[2.7] Viola sect. Himalayum 

Viola sect. Himalayum Marcussen, sect. nov.

Type: Viola kunawurensis Royle

Description.—Dwarf perennial herb. Axes not morphologically differentiated. Stems subterranean from deeply buried pleiocorm, appearing aboveground as proximal or tufted rosettes. Stipules adnate to ¾ of their length. Lamina subentire with 0–2 shallow crenulae, spathulate, gradually tapering in a long petiole. Corolla c. 10 mm, light violet with dark striations. Lateral petals beardless. Spur as long as tall, saccate, 1–1.5 mm, obtuse. Style clavate, unmargined. Cleistogamous flowers produced; cleistogamy seasonal. Allo-octoploid (CHAM + MELVIO). Secondary base chromosome number x’ = 10 (needs confirmation). ITS sequence of MELVIO type.

Diagnostic characters.—Stipules adnate AND lamina spathulate and subentire AND spur as long as tall, 1–1.5 mm AND cleistogamous flowers produced.

Ploidy and accepted chromosome counts.—8x; 2n = 20?

Age.—Crown node age not applicable (monotypic section), stem node age probably 17.8–19.3 Ma.

Included species.—1. 

Viola kunawurensis Royle

Distribution.—High mountains surrounding the Tibetan Plateau: Tian Shan, Pamir, the Himalayas, Hengduan Shan, and Qilian Shan (Figure 1)

Etymology.—The name Himalayum refers to the distribution in the Himalayas and adjacent mountain ranges.

Discussion.—Section Himalayum comprises a single species, V. kunawurensis (=V. “kunawarensis”, V. thianschanica Maxim.), occurring at high elevations (3000–5000 m) in the Central Asian high mountains surrounding the Tibetan plateau (Figure 1). Viola kunawurensis differs from similar species of sect. Plagiostigma subsect. Patellares in having a very short spur and frequently elongated internodes arising from the deep-buried pleiocorm, as well as in chromosome number, and from sect. Spathulidium in style shape and in producing cleistogamous flowers.Mining GPI sequences from the sequence reads archive of the reference sequence genome of V. kunawurensis (as V. “kunawarensis”; NCBI accession PRJNA805692) strongly indicates the presence of four homoeologs, confirming that sect. Himalayum is an independent CHAM + MELVIO allotetraploid lineage and further suggesting that the extant species is octoploid as a result of a secondary autopolyploidisation . The single chromosome count of 2n = 20 [2] is doubtful as this number reflects 4x in sect. Viola and sect. Delphiniopsis and therefore seems at odds with the octoploid condition of V. kunawurensis.Becker originally placed Viola kunawurensis in grex Gmelinianae [3], later in grex Adnatae [1]; see note under sect. Plagiostigma subsect. Patellares. Sun and coworkers placed V. kunawurensis in sect. Viola subsect. Rostratae based on the (allegedly) shared chromosome number 2n = 20 [2] and numerical taxonomy of 58 traits [4].

References.—

1. Becker, W. Viola L. In Die Natürlichen Pflanzenfamilien 21. Parietales und Opuntiales; Engler, A., Ed.; Wilhelm Engelmann: Leipzig, Germany, 1925; Volume 21, pp. 363–376. [Google Scholar]

2. Sun, K.; Wang, C.-j. Notes on the Chinese Violaceae (II)—Cytological studies on some violets of China. Bull. Bot. Res. 1991, 11, 69–72. [Google Scholar]

3. Becker, W. Violae Asiaticae et Australenses. II. Beih. Bot. Centralbl. Abt. 2 1917, 34, 373–433. [Google Scholar]

4. Sun, K.; Wang, Q.; Wang, Y. Numerical taxonomic study of Viola in China. J. Northwest Normal Univ. (Nat. Sci. Ed.) 1997, 33, 69–73, (In Chinese, with English Abstract). [Google Scholar]