Hypericum erectum (Nomenclature)
Perennial herb (0.1)0.3–0.8(–1) m tall, erect or ascending from branching and rooting woody base, with stems solitary to few (2–3) or sometimes caespitose, unbranched below inflorescence or sometimes virgately branched above. Stems slightly 2-lined when young, soon terete, eglandular; internodes 14–55 mm, shorter to longer than leaves. Leaves sessile; lamina (12–)15–50(–60) × (5–)7–15(–30) mm, narrowly triangular-ovate to narrowly elliptic, paler beneath, chartaceous; apex acute or attenuate to rounded or slightly emarginate, margin plane or recurved, base cordate-amplexicaul; venation: 3–4(–5) pairs of main laterals from lower third to half of midrib, tertiary reticulation rather dense; laminar glands black, punctiform, ± dense, small, rarely also pale, small, punctiform to shortly striiform; intramarginal glands black, dense and rarely pale, minute. Inflorescence many-flowered, from 1–3 nodes, often with branches from up to 6 nodes below, the whole corymbiform to cylindric; pedicels 1.5–3 mm; bracts and bracteoles reduced foliar, entire. Flowers c. 15 mm in diam., substellate; buds narrowly ellipsoid, acute. Sepals 5, subequal to unequal, c. 3.5–7 × 1–3 mm, ovate-lanceolate to oblong-elliptic, obtuse (or rarely rounded) to acute, entire or rarely with irregular prominent glands; veins 5, outer often branching; laminar glands black or rarely also pale, linear to punctiform; intramarginal glands black, rather dense to sparse, or absent. Petals 5, bright yellow, not tinged red in bud, 6–10 × 2–2.5 mm, 2.5–3 × sepals, obovate to oblong, entire or with scattered prominent glands; laminar glands black, striiform to punctiform, distal; marginal glands black, intramarginal to prominent, irregular. Stamens c. 25–40, ‘3’-fascicled, longest 5–7 mm, c. 0.7 × petals; anther gland black. Ovary 2.5–3.5 × 1 mm, ovoid; styles 3, free, widely divergent, (3–)3.6–4(–4.5) mm, 1–1.2(–1.5) × ovary; stigmas narrowly capitate. Capsule 5–11 × 3.5–6 mm, 1.2–2.5 × sepals, narrowly to broadly ovoid. Seeds mid-brown, (0.67–)0.8–1 mm; testa scalariform.
2n = 16 (Suzuka, 1950; Kogi, 1984).
Grassy slopes; 450–2250 m (China), 100–2100 m (Japan) [Dry woods and thickets – Gorschkova, 1974: 180].
Russia (southern Sakhalin, incl. Shikotan I.), Korea (south and Cheju-dō [Quelpart I.]), Japan (incl. Ryūkyū Is. and Okinawa), Taiwan (rare), China (southeastern). Masamune (1934) includes ‘Manchuria’, but the species is not in Noda (1971).
The above synonyms are those that have not been allocated to infraspecific taxa. For others see below.
Hypericum erectum is very variable and widespread (from Sakhalin and the southern Kurile Is. to Taiwan and eastern China), but its origins seem to have been in or near Hokkaidō. Its nearest relative (30. H. nuporoense) is so far known from northern Rumoi province only, and its most primitive form (f. vaniotii) is confined to Sakhalin, Kunashiri, Hokkaidō and northern Honshū.
The wide variation in H. erectum has resulted in a long synonymy. It does not seem possible, however, to recognise discrete infraspecific taxa, as the variation is continuous. Kimura (1951) recognised six varieties of which three are removed from H. erectum here (vars.subalpinum, parviflorum and longistylum) and two are extreme variants of that species (var. caespitosum with many relatively short stems rather than only one or rarely two to three; var. deviatum with leaves, sepals and petals with pale striate laminar glands as well as black, the petals with some prominent marginal glands). The rest of the variation he accommodated in var. erectum with eight formae. Some of these (e.g. f. vaniotii or f. angustifolium) are parts of general morphological trends, while others (e.g. f. perforatum or f. papillosum) appear to be probably the result of single mutations.
In general, there is a north–south trend from large, broad-leaved plants (f. vaniotii), through intermediate forms (f. erectum) to small, narrow-leaved plants (f. angustifolium), with f. erectum also present in Korea and China. Kimura includes the Taiwanese H. taisanense in the synonymy of f. angustifolium; but this appears to be a reduced form of f. lutchuense (Yakushima and Okinawa Is.), which differs from f. erectum merely in its robust habit and sometimes in having minute pale marginal leaf and sepal glands as well as the black ones.
Rather than treat H. erectum as a single variable taxon, I have keyed out Kimura's varieties and forms and allocated synonyms to them where possible. The specimens, however, are cited in a single list.
It appears that, although Murray's Syst. Veg. edn. 14 was published in July 1784, before Thunberg's Flora Japonica (August 1784), Murray's text was provided by Thunberg (Bartholomew et al., 1997, p. 311). Thunberg must therefore be accredited with the authorship of the new names in his Flora (Greuter, 1994, Art. 46.2).
Kimura (1951) cited ?H. choji-suzukii as a possible synonym of H. penthorodes.