YASIN J. NASIR
Androsace villosa var. robusta f. longiscapa Knuth
Fl. Per.: June-July.
Lectotype designated: Tibet, 12-14000', 29.6.1848, T. Thomson s.n. CGE! (isolectotype at K! E! M! BR! G!).
Distribution: Baltistan, Ladak and eastward to Lahul.
The ssp. robusta differs from the European Androsace villose in its robust habit and with larger plant parts. Its closest ally is Androsace jacquemontii, which is treated here as a subspecies. J.D. Hooker (l.c. 499) found it difficult to distinguish it (i.e. Androsace jacquemontii) from states of his Androsace villose (non L., and which I consider as identical with Androsace robusta). Knuth (l.c.) treated Androsace jacquemontii as a variety of Androsace villose. The only reliable character separating the two taxa is the rusty brown indument in Androsace jacquemontii, and the fact that it occurs geographically disjunct.
Knuth’s Types at Berlin were destroyed during the war in 1943. He (op. cit.) did not designate any Type for his f. longiscapa, but he cites seven syntypes. Except for Griffith's specimen from Afghanistan, I have examined duplicate sheets (i.e. isosyntypes). It is not likely that the Afghanistan specimen belongs to the f. longiscapa, as it is quite out of place geographically (the taxon ssp. robusta f. longiscapa is found in the trans Himalayan ranges).
The first specimen that Knuth cites (Kashmir, Falconer) and the last one (Potanin, from Shansi, N. China) are misidentifications. The former is Androsace duthieana and the latter is Androsace lehmanniana Spreng. Out of the remaining four specimens, which agree to the f. longiscapa, (Schlaginweit no. 5164; Duthie s.n. in 1892; Hooker and Thomson s.n. and Schlagintweit 6520), I have chosen T. Thomson’s collection (from Tibet, 12-13000', Ladak), as represented at the Cambridge Botany School Herbarium as the lectotype.
Knuth’s Type specimen of f. breviscapa of ssp. robusta (W. Himalaya, Strachey & Winterbottom 10) is no longer extant.
The species Androsace robusta has also been confused with and included in Androsace muscoidea by several authors (Handel-Mazzetti 1927, p. 163 and Gould 1982). The two taxa are quite distinct not only morphologically but also occupy to a greater extent different areas.
There is a form or at least a related taxon of the ssp. robusta in Nepal. At the BM, I have seen annotated specimens of it labelled as Androsace muscoidea f. longiscapa.
At the herbarium at Geneva (G), they are represented by Polunin, Syke’s and William’s nos. 1095; 2180; 2216 and Dombremez 2994. This taxon differs from the ssp. robusta in the compact leafy rosettes, leaves with a silvery sheen and flowers a deeper pink or violet-rose. The ssp. robusta occurs in dry exposed areas from altitudes of 2438-5182 m.
