A densely hispid Onosmodium was one of the most striking of several unfamiliar glade plants that we observed on the initial Memorial Day weekend canoe trip. As this took place a few weeks after the end of anthesis in this plant, we had to wait nearly a year to see it in flower. In the interim, its more uniform leaf pubescence seemed to rule out an identification as any of the various taxa that are considered allied to (e.g., Turner 1995) or conspecific with (e.g., Das 1965) O. molle. We were also able to determine early on that the Ketona Glade plant was not merely an extremely hispid form of the sympatric O. virginianum (L.) A. DC.: by rehydrating some blackened, withered corollas that had persisted on each, we were able to determine that the corolla lobes of the Ketona Glade plant were more broadly triangular than those of O. virginianum.
Onosmodium decipiens J. Allison, sp. nov. TYPE: Alabama: Bibb County, ca. 13.7 km NE of Centreville, "Fern Glade," Ketona Dolomite outcrop above the right (N) bank of the Little Cahaba River, 1 Nov 1993, James R. Allison and Timothy E. Stevens 8139 (holotype, NY; isotypes: AUA, DUKE, GA, GH, JSU, MICH, MO, UNA, US, VDB). Figure 10.
Corollis flavidis et pilis longitudinis aequabilis intervenio in superficiebus ambabus foliorum sessilium O. virginianum (L.) A. DC. accedens, sed lobis corollae tantum acutis et apicibus antherarum ad sinus corollae attingentibus ab illo recedens.
Perennial herb, (2.6) 3-6.5 (8.4) dm tall, brittle when dried, coarsely hairy, the hairs whitish or in youth drying golden. Stems 1-several arising from a rootstock, green becoming yellowish, erect or ascending, 3-8 mm in diameter, often branching above the middle, densely hispid, the straight or somewhat falcate, pustular-based hairs (2) 3-5 mm long, those at base of stem sometimes deciduous with age but leaving distinct scars. Leaves proximally clustered in a conspicuous basal rosette, within a few nodes upwards usually distinctly reduced and spaced, thereafter only very gradually reduced, sessile or sometimes tapering to a subpetiolar base, nearly always 5-nerved, nerves and blade surfaces densely hispid with hairs ca. 2 mm long that have conspicuous pustular bases which are transparent at first and then white with age, the hairs along the nerves on the lower surface oriented perpendicular to the nerves, those of the upper surface weakly antrorse-ascending, nerves of both surfaces also with shorter, antrorse hairs; rosette leaves spreading or weakly ascending, (4.5) 6-14 cm long, 0.9-2.8 cm wide, oblanceolate to spatulate, with a round or obtuse apex, withering only after anthesis; cauline leaves spreading or somewhat ascending, 4-9.5 cm long, 0.9-3.2 cm wide, oblanceolate to elliptic, obtuse or acute. Inflorescence (6) 10-31 cm long, of few to several scorpioid cymes, the cymes short and congested early in anthesis, then lengthening and loosening, ultimately (2.5) 3.9-20.5 cm long on a peduncle (0.3) 0.5-4 cm long; bracts elliptic to ovate, accrescent, to 11 mm long and 3 mm wide in flower and 8-45 mm long, 2.5-20 mm wide in fruit; pedicel to 4 mm long in flower and 10 mm long in fruit; sepals at anthesis 4.2-7.5 mm long, 0.5-1.0 mm wide, somewhat accrescent, linear-lanceolate, pubescent, abaxial hairs ascending, ca. 1 mm long, adaxial hairs appressed, 0.3-0.5 mm long, margins ciliate with hairs 1.5-1.7 mm long; corolla 7.3-11.0 mm long, pale yellow, lobes triangular, 2.1-3.0 mm long, 1.1-2.0 mm wide at the base, externally pubescent with hairs of two lengths, the longer hairs 1.2-2.0 mm long, usually straight or apically undulate or curled, the shorter ca. 0.5 mm long and extending down to the middle of the corolla; anthers 2.0-2.5 (2.7) mm long, apices reaching approximately to the corolla-sinuses; style 12-17 mm long. Fruit an ovoid nutlet, (1.8) 2.1-2.8 (3) mm long, lustrous white, often with brownish tinges in age, sparsely to abundantly pitted, tapered gradually to the truncate basal scar, scar (0.8) 1-1.2 mm across. Chromosome number unknown.
Flowering April-early May, fruiting June-August.
English Name: Deceptive Marbleseed.
Paratypes. Alabama: Bibb Co., N of Centerville [sic], limestone [sic] glades and bluffs above Little Cahaba River, 23 May 1993, R. Kral 82507 (VDB); Limestone [sic] glades along Little Cahaba R., 9 Aug 1992, R. Kral 81282 (VDB); N of Centreville, dolomitic outcrop by Ala. Hwy. 5, 29 May 1970, R. Kral 39318 (AUA, JSU, VDB); 28 Jul 1972, R. Kral 47850, VDB); N of Centreville, 0.5 mile E of jct. AL 291 and AL 5, limestone [sic] outcrop in clearing, 13 May 1983, R. D. Whetstone and K. E. Landers 13395, JSU); 6 miles N of Centreville/Brent, Hwy. 5, Ketona Dolomite glade, 7 Jun 1996, John R. MacDonald 9537 (VDB); 15.1 km NE of Centreville, "Brown's Dam North Glade West," 1 May 1994, A. and S. 8235 (FLAS, FSU, GA, GH, MICH, MO, NY, TENN, UNA, US, USCH, VDB); 15.3 km NE of Centreville, "Desmond's Glade," 29 Apr 1994, A. et al. 8215 (AUA, GA, JSU, UNA).
Among the endemics, only Erigeron strigosus var. dolomiticola and the new Onosmodium are characteristic elements of even the smallest Ketona Glades. While the Erigeron is absent only from a single glade near the western periphery of the glade "archipelago," O. decipiens is absent only from the very easternmost glade. Like the Erigeron, O. decipiens is a plant of full or partial sunlight, found on the open glade or sometimes along edges.
The literature (Gandoger 1918) contains a taxon named Onosmodium alabamense, described by the notorious Michel Gandoger (1850-1926), who, in the words of Correll and Johnston (1979), "named thousands of unacceptable species." From the scanty descriptive information provided, it would seem that Gandoger's plant, with its "stylus inclusus" is not an Onosmodium at all, at least in the modern, post Mackenzie (1905) sense. The "elongati lineares" corolla lobes further disqualify Gandoger's name from possible application to the Bibb County plants.
Onosmodium decipiens exhibits characters alternately either of O. molle Michx. ssp. hispidissimum (Mackenzie) Boivin or of O. virginianum. Table 1 summarizes the principal differences among these three taxa of Onosmodium. Both leaf indument and maximum cauline hair length are ideal characters as they can be used throughout the growing season. The double indument of the leaf surfaces of O. molle ssp. hispidissimum, a shared trait of the several subspecies of O. molle (Das 1965), consists of short, appressed hairs beneath the longer, more spreading ones.
Table 1: Morphological comparison of three Onosmodium taxa
O. molle ssp. hispidissimum
O. decipiens
O. virginianum
plant height
to ca. 12 dm
to ca. 6.5 (-8.4) dm
to ca. 5 dm
stem hair length
> 2.5 mm
< 2.0 mm
leaf indument
double
simple
corolla color
white w/ greenish lobes
light yellow
corolla lobes
nearly deltoid
acuminate
anther apices
at corolla sinuses
at sinuses
below sinuses
nutlet base
constricted
unconstricted
The new Onosmodium is apparently the only Ketona Glade endemic other than Dalea cahaba to have been collected prior to 1992 (see paratypes, above). Specimens at VDB from Robert Kral's 1970 and 1972 Bibb County collections, determined by him as O. hispidissimum Mackenzie, were examined by Jerry and Carol Baskin in 1982. They annotated the specimens as "Onosmodium cf. virginianum (L.) A. DC.," recognizing that while they could not be taxon hispidissimum, they also did not fit comfortably within O. virginianum. The specific epithet, decipiens, is an acknowledgment of the plant's "deceiving" morphology.
While the closely related Onosmodium virginianum is occasional in rocky places in the vicinity of the Ketona Glades, it never replaces O. decipiens as a component of the open glade. We found both species in close proximity at only a single site, with O. decipiens occupying the more exposed habitat of the open glade and O. virginianum limited to the glade-woodland ecotone. At this site we found plants of intermediate morphology, representing hybrids and putative backcrosses (A. and S. 8231, GH, NY, UNA, US). Unlike the situation with Liatris oligocephala, O. decipiens is too abundant locally, and hybridization with it seemingly too rare, for it to be threatened with genetic "swamping" by its more widespread congener.