Rocky Mountain or oblong-leaved or marsh gentian

Pneumonanthe affinis (Grisebach) Greene

Herbs perennial, 0.5–8 dm. Stems 1–10(–20), terminal from caudex, decumbent to erect, glabrous or puberulent in lines below leaf bases or more extensively. Leaves­ cauline, variably spaced; blade oblong or elliptic to ovate, lanceolate, or nearly linear, 1–5 cm × (2–)3–20(–25) mm, generally 2+ times as long as wide, margins ciliate, apex obtuse to acute. Inflorescences racemoid thyrses of ± dense 1–6-flowered cymules, terminating main stem and usually on short branches at distal 1–6(–12) nodes. Flowers: calyx 5–18(–23) mm, tube occasionally deeply cleft, lobes linear to narrowly elliptic-lanceolate or occasionally some rudimentary, (0–)1–13 mm, margins ciliate; corolla blue, sometimes with pale dots adaxially on lobes, or rarely pale violet or white, tubular-funnelform, open, (12–)18–40(–45) mm, lobes ± spreading, oblong-ovate, 3–7(–10) mm, free portions of plicae divided less than 1/2 their length into 2 ± triangular, lacerate segments; anthers distinct. Seeds winged.

Varieties 2 (2 in the flora): w, c North America.

Gentiana affinis is highly variable, and some authors have divided it into several species, or have recognized more varieties than the two accepted in this flora. N. H. Holmgren (1984b) speculated that further studies might disclose patterns that would warrant the recognition of additional infraspecific taxa in G. affinis. From studies for the present and earlier works, however, only two varieties appear to be well differentiated, and even these intergrade in Idaho, Montana, Oregon, and northern California, especially in Del Norte County, California. Despite the very different vegetative aspects of the extremes, which sometimes occur more or less sympat­rically, intergradation is so extensive that the recogni­tion of these taxa as distinct species seems inappropri­ate. The treatment of G. affinis presented here, although based on an independent approach involving the study of numerous specimens, is essentially in agreement with that of Holmgren and is supported by biometric analy­ses by J. R. Spence (unpubl.), which confirmed the inter­gradation between var. affinis and var. ovata.

Only var. affinis approaches the range of Gentiana puberulenta. Large plants of G. affinis generally differ from small plants of G. puberulenta in the proportionate lengths of the corolla lobes and the free portions of the plicae. In G. affinis, the lobes are less than two times as long as the free portions of the plicae; in G. puberulenta they are generally two or more times as long.

Lower Taxa

• Gentiana affinis var. affinis
• Gentiana affinis var. ovata A. Gray