Rhododendron bakeri vs. R. cumberlandense Revisited
Donald H. Voss
Vienna, Virginia

Misunderstanding persists among azalea fanciers with respect to K. A. Kron's revision citing Rhododendron cumberlandense E. L. Braun as the correct name for the red azalea of the Cumberlands (Kron, 1993). This taxon was known widely in the past half century as R. bakeri (Lemmon & McKay) Hume. (A taxon is a group of plants within a category in a classification scheme, e.g., an individual species.) This note expands upon a brief treatment of this topic in Voss, 1997.

As of the mid-1930s, the red azalea of the Cumberlands had not been recognized as a separate species. Braun (1941) attributes this to the plant's resemblance to the flame azalea, R. calendulaceum . Braun mentions that J. K. Small, an authority on higher plants of the Southeast, stated in his 1933 Manual of the Southeastern Flora that: "a red azalea native in the Cumberland Mountains...may be distinct." Braun also cites her own 1935 publication of "red azalea" as a vernacular name for an azalea of the Cumberlands and W. H. Camp's 1936 mention of "red azalea of Black Mountain, Kentucky." Thus the plant was being noted as something different from previously named taxa, but it had not been formally described and given a Latin botanical name.

This situation soon changed. In 1938 Lemmon and McKay applied the name Azalea bakeri to a population of azaleas in Union County, Georgia, and described its province as the Blue Ridge Mountains of Georgia and North Carolina at elevations above 3,000 feet (900m) (Lemmon, 1938). Three years later, Braun published R. cumberlandense as the name for the red azalea of the Cumberlands (Braun, 1941). She adopted a cautiously narrow view of the province of this azalea, specifying "Cumberland Plateau and Cumberland Mountains, Kentucky" in the formal description and suggesting that "In the southern Blue Ridge province it has hybridized freely resulting in the apparently variable flame azalea of that section."

Kron provides a more detailed description of the province of the red azalea of the Cumberlands, from "Westernmost Virginia and eastern Kentucky in the Cumberland Mountains and Plateau, south through Tennessee to northern Alabama, and east of the Tennessee River Valley in the southern Blue Ridge, along the border of Tennessee and North Carolina, south to northern Georgia." Kron lists over forty representative herbarium specimens of R. cumberlandense from the range of this species; nine of these are from northern Georgia.

Hume (1948) made the new combination Rhododendron bakeri and treated this as a species separate from R. cumberlandense . With respect to the latter, he noted that "There is every reason to believe that the Cumberland azalea will prove to be a valuable addition to North American garden shrubs." Lee et al. (1952) recognized R. cumberlandense as the Cumberland azalea, noting that it blooms two to four weeks after R. calendulaceum . They stated that "Bakeri as described...is essentially similar and evidently represents the somewhat yellower end of the same complex."

A major change in usage occurred following publication of Skinner's description of a 1951 collecting trip through the southeastern United States "In Search of Native Azaleas" (Skinner, 1955). Part of the trip focused on finding plants of the red azalea of the Cumberlands at sites from Yahoo Ridge, Kentucky - the type locality for R. cumberlandense - to northern Georgia. In early June, he found the plant to be abundant on mountain slopes in Union County, Georgia, including those in Vogel State Park. Skinner found the display of:

...both colors and plants so reminiscent of those...in Kentucky that even before making a detailed check of less obvious characteristics one could scarcely doubt that this was the same Kentucky azalea. But was it?...Since both descriptions [i.e., Lemmon's for R. bakeri and Braun's for R. cumberlandense - DV] fit these plants with reasonable accuracy it would seem that this gay little bush of the Cumberland Plateau must soon shed its dual personality to be recognized by the single, prior, though less happily descriptive name of R. Bakeri .

Skinner appears to have based his somewhat tentative ("it would seem") conclusion that both names were meant to apply to the same taxon on his observation of R. cumberlandense and of R. bakeri near their respective type locations and on similarities in the respective descriptions published by Lemmon and Braun. He does not mention any inspection of the type specimens. Acceptance of R. cumberlandense as the correct epithet for the red azalea of the Cumberlands was generally withdrawn after publication of Skinner's conclusion that R. cumberlandense and R. bakeri were different names for the same taxon. Under the rules governing botanical nomenclature (Greuter et al., 2000), the first legitimate name for a taxon is the only one that may be used. Thus, the epithet bakeri (published in 1938) came to be accepted over cumberlandense (published in 1941).

In their books on azaleas, Frederic P. Lee (1958, 1965) and Fred C. Galle (1987) treated R. cumberlandense as a synonym of R. bakeri . (Galle notes that some observers consider these names to represent different species.) For reasons discussed below, Kron (1993) states that the correct name for the Cumberland azalea is R. cumberlandense . The Rhododendron Handbook 1998 (Argent et al., 1997) accepts R. cumberlandense as the correct name for the Cumberland azalea.

The Worm in the Apple

Key to unraveling the bakeri - cumberlandense confusion is under standing the dominant role of the type specimen in fixing application of a botanical name. When a botanist finds a population of plants that warrants naming as a new species, he must publish the designation of a type specimen (the "holotype"), a description or a diagnosis in Latin distinguishing the new species from others, and the name of the herbarium in which the type is deposited. The type specimen is usually a dried, pressed specimen mounted on an herbarium sheet and labeled with information about the circumstances of collection. The nomenclatural rules establish a tight link between the botanical name and the specimen designated as the type in the original publication. The Botanical Code states:

A holotype of a name of a species... is the one specimen or illustration used by the author, or designated by the author as the nomenclatural type.

As long as a holotype is extant, it fixes the application of the name concerned [emphasis added]. (Greuter et al., 2000; Article 9.1)

Skinner apparently saw the Cumberland azalea in northern Georgia and was aware of the similarities in the descriptions provided by Lemmon (bakeri) and Braun (cumberlandense). He probably (quite naturally but erroneously) assumed that the herbarium specimen designated by Lemmon as the type for A. bakeri was from the taxon described (and presumably intended) by Lemmon. This erroneous assumption - the worm in the apple - went unchallenged for nearly four decades.

The following points shed light on Skinner's conclusion that the epithets bakeri and cumberlandense refer to the same taxon and on Kron's finding that the type of R. bakeri is "not at all similar" to the type of R. cumberlandense - and, indeed, also differs in taxonomically important features from Lemmon's own description in the protologue for A. bakeri . (The type of a new combination such as Hume's R. bakeri , which changed the genus name from Azalea to Rhododendron , is the type of the "basionym," here Azalea bakeri .)

• The bakeri type has young expanding leaves; cumberlandense flowers well after the leaves have expanded. [Lemmon's description of bakeri states that the flowers expand after the leaves unfold.]

• The abaxial (lower) leaf surfaces of the bakeri type are densely pubescent; cumberlandense leaves are usually glabrous beneath, never densely pubescent. [Lemmon described the lower surface of bakeri leaves as glabrous, often shiny, or pubescent.]

• The floral bud-scale margins of the bakeri type are all unicellular ciliate; on cumberlandense the margins are usually ciliate at the apex and consistently glandular below. [In discussing bakeri , Lemmon refers to the bud-scale margins only as ciliate.]

• The corolla of the bakeri type has both glandular and eglandular multi-cellular hairs; in cumberlandense , the multi-cellular hairs of the outer surface of the corolla are consistently glandular. [Lemmon's description of bakeri states that the external surface of the corolla is glandular.]

Despite some differences in the Lemmon ( A. bakeri ) and Braun ( R. cumberlandense ) descriptions, the similarities suggest that Lemmon may well have intended his description to characterize the Cumberland azalea. But - whatever he had in mind - the characteristics of the herbarium specimen designated as the type for his A. bakeri are not consistent with those of the geographically widespread populations of the Cumberland azalea.

To speak bluntly, the evidence indicates that Lemmon was careless when he designated a herbarium specimen as the type for A. bakeri . The type specimen is from an area (Neels Gap, Vogel State Park, Union County, Georgia) where the red azalea of the Cumberlands was present. But Kron's detailed examination of the bakeri type has shown that it is from a hybrid plant, probably R. flammeum x R. canescens . We will never know the reason for Lemmon's mistake, but because the type designated for a botanical name fixes the application of that name, use of the epithet bakeri for the red azalea of the Cumberlands is foreclosed. The earliest legitimate, validly published name for the red azalea of the Cumberlands is R. cumberlandense E. L. Braun. Replacing the epithet bakeri with cumberlandense is not a matter of changing the circumscription of the widely distributed Cumberland azalea familiar to Skinner, Lee, Galle, and others. Nor is the presence of the Cumberland azalea in northern Georgia challenged. The nomenclatural change is, rather, the consequence of Kron's discovery that the specimen designated by Lemmon as the type for R. bakeri is from a hybrid plant and is not the red azalea of the Cumberlands.

In sum:

• Lemmon & McKay based the epithet bakeri on a population of azaleas found in Union County, Georgia, and in the Blue Ridge to North Carolina.

• Following Skinner's 1955 statement (apparently based on published descriptions, not on examination of types) equating R. bakeri and R. cumberlandense , the earlier-published name R. bakeri came into widespread use for the Cumberland azalea.

• During her extensive field and herbarium investigations of plants in Rhododendron section Pentanthera (deciduous azaleas), Kron concluded that the type for R. bakeri is from a hybrid plant.

• The type fixes application of a botanical name. Because the characteristics of the specimen designated by Lemmon as the type for A. bakeri (the basionym of R. bakeri ) are not consistent with those of the Cumberland azalea, the name does not apply to that species. The earliest available name for the species is R. cumberlandense .

• Kron has not changed the circumscription of the Cumberland azalea, a taxon long known as R. bakeri . It is the same group of plants; only the name has changed.

[Note: The author noted a discrepancy between the collection date for the type of A. bakeri as cited by Lemmon and as cited by Kron. Lemmon's protologue in Bartonia lists the date as "June 6, 1936." Kron (1993) shows the date as "1 vi 1936" (i.e., June 1, 1936). The protologue states that the type was deposited in the herbarium of the Academy of Natural Sciences in Philadelphia. Collection Manager James Macklin of the Academy herbarium kindly furnished a xerographic copy of the type sheet. He also searched the type and general collections for a Lemmon specimen of A. bakeri dated June 6, 1936 but found none (personal communication). The date on the holotype is "June 1 - 1936," and the label carries the endorsement in script "Type specimen W. P. Lemmon." It thus appears that the June 6 date in the protologue is a transcription or typographical error. The author expresses appreciation to Dr. K. A. Kron for her constructive comments on an earlier draft of this article. The author is solely responsible for any errors or infelicities.]

References
Argent, G. et al. 1997. The Rhododendron Handbook 1998: Rhododendron Species in Cultivation . London: The Royal Horticultural Society.
Braun, E.L. 1941. The red azaleas of the Cumberland, Rhodora 43:31-35. Galle, Fred C. 1987. Azaleas . Rev. & enl. ed. Portland, OR: Timber Press.
Greuter, W. et al. 2000. International Code of Botanical Nomenclature . Koenigstein, Germany: Koeltz Scientific Books.
Hume, H. Harold. 1948. Azaleas: Kinds and Culture . New York: Macmillan Co.
Kron, K.A. 1993. A revision of Rhododendron section Pentanthera. Edinb. J. Bot . 50(3):309-311.
Lee, Frederic P. et al. 1952. The Azalea Handbook . Washington, DC: The American Horticultural Society.
Lee, Frederic P. 1958. The Azalea Book . Princeton: D. Van Nostrand. [Lee, p. 162, states that his descriptions of the eastern native azaleas are based on Skinner (1955).]
Lee, Frederic P. 1965. The Azalea Book, ed. 2 . Princeton: D. Van Nostrand.
Lemmon, W.P. 1938. Notes on a study o f southeastern azaleas with descriptions of two new species. Bartonia 19:14-17. [Vol. 19 was nominally dated "1937" but was issued in March 1938.]
Skinner, H.T. 1955 In search of native azaleas. Morris Arbor. Bull . 6:15-22.
Voss, D.H. 1997. Rhododendron bakeri vs. R. cumberlandense . Journal American Rhododendron Society 51(4):197.

Don Voss is a member of the Potomac Valley Chapter.