135.  Dalea sericocalyx (Rydberg) Riley

(Plate CXXIII)

Probably shrubby but of unknown stature, except for glabrate older branches thinly pilosulous throughout with fine ascending and spreading hairs up to 0.6-0.7 mm long, the ample thin-textured foliage green, the leaflets pubescent both sides, charged beneath with many scattered blister-glands; leaf-spurs less than 1 mm long; stipules linear, livid becoming dry and fragile, up to 2.5 mm long; intrapetiolular glands 0; post-petiolular glands small but prominent; leaves subsessile, the main cauline ones 5-7 cm long, with slender, ventrally flattened but scarcely margined rachis and up to 9 pairs of obovate or obovate-elliptic, short-acuminate, flat leaflets 1-1.4 cm long, the terminal one distant from the last pair, the leaves of axillary spurs smaller, with only 3-5 pairs of shorter leaflets; peduncles terminal to all ultimate branchlets, 1-2.5 cm long; spikes subracemose, loosely 10-20-flowered, the flowers ascending, the axis 2-4 cm long; bracts deciduous, narrowly lance-acuminate or -caudate, 7-9.5 mm long, at the incurved, submembranous-margined, keeled and loosely folded base ± 1 mm wide, distally firm and brownish, dorsally pilose and subglandular; pedicels up to 0.2 mm long, subtended by 2 gland-spicules; calyx ± 11 mm long,'hirsute with fine straight ascending hairs up to 2 mm long, the tube (measured to dorsal sinus) 3 mm long, strongly recessed behind banner, the ribs slender but prominent, the membranous intervals charged with 0-3 minute glands, the slenderly lance-acuminate or -aristate teeth subequal, ± 8 mm long, straight, gland-spurred, plumose; petals (known only from antique specimens) apparently brownish-purple, glandless, the epistemonous ones perched less than 1.5 mm above hypanthium; banner ± 5.7 mm long, the claw 3 mm, the deltate-cordate subacute blade recurved through ± 40°, at base recessed down the claw into a comet opening adaxially, 3.4 mm long, 3.8 mm wide; wings 7.3 mm long, the claw 2.4 mm, the blade lance-oblong, 5 mm long, 2.2 mm wide; keel 9.4 mm long, the claw 2.9 mm, the blades broadly elliptic, 6.9 mm long, 3.5 mm wide; androecium 10-merous, 7-7.5 mm long, the longer filaments free for nearly 3 mm, the connective gland-tipped, the oblong anthers 1-1.2 mm long; ovary and young pod glabrous proximally, the valves charged with 2-3 large linear-oblong, vertical and subparallel blister-glands, the ripe pod unknown. — Collection: 1 (typus).

Habitat unknown, but to be sought in the pine-oak belt on the Pacific slope of Sierra Madre in s. Sinaloa or adjacent Durango, very likely flowering in winter.

Dalea sericocalyx (Rydb.) Riley, Kew Bull. 1923: 338. 1923, based on Parosela seri- cocalyx (with silky calyx) Rydb., N. Amer. Fl. 24: 63. 1919. —"Type collected in Sierra Madre (Sinaloa or Durango) by Berthold Seemann..." — Holotypus, GH! isotypi, BM (numbered 2182), K!

This is one of several remarkable and still little known species discovered by Berthold Seemann on his journeys across Sierra Madre, coming and going between the coast and Ciudad Durango during the winter of 1848-49, while H.M.S. Herald was undergoing repairs in the harbor at Mazatlan. The specimens show only the top of the plant, and its stature can only be guessed at; it should nevertheless be easily recognized by its ample pilosulous leaflets and loose barbate spikes of exceptionally long-toothed calyces. The petals, described by Rydberg from material already 80 years old as "reddish-purple", are now brownish; it is impossible to say whether they were originally bright purple or greenish-yellow and rubescent. Until more is learned on these points, it seems idle to speculate, except negatively, on the relationships of D. sericocalyx. The general aspect of the foliage, with its expanded plane leaflets, the terminal one stalked, is rather suggestive of D. carthagenensis sens, lat., but the flower is less so. The type-collection was originally misidentified by Seemann (Bot. Herald 279. 1857) as D. lachnostachya Gray and was still associated with that species by Rydberg (1919). in Parosela subgen. Trichopodium. Conforming to the diagnosis of the Lachnostachydes, the flowers were described as "subsessile, reflexed after anthesis," but the specimens do not confirm this. There are vestigial pedicels, it is true, but the form of the flower is entirely foreign to D. lachnostachya and its near kindred, and wherever the relationship of D. sericocalyx may prove to be, it is certainly not to be sought in the subgen. Theodora of this account. In the rather unusual outline and form of the banner and in the extremely long-toothed calyx there are suggestive resemblances (cf. Pl. CXIII, CXXIII) between D. sericocalyx and D. (Versicolores) lamprostachya, another rare species described from Sierra Madre inland from Mazatlan. The leaflets of D. lamprostachya are much fewer, narrower, and carnosulous, the bracts persist between the flowers, and the calyx itself is sessile; the similarities may be purely coincidental, but the point will need reconsideration when fresh flowers of D. sericocalyx are secured. The curious elongate glands on the forming pod of D. sericocalyx have no known match in ser. Versicolores; the ripe fruit will hold much interest.