Creator/Copyright holder	Licence
400861188	226058090	Z. muehlenbergii showing more jagged stark eye, Santa Clara county, California	Joe MDO (joemdo)	CC BY-NC
79872047	50331884	Z. muehlenbergii showing faint or streaked eye, Capital District, British Columbia; Z. muehlenbergii stigmas barely exceeding corolla orifice	Dan Tucker (dbltucker)	CC BY-NC
295168890	170202355	Z. muehlenbergii with distinct jagged border and spacious orifice, Alameda county, California	Helen Hancock (helenh)	CC BY-NC
290145407	167484511	Z. muehlenbergii corolla orifice, Marin county, California	Krissa Klein (graysquirrel)	CC BY-NC
401401076	226338560	C. tenuiflorum with small, rounded, orifice and nebulous 3 pronged eye	Ron Vanderhoff (ronvanderhoff)	CC BY-NC
396754300	223892634	C. tenuiflorum with small, rounded, orifice and nebulous 3 pronged eye	Erica Fleniken (ericalsf)	CC BY-NC
297176554	171270264	C. erythraea, Douglas county, Oregon	Morgan Stickrod (morganstickrod)	CC BY-NC
201068337	118970955	Z. davyi corolla with prominent eye	Irene (aparrot1)	No rights reserved
38503346	24891586	Z. davyi corollas with prominent eye	Alex Heyman (arheyman01)	No rights reserved
79237160	49905958	Z. muehlenbergii calyx lobes	Steve Beatty (steve-beatty)	CC BY-NC
424183073	238274255	C. tenuiflorum calyx lobes	Ryan McDaniel (rymcdaniel)	All Rights Reserved
404357447	227902612	Z. davyi calyx keels	Cat Chang (catchang)	CC BY-NC
200593025	118706199	Zeltnera texensis stigma abaxial view	Ryan McDaniel (rymcdaniel)	All Rights Reserved
202295989	119653908	Centaurium tenuiflorum stigma abaxial view	Ryan McDaniel (rymcdaniel)	All Rights Reserved
291839238	168399955	Z. calycosa stigma lobe (profile view)	Ryan McDaniel (rymcdaniel)	All Rights Reserved
290143690	167483440	Zeltnera muehlenbergii stigma from an angle	Krissa Klein (graysquirrel)	CC BY-NC
424182640	238274255	C. tenuiflorum appressed stigma lobes	Ryan McDaniel (rymcdaniel)	All Rights Reserved
424185048	238275714	C. tenuiflorum separated stigma lobes and styles	Ryan McDaniel (rymcdaniel)	All Rights Reserved

References

Abrams, L. R. 1951. An Illustrated Flora of the Pacific States Washington, Oregon, and California, Vol 3. Gerianaceae to Scrophylariaceae/geraniums to figworts. Stanford University Press, Stanford, CA. https://doi.org/10.5962/bhl.title.4657.
Bentham, G. 1839-?. Plantas Hartwegianas :imprimis mexicanas adjectis nonnullis Grahamianis enumerat novasque. G. Pamplin, London, U.K. https://doi.org/10.5962/bhl.title.437.
Broome, C. R. 1973. Systematics of Centuarium (Gentianaceae) of Mexico and Central America. Ph.D. Thesis, Duke University, Durham, North Carolina.
Cooke, W.B. and C.G. Shaw. 1952. The Suksdorf Fungus Collections. Research Studies of the State College of Washington. Vol XX, no 4.
https://rex.libraries.wsu.edu/esploro/outputs/report/The-Suksdorf-Fungus-Collections/99901025838801842#file-0
Dunn, R. A. 1967. A Revision of the Genus Centaurium of the Continental United States. Ph.D. Thesis, The Catholic University of America, Washington, D.C, USA.
Eckel, P.M. 2023. A Grammatical Dictionary of Botanical Latin. https://www.mobot.org/mobot/latindict/keyDetail.aspx?keyWord=umlaut
Excoff, P. 2025. iNaturalist observation: https://www.inaturalist.org/observations/27477275. Accessed on 16 Feb 2025.
Garry Oak Ecosystems Recovery Team. 2024. SAR factsheet Zeltnera Muhlenbergii. https://goert.ca/wp/wp-content/uploads/SAR-factsheet-zeltnera-muehlenbergii.pdf
GBIF.org (24 February 2025) GBIF Occurrence Download https://doi.org/10.15468/dl.s4s72n
Grisebach, A.H.R. 1839. Genera et Species Gentianaearum Adjectis Observationibus Quibusdam Phytogeographycis. J. G. Cotta, Stuttgart and Tübingen. https://doi.org/10.5962/bhl.title.51527.
Historical Society of Pennsylvania. n.d. Biographical/Historical Note in Henry Ernest Muhlenberg papers. https://www2.hsp.org/collections/manuscripts/m/Muhlenberg0443.html. Accessed February 21, 2025.
Hickman, J.C. 1993. Centaurium in J.C. Hickman (ed.) The Jepson Manual: higher plants of California. University of California Press, Berkeley, CA.
Howell, T. 1903. A Flora of Northwest America. Portland, OR. https://doi.org/10.5962/bhl.title.21101.
iNaturalist community. 2024 - 2025. Gentianaceae of California, Oregon, Washington, Idaho, and British Columbia. accessed 2024-2025.
iNaturalist community. 2024. Research Grade observations of Zeltnera muehlenbergii. Exported on 30 July 2024.
iNaturalist community. 2025. Research Grade observations of Zeltnera muehlenbergii. Exported on 6 May 2025.
Jepson, W. L. 1925. A Manual of the Flowering Plants of California. Associated Students Store, University of California, Berkley, CA. https://www.biodiversitylibrary.org/bibliography/176756.
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.
Mansion, G. & L. Zeltner. 2004. Phylogenetic Relationships within the New World Endemic Zeltnera (Gentianaceae-Chironiinae) Inferred from Molecular and Karyological Data. American Journal of Botany 91:2069-2086.
Mansion, G., Zeltner, L., & Bretagnolle, F. 2005. Phylogenetic patterns and polyploid evolution within the Mediterranean genus Centaurium (Gentianaceae - Chironieae). Taxon 54(4): 931-950.
McDaniel, R. T. 2022. Zeltnera texensis and Centaurium species in Texas. https://www.inaturalist.org/journal/rymcdaniel/66657-zeltnera-texensis-and-centaurium-species-in-texas.
McDaniel, R. T. 2025. Zeltnera davyi and its undocumented eye. https://www.inaturalist.org/journal/rymcdaniel/106865-zeltnera-davyi-and-its-undocumented-eye.
McKelvey, S. D. 1955. Botanical exploration of the trans-Mississippi West 1790-1850. The Arnold Arboretum of Harvard University, Jamaica Plain, MA. Courtesy of HathiTrust, https://www.hathitrust.org/.
Miles,W. 2002. Stewardship account for Muehlenberg's Centaury. https://goert.ca/wp/wp-content/uploads/steward-account-zeltnera-muehlenbergii.pdf
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. 2012. Zeltnera, in Jepson Flora Project (eds.) Jepson eFlora, https://ucjeps.berkeley.edu/eflora/eflora_display.php?tid=81825, accessed on August 19, 2024.
Pringle, J. S. 2023a. Centaurium. In: Flora of North America Editorial Committee, eds. 1993+. Flora of North America North of Mexico. 20+ vols. New York and Oxford. Vol. 14. http://www.efloras.org/florataxon.aspx?flora_id=1&taxon_id=106016
Pringle, J. S. 2023b. Zeltnera. In: Flora of North America Editorial Committee, eds. 1993+. Flora of North America North of Mexico. 20+ vols. New York and Oxford. Vol. 14. http://www.efloras.org/florataxon.aspx?flora_id=1&taxon_id=318122
Wittrock, V. B. 1886a. Erythraeae exsiccatae Fasc. II. Botaniska Notiser 1886: 65-71. https://www.biodiversitylibrary.org/bibliography/6050.
Wittrock, V. B. 1886b. Erythraeae exsiccatae Fasc. II. Botanisches Centralblatt (1886) 26: 315-319. https://www.biodiversitylibrary.org/bibliography/6954.

Revisions

0.1 - May 16,2025 - Draft.

Posted on May 16, 2025 05:50 PM by

rymcdaniel | 9 comments | Leave a comment

February 21, 2025

Zeltnera davyi and its undocumented eye

TLDR (Abstract)
The species Zeltnera davyi (Jeps.) G. Mansion likely has a prominent, white, pentagonal corolla eye, which has not been documented and may differentiate it from Zeltnera muehlenbergii (Griseb.) G. Mansion. A number of other Zeltnera species may also have undocumented eyes. However, the lack of relevant identifying characteristics present in photographic evidence on iNaturalist prevents a rigorous correlation of the current species description (e.g. Pringle 2023) with the presence of the eye.

It's always surprising to find an undocumented aspect of a plant, even more so when it seems quite obvious. Such was the case in the summer of 2024 when I realized that the literature on Zeltnera seemed to completely overlook the corolla eye ("the marked or contrastingly colored center of a flower" (Diggs et al. 1999)) of Zeltnera davyi (formerly Centaurium davyi). I had been researching California Zeltnera for a couple of months in hopes of sorting out the confusion between Zeltnera muehlenbergii and Centaurium tenuiflorum on iNaturalist (post in the works), and thus became familiar with some of the other California species, such as Z. davyi, Z. trichantha and Z. venusta, that are thought to be sympatric with Z. muehlenbergii.

As part of this research, I examined all of the verifiable observations identified as Z. davyi at the time(464) at least once, though upon further review I missed a few due to some quirks in the operation of iNaturalist's Identify interface. I checked, among other things, what differentiating features might be visible. It quickly became clear that most of the features used to distinguish Z. davyi from other species in the literature(Mansion 2004; Pringle 2010,2012,2023) such as keeling in the calyx lobes, pedicel length, and corolla dimensions among other things, were not discernible or documented in most observations. However, the vast majority of them did have a fairly regular color pattern on the corolla with an eye composed of a distinct, prominent white region at the base of the usually red or pink corolla lobes. It was so prevalent that I did not even bother counting the number of times it did not occur. Only white corolla specimens and a few intermediate specimens which remain difficult to identify appear to lack this. I assumed this eye was a known aspect of Z. davyi, but upon reexamining the literature found that it was never mentioned, in addition to finding that the eyes of other Zeltnera were rarely described as well.

This post primarily is an exploration of the concept of the corolla eye in Z. davyi, whether the photographic evidence present on iNaturalist is enough to conclude that Z. davyi has an eye, and whether it can be used as a distinguishing feature from Z. muehlenbergii. It touches on various topics like the concept of a corolla eye, the history of the genus, the difficulties of using photographic evidence for identification, and also the possible usefulness of using the eye concept more broadly in the genus Zeltnera. Ultimately, while the current photographic evidence on iNaturalist very strongly indicates that Z. davyi usually has a prominent eye, it may not be sufficient to correlate the traditionally documented morphological aspects of the species with the presence of an eye, so examination of fresh specimens will likely be needed to make that correlation more rigorous.

Z. davyi corolla with conspicuous eye

by Irene (aparrot1) (iNat Ob. 118970955)

The concept of a corolla eye and its application in Zeltnera

The concept of a corolla eye, though sometimes used in descriptions of a number of taxa, is not necessarily clearly defined. At least some botanical glossaries do not contain a definition of it(e.g. Jepson Flora Project 2025; Harris and Harris 2001). Two Texas floras do contain a definition. Correll and Johnston (1970) define an eye as "the marked center of a flower" while Diggs et al. (1999) adds "the marked or contrastingly colored center of a flower." In relation to Centaurium, the concept of an eye seems to have first been introduced by Broome (1973), where she mentions "Some species ... have a small white band, "star" or "eye", at the base of the corolla lobes".

The earliest descriptions of species of Zeltnera (as Erythraea and then Centaurium) usually only indicated the reddish color of the corollas (Grisebach 1838, 1845) for each taxonomic section, not even differentiating between parts of the corolla such as the tube or lobes. Gray (1876) seems to be the first to note something similar to an eye in his description of Erythraea venusta where he noted a yellow center (the yellow green throat is usually visible in Z. venusta in southern California). Jepson(1925, 1939) later described C. venustum as having a white throat with red spots, and a white throat was added for C. trichanthum and yellow throat for C. muehlenbergii (which later authors treated as C. davyi) in 1939. Abrams (1951) contained similar descriptions for C. venustum and C. trichanthum, but introduced C. davyi as a separate species with a yellow throat. Broome's description of the white band at the base of lobes appears to not only have introduced the eye concept to Centaurium but also appears to be the widest application of it as she noted its occurrence in five species (C. venustum, C. calycosum, C. trichanthum, C. wigginsii, and C. capense). She differed with Jepson and Abrams in noting that the white band was at the base of the corolla lobes and not in the throat, which is part of the tube. She probably had more insight into the appearance of a broad range of fresh Centaurium specimens since she raised numerous species for research. The eye, however, has not been well documented even after her work. Descriptions of Z. calycosa (which Broome noted as having an eye) in Texas treatments (Correll and Johnston 1970, Diggs et al. 1999, Turner 1993) neglect to mention it at all. Even the more recent treatments by Mansion(2004) and Pringle(2023) mention the eye primarily in the genus description, but only specifically calling it out for Z. venusta.

The lack of documentation of the eye is likely due to a reliance on pressed material. Pressed flowers are generally pressed closed, so that they eye would not be visible, and though the color patterns on the outside of the corolla (abaxial side) may be similar to those inside (adaxial), they may not always be exactly the same. Additionally, colors tend to change and white areas are sometimes discolored by other pigmented parts of the plant. Photos of Zeltnera from the United States on iNaturalist show that even more species than those noted by Broome contain an eye, even a couple which she raised but did not note such as Z. beyrichii and Z. texensis. The presence of photographic evidence may be useful in the future for amending species descriptions and addressing some taxonomic issues in the genus.

The eye in most Zeltnera, like Broome indicated, usually is formed by some pattern of white at the base of the corolla lobes. This often forms a rough pentagon but can also appear bluntly star shaped, sometimes with 5 points and sometimes with 10. The eye can also be influenced by portions of the corolla throat when visible, which is often a yellowish color. Pringle(2023), in his overview of the genus, mentioned the possibility of a pale yellow or green eye, and this may be what he was referring to. The eye can also include reddish patterns on the background of white or lightened pigment. Some species, usually smaller flowered ones, may lack an apparent eye or have a much less conspicuous one. The following table is an attempt to describe the eyes of some of the more commonly photographed species using accepted species in the U.S. from Pringle(2023), but it is preliminary and more research needs to be done.

Eye presence and pattern in U.S. Zeltnera

Species	Eye	Description
Z. arizonica	yes	tbd, but often with greenish yellow center and irregular white border
Z. beyrichii	yes	small, pentagonal with yellow throat visibly bordered by small white band
Z. calycosa	yes	either white 10 point star with central yellow region or more irregular bordered mostly white band
Z. davyi	usually	prominent white band about 1/3 to 1/5 of corolla lobe length, forming rough pentagon or star, sometimes with yellow center
Z. exaltata	unclear	inconspicuous at best, possibly obscured by angle of throat
Z. glandulifera	unclear	only one known photo
Z. maryanniana	yes	yellow 10 point star similar to some Z. calycosa
Z. muehlenbergii	unclear	variable from small white band or projection at base of corolla lobe to more irregularly pigmented base of lobe, needs more research
Z. multicaulis	unclear	only one photo
Z. namophila	inconspicuous	small,irregular
Z. nudicaulis	minimal	perhaps a small star shape
Z. texensis	yes	small white band from about 1/8 to 1/10 the length of the corolla lobe producing pentagonal or star shape
Z. trichantha	yes	white star or pentagon
Z. venusta	yes	varying from prominent white band in southern California to various patterns in the Sierra Nevadas (possibly Z. abramsii)

While this post focuses primarily on the eye of Z. davyi, I noticed some other trends which may provide some insight into some of the historical problems in the genus.

Eye patterns in Z. venusta in the Sierra Nevadas differ from those in Southern California
Specimens identified as Z. venusta in the Sierra Nevadas have sometimes been treated as the subspecies C. venustum var. abramsii (Munz and Keck 1959, Broome 1973) or as a separate species, Z. abramsii (Mansion 2004). Pringle (2023) chose not to recognize them as a separate taxon. However, a brief look at photos of specimens in the Sierras shows a number of different eye patterns which seem to differ from specimens in southern California, which appear to mostly have a simple, prominent white eye. The different eye patterns may provide more evidence for support of Z. abramsii as a separate species, but further research is needed.

Eye patterns of some Z. calycosa in the Texas Hill Country and west Texas are very similar to those of Z. maryanniana
Both Z. maryanniana (found in southern New Mexico and the Trans-Pecos of Texas) and some specimens identified as Z. calycosa in central and west Texas have an eye consisting of a 10 point star. In Z. maryanniana, this star is mostly yellow. In specimens of Z. calycosa ranging eastward all the way to the central Edwards Plateau, specimens can be found with a 10 point star that is white with a yellow center. Other specimens identified as Z. calycosa on the Edwards Plateau usually have a more irregular white band without the yellow center and they may correspond with specimens previously identified as C. calycosum var. nanum(Broome 1973, and maybe others). In central Texas, the plants with the 10 point star often appear larger with larger blooms than those without the star. While Z. maryanniana and Z. calycosa reportedly have a different base chromosome number (Mansion and Zeltner 2004), this may provide evidence to reexamine that relationship. Alternatively, the different eye patterns on the Edwards Plateau may provide further evidence for some level of taxonomic recognition such as the varietal levels recognized by Broome (1973). It is also possible that the similar or different eye pattern may just be genetic variation among the currently recognized taxa. There are a number of different possibilities which may warrant further research.

A short history of Zeltnera davyi

A short history of Z. davyi can be found in Pringle (2010), mostly in comparison to Z. muehlenbergii. According to Pringle, the species now known as Z. davyi was first treated as a variety of Centaurium exaltatum by Jepson (1925), though he later (Jepson 1939) called it C. muehlenbergii. However, Abrams (1951) elevated it to species as Centaurium davyi, and this was mostly followed by later authors (Dunn 1967, Hickman 1993, Munz and Keck 1959) until it was moved to Zeltnera by Mansion (2004), when most north American taxa of Centaurium were moved to Zeltnera.

As mentioned above, the eye is never mentioned in treatments of Z. davyi. Only in Jepson (1939) and Abrams (1951) does the description include a yellow throat, but never is a white area in the lobes mentioned. Illustrations of Z. davyi are found in Abrams (1951) and Pringle (2012) and both show a lighter throat bordered by darker lobes, but this does not seem to convey accurately the extent to which the eye extends into the corolla lobes. None of the major floras seem to have photos of Z. davyi, but it would not surprise me if some local floras may have photos. I have not done an exhaustive search, and if there are smaller local floras, they are not known or easily available to me. The first photo I can find online appears to be one linked to the Wikipedia page for Centaurium davyi, which was linked in 2010 (thank you Wikipedia for recording the date it was uploaded) and sourced from the flicker account of Jerry Kirkhart. The first iNaturalist observation of Z. davyi (18703) was added by the venerable Ken-ichi Ueda in May of 2011. Photos of Z. davyi on Calflora seem more recent, and not always correct.

© Jerry Kirkhart (CC-BY), © Ken-ichi Ueda (CC-BY) (iNat Ob. 18703)

Z. davyi occurs mostly on the central California coast, where it is mostly the only Zeltnera present. It has historically been noted to occur as far north as Mendocino county and range to the south as far as San Luis Obispo county and also the Channel Islands (Abrams 1951; Mansion 2004). Pringle (2010) extended the range in the north to Humboldt county and in the south also included mainland Santa Barbara county. Its range overlaps mostly with Z. muehlenbergii, but also somewhat with Z. venusta and Z. exaltata, which means that those are the three species that it is most important to differentiate it from.

Traditional Differentiation from nearby species

Z. davyi has traditionally been differentiated from its near neighbors by quantitative measures of parts such as the pedicels, corolla lobes and even anthers, along with the presence of a keeled calyx, which may occur only in Z. davyi among U.S. species if Pringle (2023) is correct. The following tables show characteristics typically used to differentiate Z. davyi from Z. exaltata, Z. venusta, and Z. muehlenbergii, based on Pringle (2010,2023).

Z. davyi vs. Z. exaltata

Feature	Z. davyi	Z. exaltata
Corolla lobes	5	4
Pedicels	(2–)4–25(–55) mm, all or most shorter than closed corollas	(2–, on ultimate branches)10–70 mm, all or most longer than closed corollas

Of the three, Z. exaltata seems the easiest to differentiate from Z. davyi due to the longer pedicels and typically 4-merous corollas, though many Zeltnera corollas occasionally have only 4 lobes. The 4-merous corollas of Z. exaltata does not seem to have been noted before Pringle(2012) though, and I have seen a number of iNaturalist observations of it which appear to have both 4 and 5-merous corollas, so its reliability as a distinguishing character may need more research. However, Z. exaltata seems fairly rare in the range of Z. davyi, so the distinction may not need to be made often.

Z. davyi vs. Z. venusta

Feature	Z. davyi	Z. venusta
Corolla lobe length	3–7 mm	(5–)6–20 mm
Calyx keeling	keeled	+-sharply ridged (from key not text)

Due to its smaller corollas, Z. davyi has often been grouped with other small flowered species in earlier floras. However, Broome (1973), though she did not offer a treatment of Z. davyi noted in her treatment of C. venustum:

Centaurium davyi might, in fact, have been a direct offshoot of C. venustum from which it differs only in the smaller corollas and the extremely carinate calyx-lobes.

Borderline specimens likely need to be examined for both corolla size and the presence of the calyx keels. Fortunately, the range of the two species likely only overlaps in a small area, perhaps even limited only to the Channel Islands. Z. venusta specimens are documented to have a corolla eye, and the specimens in southern California appear to have a simple white one, like Z. davyi, but it possibly does not extend as far into the corolla lobe.

Z. davyi vs. Z. muehlenbergii

Feature	Z. davyi	Z. muehlenbergii
Mid-stem leaves	elliptic to ovate leaves over 5 mm wide (except on the smallest plants) well into the inflorescence	elliptic to narrowly ovate leaves, when present, are usually limited to the proximal one-third or less of the plant, with the distal leaves being narrower
Pedicels	consistently present pedicels 4–30 mm long, (2–)4–25(–55) mm	proximal flowers in center of cyme divisions sessile or to 12 mm, distal flowers generally sessile or pedicels to 4 mm
Calyx keeling	keels proximally 0.3–0.6 mm wide	not keeled or proximally only, less than 0.25mm (2010)
Corolla lobe width	ovate-elliptic 3–7 × 2–3 mm	elliptic-oblong 2–7 × 1–2 mm

The most difficult species to differentiate according to the traditional treatments are Z. davyi and Z. muehlenbergii as many of the characteristics are similar and their ranges overlap more substantially, mostly around the Bay Area. Pringle (2010) spent significant effort to determine that the type specimens of Z. muehlenbergii were not actually Z. davyi. To add to the confusion, he also noted the possible presence of a smaller keel on Z. muehlenbergii. However, Pringle (2023) omitted this, so it is unclear if this can occur or not.

Complications with the calyx keel

There are two additional complications regarding the keeled calyx lobes. First, both Mansion (2004) and Pringle (2023) draw a distinction between being keeled and simply sharply ridged. Second, in practice it appears difficult to capture the calyx keel in photos, requiring either skill or luck to get the right light to show the subtle change in width from the adaxial to abaxial side of the lobe. I suspect many people mistake sharply ridged calyx lobes for keeled ones. Note in the photo below how the keel on the central calyx lobe (running horizontally) is shaded darker than the rest of the lobe. This feature is not noticeable on many of the other lobes of the same plant in this photo.

Z. davyi keeled calyx

An aside on troubles with Z. muehlenbergii

Though I will explain these issues in more detail in a separate post, in addition to the problems with morphological overlap, there are number of other issues with Z. muehlenbergii which make identification of Z. davyi more difficult. Primarily, the issue is that there is no clear concept of what Z. muehlenbergii corollas look like as fresh specimens. Historically, as explained in Pringle (2010), Z. muehlenbergii has been confounded with the old World Centaurium now known as Centaurium tenuiflorum due to morphological similarities of the corolla size and pedicel length. On iNaturalist, this similarity has caused mass misidentifications of Centaurium specimens as Z. muehlenbergii, primarily in northern California, as the primary differentiating feature (the stigma shape) is almost never photographed (a similar phenomenon occurred in Texas with Z. texensis). This has likely masked what the corollas of Z. muehlenbergii really look like, at least in parts of its range. Once the massive numbers of Centaurium are dispensed with, it leaves a fairly small set of around 100 observations which are likely Z. muehlenbergii ranging from San Luis Obispo county, California in the south to Vancouver Island, British Columbia in the north. Unfortunately, there is some variation over that wide range in the appearance of the corolla, which, though it is not surprising over such a wide range, casts some doubt over whether they are all the same species. Where this concerns Z. davyi is that a number of these specimens come into close contact with what appear to be Z. davyi (as I will explain further down), and this occurrence also raises the possibility that some of these may be Z. davyi, or at the very least that some may be hybrids. Specimens on Vancouver Island (thought to be Z. muehlenbergii) show some resemblance to those around the Bay Area (Marin, San Mateo, Santa Cruz counties) that are in close contact to specimens thought to be Z. davyi. Those specimens seem to have some variation in the eye area from having mostly white eye with faint red dots, to having an eye with a central red region with lateral white or lightened areas, to having a darker central red region. Another variety, which seems to be favored on slightly more inland specimens (Santa Clara,Alameda, Contra Costa counties, etc), seems to have a more defined, yet jagged border with the white area in the throat and base of the lobe compared to the rest of the pigmented lobe.

Possible Z. muehlenbergii showing faint or streaked eye, Capital District, British Columbia

© Dan Tucker (CC BY-NC) (iNat ob. 50331884)

Possible Z. muehlenbergii showing more jagged stark eye, Santa Clara county, California

© Joe MDO (CC BY-NC) (iNat ob. 226058090)

Another possible issue, which I think is unlikely, is that since neither the description of Z. davyi nor Z. muehlenbergii mention the presence of an eye, that Z. muehlenbergii may have a prominent eye and that the only differentiating features are those previously mentioned.

Eye area not visible in Herbarium specimens

The fundamental problem with herbarium specimens not only of Zeltnera is that flowers are usually pressed closed, so there is no visibility to the actual eye area. One might assume that the pigmentation is the same on the outside of the flower as on the inside, but I don't think that has been proven to be the case. Even if they are similar, actual herbarium specimens present an inconsistent view of the area that would constitute the eye (the corolla throat and the base of the corolla lobes). What can usually be seen are three different colored areas: a greenish area consisting of the corolla tube and calyx, a light or yellowish area consisting of the corolla throat and sometimes the base of the corolla lobes, and a dark reddish area consisting of the majority of the length of the corolla lobes. The light area that would normally make up the eye is inconsistent. While the narrowed throat is almost always light or yellowed, the base of the corolla lobes varies from being distinctly light (suggesting an eye), to blending gradually from light to dark, to being almost completely dark. One collection identified by Pringle, Triano 204 (CHSC), contains a number of small plants that illustrate this range of conditions and show how it might be difficult to generalize about the presence of an eye in Z. davyi based on herbarium specimens. It is unclear if this variation represents actual variation in the population (which may be possible as reflected by at least one population on iNaturalist), or if it simply shows variations in the ways that colors can change in specimens during preservation.

I only examined a small number of images of specimens of Z. davyi as it quickly became apparent why the eye was not documented. While there are probably a couple hundred images of possible Z. davyi herbarium specimens online, it is not always clear if they have been identified correctly. The type, Davy 396 (UCJEPS), does not have an image online, nor do any of the other specimens at UCJEPS. Of the studies done on Centaurium and Zeltnera in the U.S., only Dunn (1967) contains a list of representative specimens. A number of specimens have been annotated by her at GH, but it is unclear if her concept of Z. davyi would be the same as later authors. Only one specimen identified by Pringle as Z. davyi is currently online, Triano 204 (CHSC). Other specimens I examined appeared to be Z. davyi based on pedicel length, corolla size, and sometimes relatively inflated looking corolla tubes.

Specimen images examined: California.
Marin Co.:
East side of Inverness Ridge along Point Reyes Road, 19 Aug. 1945, J.T. Howell 21420 (DAV);
Monterey Co.:
Point Lobos, back of Hidden Beach, 7 May 1944, L. C. Wheeler 5967 (UCR);
Ellicott Slough Long-Toed Salamander Preserve, 16 May 1980, B.M. Pavlik and B. Lowe s.n. (DAV);
San Luis Obispo Co.:
Los Padres National Forest. Just S of junction with Hwy 58 on Red Hill Rd, 4 May 1998, D. J. Keil 27064 (OBI);
San Mateo Co.:
Near Pescadero, 26 Jun. 1938, A. Eastwood and J. T. Howell 6060 (GH);
Sand dunes at Point Año Nuevo, 18 Jun. 1954 , R. Bacigalupi and G.T. Robbins 4560 (OBI);
Summer Spring Road, Woodside, 7 Jul. 1963, P. Rundel 1133 (UCI) - (presumably Summit Springs Rd.?);
18.5 miles north of San Mateo-Santa Cruz County line, 10 June 1969, D.F. Howe 4641 (SD);
Ca 750 ft S of Hwy. 1 and 2.9 mi S of Pigeon Point Light Station, 24 May 1988, S. Triano 204 (CHSC);
Santa Barbara Co.:
Vandenberg AFB. Burton Mesa at 35th St site between Main Gate and Surf, just W of RR tracks, 8 May 1986, D.J. Keil et al 19367 (OBI);
Santa Cruz Co.:
Big Rock Hole, San Lorenzo River, Henry Cowell Redwoods State Park, 14 June 1998, K. Kellman 1632 (UCSC).

iNaturalist specimens

In the summer of 2024, as part of an effort to understand what Z. muehlenbergii actually looks like, I decided to examine all known observations then identified as Z. davyi to get a general sense of what was captured in the observations and whether they could be identified by the traditional morphological methods. There were a total of 464 observations, though I later discovered that due to some quirks in how the iNaturalist Identify interface works, that I missed about 11 observations. The process was more exploratory than a rigorous examination and as such I did not keep a close record of what could be seen in each observation. My general sense after this endeavor was that they could not be identified based on the traditional metrics. Corolla measurements were almost never provided. Pedicels often were not visible and never measured. Calyx lobes, though often appearing prominent or ridged, could not determined to be keeled or not. What grouped most of these observations together was the location and the similar corollas, which included the prominent eye.

After setting the project aside for a few months, I returned to the data set to attempt to quantify what actually might be visible and reexamined the first 175 observations more closely to determine if the following characteristics, among others were visible: prominent corolla eye, pedicels, calyx lobes, calyx keels, and corolla lobe width measurement. I also kept a record of what identification I would likely give to each observation based on overall impression, which in retrospect was probably due to overall look of the corolla, including the eye. Two observations were actually duplicates, so I closely examined only 173. For clarification, a prominent eye for the purposes of this discussion, typically fits the description for the eye of Z. davyi provided in the earlier table as a prominent white band about 1/3 to 1/5 of corolla lobe length, forming a rough pentagon or star, sometimes with a yellow center.

Feature visibility

Feature	Number
Open flowers	167
Prominent corolla eye	147
Pedicels	100
Calyx lobes	157
Calyx keel possibly present	37
Corolla measurements	1
Total Specimens	173

Most observations, as might be expected, had open flowers where the eye area was visible. A handful either had closed flowers, blurry flowers, or no flowers. Most of the ones with open flowers, 147 of 167, had what appeared to be a prominent eye. Of the remaining 20, 3 probably had a prominent eye, but were photographed at an odd angle. The remaining 17 mostly had some sort of small or irregular eye. Since the corolla tube and throat are generally different colors than the corolla lobes, almost all specimens had some sort of marking, just not prominent. Centaurium species or those highly likely to be such typically had a small visible eye. Z. muehlenbergii generally had some sort of white region at the base of the lobes. A few specimens had lightened or irregular pigmentation at the base of the lobes.

The only other aspect of the corolla I checked was the presence of any information about the corolla size. Only one of the 173 observations mentioned a possible estimation of the corolla size, and its accuracy seemed questionable. On rare occasions, an observer would capture a small patch of plants showing a range of corolla sizes, and sometimes that would show a plant with corolla lobes seemingly less than half the width of others in the photo. As the width is described as being from 2-3mm (Pringle 2023), this seems to imply that the range may be larger than previously thought, perhaps wider or narrower. This could mean that corolla width range may overlap with that of Z. muehlenbergii, though this hypothesis would have to be confirmed with actual measurements.

Pedicels or the areas where they might be were sometimes visible (100 of 173), but rarely clearly so, and if so it was usually no more than one. They ranged from clearly sessile central flowers in likely Centaurium species to about as long as the calyx in some likely Z. davyi, but rarely were enough visible to determine whether the pedicels of the specimen were more consistent with Z. davyi rather than Z. muehlenbergii.

Calyx lobes were usually visible to some degree (157 of 173), but as mentioned before it was very difficult to determine if they were actually keeled or simply ridged, and at most I could only say that keels might be present (37 of 173). As determination was mostly based on subtle color changes on parts of the lobe, it became hard to tell when it was just a color change and when it might reflect a keel.

Ultimately, the data present in the photographic evidence was usually not complete enough to determine species based on key documented morphological characteristics, especially given overlapping characteristics with Z. muehlenbergii, so it did not seem possible to corollate the prominent eye to an actual species based on morphological characteristics.

Attempting Identifications

Despite the clear issues with identifying these specimens, I did attempt to identify what I most likely thought they were. None of these were rigourous. In addition to the problems identifying Z. davyi, identifying Z. muehlenbergii and Centaurium have their own issues which I will address in an upcoming post on Z. muehlenbergii. For clarification, the following table represents what I would identify the specimens as, but I often did not formally add an identification on the iNaturlist system, yet.

Initial Identifications

Taxon	Number
Z. davyi	141
Zeltnera	16
Chironiinae	10
Centaurium	4
Z. davyi duplicates	2
Plantae	1
Z. muehlenbergii	1

Unsurprisingly, given my bias towards those with a prominent eye being Z. davyi, most specimens appeared consistent with Z. davyi. The total number(141) was less than those with prominent eyes (147), due mostly to being unable to distinguish Z. davyi and Z. venusta in the islands off the southern coast of California. These represented most of the specimens which I could identify to Zeltnera, but not to species.

Problematic Zeltnera specimens

Possible taxon	Number	Comments
Z. davyi or Z. venusta	6	Channel Island specimens that could not be differentiated
possible Z. muehlenbergii or hybrid	3	small lobed specimens with mottled eye or other intermediate characteristics (Wilder Ranch, Santa Cruz County; Point Reyes, Marin County)
Z. davyi likely	2	small flowered specimens with eye but no other corroborating features
Z. muehlenbergii likely	2	further south than Pringle (2010) documented range (San Luis Obispo county)
possible Z. davyi without eye or hybrid	1	(San Mateo County, Bean Hollow)
unknown Zeltnera	1	small plants with small lobes and a large irregular eye, with a mix of 4-merous and 5-merous corollas (Milipitas Road, Monterrey)
possible Z. muehlenbergii	1	inland sample without an eye but only corolla view (Hain Wilderness, San Benito County)

Problematic Zeltnera populations which included specimens both with and without prominent eyes were encountered. I did a close examination of some of these populations, especially in cases where specimens without prominent eyes were near or in populations with prominent eyes. This usually included specimens in the area which were not in the original Z. davyi data set as they might be identified as something else.

Populations without prominent eyes

A few areas in the range of Z. davyi appear to have observations of specimens which lack a prominent eye in proximity to observations of specimens which have a prominent eye, and which are presumed to be Z. davyi. These populations occurred around the following areas: Bean Hollow State Beach in San Mateo county, Wilder Ranch State Park in Santa Cruz county, and Point Reyes National Seashore in Marin county. Additionally, there was a single occurrence near Fort Ord in Monterey county. These populations were not monolithic. While the populations at Point Reyes and Wilder Ranch appeared to have small lobed specimens with mostly light and sometimes streaked pigmentation towards the base of the lobes, a handful of specimens at Bean Hollow appeared to have no lightened area whatsoever. There is a strong possibility that the specimens without prominent eyes in these areas may simply be sympatric examples of Z. muehlenbergii, though the proximity to apparent Z. davyi in areas known to favor Z. davyi (sandy coastal areas) raises possible questions.

Bean Hollow State Beach

Of a total of 25 Zeltnera specimens at Bean Hollow, 21 had prominent eyes, 3 did not appear to have eyes (6412484,168085147,223270062), and 1 was a mixed sample (167981075) which showed plants with and without eyes. Though there was no prominent eye on the adaxial view, all 4 abnormal samples showed some amount of lighter color in the area that might correspond to the eye on the abaxial view. All of the plants appeared fairly small, but there was no scale. They occurred in different years but did appear to be in the same general area. Pedicels that were visible appeared quite short (typically no more than about 1/6 the length of an adjacent calyx lobe), but this might be expected on such small plants with few flowers crowded together. Calyx lobes appeared prominent with the usual difficulties on determining keeling. Of the 3 problematic populations, these specimens seem most likely to be only slightly aberrant Z. davyi, though it is difficult to rule out Z. muehlenbergii completely since they look similar to specimens thought to be Z. muehlenbergii in other areas.

Bean Hollow Zeltnera without eye in a population mostly with eyes

Wilder Ranch Ranch State Park

At the north end of Wilder Ranch State Park and nearby across Empire Grade Road there are 20 Zeltnera observations. Ten of these observations have prominent eyes and appear to be Z. davyi. The majority of those observations are south of Empire Grade with 5 near the junction Chinquapin and Long Meadow Trails, two a little to the northeast on a side trail off Chinquapin Trail, one further south on Chinquapin Trail, and one east of Empire Grade in what appears to be called Chinquapin Meadow. Nine observations showed more variety of pigmentation and are primarily to the north of Empire Grade Road in an area apparently known as North Meadow, approximately 3/4 a mile north of most of the Z. davyi, though two are only about 400 meters northeast on Chinaquapin Trail. These mostly appear to have small corolla lobes with varying degrees of pigmentation at the base of the lobes, but sometimes with noticeably long pedicels. Without the presence of Z. davyi nearby, these might be more easily identified as Z. muehlenbergii, but the presence of Z. davyi at least raises the possibility that they might be small abberant Z. davyi or even hybrids. One additional observation was identified as C. tenuiflorum and I could not identify it to genus based on the photos.

Observations in Wilder Ranch area - (p) = noticeably pedicelled

Type	Number	Observations
With Prominent Eyes	10	13318065, 119319973(p), 119487257, 119824765, 167973904, 170586655, 170587575, 170587638, 220159684, 231125478
Varying Eyes	9	3150046(p), 12941949(p), 47234072(p), 47358361, 50300534, 119487187, 169528011, 221658101, 221751755
Unidentifiable to genus	1	119824250

Wilder Ranch Zeltnera without prominent eye

Point Reyes National Seashore

There are 12 Zeltnera observations at various points at Point Reyes National Seashore in Marin county, all located in what are labelled as Pastoral Lands on the National Park Service map of the area. 11 of the observations are in areas roughly around the Point Reyes Beach North access to around or northeast of Bull Point Trail and one all the towards the Coast Guard site. One is across the Drakes Estero near Sunset Beach Trail. Six of the observations have a prominent eye consistent with most Z. davyi while five appear to have smaller lobes with various pigmentation at the base. One observation has closed corollas. 10 of the observations are in a roughly 2 square mile rectangle with the north beach access being the southern side, extending about a mile northward. The two types only come close with several observations at the end of short spur off the north beach access, though they were observed in different years. The smaller ones were typically observed later in July.

In the absence of the specimens with prominent eyes which are more consistent with Z. davyi, those without them would probably be identified as Z. muehlenbergii based on the small corolla sizes (though it is impossible to tell without an actual measurement). Some of corollas from specimens here look similar to those from Wilder Ranch, though all of the specimens around Point Reyes appear to be quite small. Ironically, though the corolla lobes appear small, the corollas overall don't appear exactly like observation of Z. muehlenbergii further inland, which is a problem I will likely discuss more in a later post on Z. muehlenbergii. There are any number of possibilities for what these specimens without prominent eyes are.

Type	Number	Observations
With Prominent Eyes	6	6619278, 221726206, 222907433, 223673706, 226785036, 234798789
Without Prominent Eyes	5	3689410, 6952045, 29692736, 173368345, 230018760
Flowers closed	1	173936431

Point Reyes Zeltnera without prominent eyes

Fort Ord

Though not included in the 175 I examined more closely, one observation (178907160) in the Fort Ord area of Monterey county appeared to have a similarly streaked corolla eye to some seen at Wilder Ranch and Point Reyes. According to local naturalist Fred Watson(inat Ob. 184610832), the local Flora of Fort Ord (Styer 2019, as cited in Watson 2024) notes two possible types of Z. davyi at Fort Ord (and apparently also Santa Cruz county) and a botanist named Randall Morgan believed a smaller, earlier version may have been an undescribed species. I have not been able to follow up on this reference, but it is worth noting that Pringle (2010) cited a number of instances of a collection of Z. muehlenbergii (Elmer 4378) from nearby Pacific Grove as showing possible signs of intergradation or hybridization with Z. davyi. It is possible that this one observation is representative of those concepts, but it appears to be the only one in an area with a fairly high number of Z. davyi observations.

Populations with prominent eyes

There were two areas where specimens with prominent eyes proved difficult to identify, but I did not examine them as closely for this post. This was mostly a problem in the islands off the coast of southern Californina, where the corollas looked as if they might either be large Z. davyi or small Z. venusta and it was not possible to determine if the calyx lobes were keeled.

A second area occurred inland in Monterrey county (Milpitas Road on the northwest side of Fort Hunter Legget). There were five specimens with small corollas and prominent eyes. Three of the specimens looked like they could plausibly be small Z. davyi(8611311,190444303,219222650), but it was difficult to be sure given the lack of photos. Two of the specimens(3189234,220339267) seemed to have larger eyes than regular for Z. davyi and appeared somewhat fragile with long pedicels and a fair number of 4-merous corollas. It seemed like certain aspects may be similar to Z. exaltata, and there did appear to be a rare observation of Z. exaltata (28461270) not too far further south, so perhaps some of the specimens could be hybrids.

Curious occurrences in Sonoma

Sonoma county may or may not be in the known range of Z. davyi depending on one's reference, though the distribution map in Pringle (2010) does not include it. However, there are two observations in the county which appear legitimate, both appearing to have a prominent eye. In the area where one (214918970) of them occurs, southwest of Kenwood, there are two other observations which appear similar, one with a streaked eye (214933487), and one with narrow lobes (164640412) possibly with a little streaking. This may represent some genetic variation on the edge of the range or perhaps some hybridization. Coincidentally, according to Pringle (2010) the original type specimens of Z. muehlenbergii collected by Douglas were likely collected in Sonoma county. Despite the overwhelming number of likely Centaurium observations in the county, there are few actual Zeltnera, and only a handful of those seem plausibly consistent with Z. muehlenbergii. The observations unfortunately do not show many details.

The geographic argument

One avenue I have not pursued vigorously is the geographic one. In areas where Z. davyi has been reported to exist in the scientific literature, on iNaturalist one mostly sees observations of Zeltnera with corollas with the prominent eyes as pictured earlier. While this is more difficult to divine in the northern range of Z. davyi due to the prevalence of Centaurium misidentified as Zeltnera, it is more easily seen in the southern part of the range in the counties of Monterey, San Luis Obispo, and Santa Barbara. Additionally, Z. muehlenbergii is either rare or not present in these counties according to Pringle(2010), though recent reports by Keil and Hoover (Vascular Plants of San Luis Obispo County, California, 2023, which I have not seen, cited in Excoff 2025) and observations on iNaturalist seem to extend its range to San Luis Obispo. It seems unlikely that another native species of Zeltnera would become so prevalent in the area, though certainly not impossible.

Zeltnera eye present along central coast of California

County	Prominent	Minimal	None	White corolla only
Monterey	197	1 (streaked)	1	4
Santa Barbara	56	0	0	0
San Luis Obispo	85	4	2	5

Those with prominent eyes are thought to be primarily Z. davyi (a few likely Z. vensuta), those with minimal eyes Z. muehlenbergii, and those without Z. exaltata. The one in Monterey with a minimal, streaked eye is the one from the Fort Ord area discussed earlier.

Summary

The occurrence of a corolla eye has not been well documented for most Zeltnera species, including Z. davyi. The majority of iNaturalist observations of Zeltnera within the known range of Z. davyi show a fairly consistent, prominent, white, roughly pentagonal eye. However, there is not enough information shown or provided in the observations to identify them to species based on the current species descriptions. Additionally, a small number of Zeltnera observations that lack a prominent eye either within or near populations that have the eye indicate the possibility of some variation within Z. davyi, but it is also possible that they may be Z. muehlenbergii or hybrids of the two species.

References

Abrams, L. R. 1951. An Illustrated Flora of the Pacific States Washington, Oregon, and California, Vol 3. Gerianaceae to Scrophylariaceae/geraniums to figworts. Stanford University Press, Stanford, CA.
Beidelmann, L.H. and E.N. Kozloff. 2003. Plants of the San Francisco Bay Region: Mendocino to Monterey. University of California Press, Berkley, CA.
Beidelmann, L.H. and E.N. Kozloff. 2014. Plants of the San Francisco Bay Region: Mendocino to Monterey. University of California Press, Berkley, CA.
Broome, C. R. 1973. Systematics of Centuarium (Gentianaceae) of Mexico and Central America. Ph.D. Thesis, Duke University, Durham, North Carolina.
Correll, M. S. &amp M. C. Johnston. 1970. Manual of the Vascular Plants of Texas. Texas Research Foundation, Renner, Texas.
Diggs, G. M., Lipscomb, B. L., O'Kennon, B., Mahler, W. F., &amp L. H. Shinners. 1999. Shinners &amp Mahler's Illustrated Flora of North Central Texas. Botanical Research Institute of Texas.
Dunn, R. A. 1967. A Revision of the Genus Centaurium of the Continental United States. Ph.D. Thesis, The Catholic Univserity of America, Washington, D.C, USA.
Excoff, P. 2025. iNaturalist observation: https://www.inaturalist.org/observations/27477275. Accessed on 16 Feb 2025.
Gray, A. 1876. “Erythraea”. Pp. 477–480 in: Brewer, W. H. &amp Watson, S. (eds.), Polypetalae. California Geological Survey. Botany of California, vol. 1. J. Wilson, Cambridge.
Grisebach, A.H.R. 1839. Genera et Species Gentianaearum Adjectis Observationibus Quibusdam Phytogeographycis. J. G. Cotta, Stuttgart and Tübingen.
Grisebach, A.H.R. 1846. Gentianaceae. Pp. 39–141 in: Candolle, A. de (ed.), Prodromus Systematis Naturalis Regni Vegetabilis, vol. 9. Masson, Paris..
Harris, J. G. &amp M. W. Harris. 2001. Plant Identification Terminology : An Illustrated Glossary. 2nd edition. Spring Lake Publishing, Springlake, UT.
Hickman, J.C. 1993. Centaurium in J.C. Hickman (ed.) The Jepson Manual: higher plants of California. University of California Press, Berkeley, CA.
iNaturalist. Available from https://www.inaturalist.org. Accessed 2024 and 2025.
Jepson, W. L. 1925. A Manual of the Flowering Plants of California. Associated Students Store, University of California, Berkley, CA.
Jepson, W. L. 1939. A Flora of California. Vol 3, pt. 1, Sympetalae Hypogyneae. Associated Students Store, University of California, Berkeley, CA.
Jepson Flora Project (eds.). 2025. Jepson eFlora, https://ucjeps.berkeley.edu/eflora/, accessed on January 24, 2025.
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.
Mansion, G. &amp L. Zeltner. 2004. Phylogenetic Relationships within the New World Endemic Zeltnera (Gentianaceae-Chironiinae) Inferred from Molecular and Karyological Data. American Journal of Botany 91:2069-2086.
Munz, P.A. &amp D.D. Keck. 1959. A California Flora. University of California Press, Berkley, CA.
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. 2012. Zeltnera, in Jepson Flora Project (eds.) Jepson eFlora, https://ucjeps.berkeley.edu/eflora/eflora_display.php?tid=81825, accessed on August 19, 2024.
Pringle, J. S. 2023. Zeltnera. in: Flora of North America Editorial Committee, eds. 1993+. Flora of North America North of Mexico. 20+ vols. New York and Oxford. Vol. 14. http://www.efloras.org/florataxon.aspx?flora_id=1&taxon_id=318122
Watson, F. 2024. iNaturalist observation: https://www.inaturalist.org/observations/184610832. Accessed on 16 Feb 2025.

Note: Though I have made an attempt to cite most of my sources, I have not cited the particular iNaturalist observations unless they contained textual comments which I referenced in the text. Copyright for photos is listed separately. The observations I used formed no particular cohesive set such as one that could be easily exported or referenced on GBIF. I can provide a list of the observation numbers of the original Z. davyi observations I examined if any one is curious.