Taraxacum lilacinum (Krasn.) Schischkin (1937: 4)

Kirschner, Jan & Štěpánek, Jan, 2017, A revision of Taraxacum sect. Atrata, a dandelion group centred in the Middle Asia, and the problem of Taraxacum brevirostre, Phytotaxa 305 (4), pp. 225-261 : 230-231

1. Taraxacum lilacinum (Krasn.) Schischkin (1937: 4) ≡ Taraxacum glabrum var. lilacinum Krasnov (1887: 76) [as

“lilacina”].

Type indication (see Fig. 2 , and Note 2):—Prope Ketmenpass et fl. Kuelu.

Type:— KYRGYZSTAN. “Prope glacies aeternas Chan Tengri“, 1886, Krassnow [Krasnov], [under the name (sic!) “ Tragopogon glabrum v. lilacinum Krass. ”] (LE, no. det. 6511, lectotype, fide Schischkin & Tzvelev 1964: 530).

Note 1 (typification):— Schischkin (1937) probably was aware of the validity of the varietal name published by Krasnov (1887, see Fig. 2 ) and introduced a new combination only. He published a new, more complete description of his plants and, importantly, he designated a lectotype, allegedly deposited in LE (judging from the introductory phrase of Schischkin, 1937: 1: [translated from Russian] “When elaborating the Kazakhstan representatives of the genus Taraxacum in the ample material preserved in the herbarium of Institute of Botany, Academy of Sciences of the USSR, .... I succeeded in observing a few species I was unable to accomodate within the framework of known species. ” Schischkin (1937) cited a plant coming from the Ketmen Tau Range, from the border region of Kazakhstan and China. However, we were unable to find the Ketmen Tau plant in LE, nor was it seen by Handel-Mazzetti (1907) or Tzvelev. The other possible syntype specimen (fl. Kuelu, most probably Kyulyu River in E. Kyrgyzstan, approximately at 42° 10‘ N, 79° 05‘ E) is not extant, either. Thus, in accordance with Art. 9.2 and 9.3 of the Code ( McNeill et al. 2012), the lectotype must be selected from among the other elements of the original material, those not cited in the protologue. The type selected by Tzvelev in Schischkin & Tzvelev (1964) does not come from the Ketmen Tau but from the region of the second highest peak of the central Tian Shan, Khan Tengri (7010 m), in the Terskey Alatau, Kyrgyzstan- Kazakhstan border.

In the herbarium LE, there is another, rather incomplete specimen of T. lilacinum collected by Krasnov, with the locality “Prope Sary Jassy“, which, in all likelihood, means the Sarydzhaz River in NE Kyrgyzstan, not far from Mt. Khan Tengri (see the list of specimens seen).

Note 2: —The layout of the account of the flora of the eastern Tian-Shan in Krasnov (1887) is not uniform but, as a rule, a list of localities immediately follows the name of each taxon (species or variety) or its diagnosis. Then, as a separate sentence, overall distribution is given. As a typical example, we can quote Camphorosma ruthenicum : “Prope Ilijskoje non rara, v. erecta Krassn. [... diagnosis]. Prope Ili non procul a Chorgos. Ar. g. Rossia media et australis ...”. However, the protologue of T. glabrum var. lilacinum makes an exception. Here the localities are listed after the species name ( T. glabrum ), then (after a comma) follows the name and a diagnosis of the variety. The paragraph is completed by a brief information on the geographical range (Tianschan) that obviously refers only to the variety (because T. glabrum was described from southern Siberia and probably is absent from the Tian Shan). Two interpretations are possible, first that the localities refer to T. glabrum s. str. and those of var. lilacinum were omitted by mistake, or secondly, the data order was changed in the case of var. lilacinum , and the whole text exclusively refers to the latter. We prefer the latter alternative as more plausible.

Description: —Plants small, 4–12 cm tall. Petiole pinkish to purple, long, narrow, unwinged; leaves deep green, ± spatulate-oblanceolate, 2.5–8 × 0.5–1.7 cm, glabrous, undivided, often entire, not rarely remotely dentate with 2–4 pairs of short to lobulate teeth, apex subobtuse. Scapes purplish brown, glabrous, usually overtopping leaves. Capitulum 2–3 cm wide. Involucre 6–9 mm wide, base rounded to narrowly rounded. Outer phyllaries (8) 9–11 (13), ± not imbricate, ± lanceolate to narrowly oblong-lanceolate, usually broadest in the middle, outermost ones 4–5.5 (–8) × 1.5–2.3 mm and 1/2–2/3 as long as inner ones, appressed, sometimes unbordered, usually with a paler to whitish border (0.1–) 0.2–0.3 (–0.5) mm wide, with ± abrupt transition into deep blackish green middle part, margin glabrous, apex flat; inner phyllaries 1.1–1.2 (–1.4) cm, apex ± flat. Ligules deep pink to light pinkish violet from both sides (see Kirschner & Štěpánek 1993, Plate 1(a); outer ligules flat, ± not striped or faintly striped deeper pink or purplish outside, inner ligule teeth short, purple or black-purple. Stigmas greyish green, with blackish pubescence outside. Anthers polliniferous, sometimes very sparsely so; pollen grains irregular in size. Achene ( Fig. 10F ) light (greyish) stramineous-brownish, 4–4.5 × ca. 0.9 mm, smooth or with sparse minute squamules or tubercules above, zone between the rugulosity and the achene top ca. 0.5–0.6 mm long, but cone indistinct, body ± gradually narrowing into beak; beak 2.5–3.5 (–4) mm, thickened (ca. 0.3 mm). Pappus ± white, 5–5.5 mm.—Agamospermous.— Fig. 4A , see also Kirschner & Štěpánek (1993: Plate 19a).

Taxonomic notes: —The name T. lilacinum was generally used for the members of T. sect. Atrata with aberrant flower colours. Here the name is restricted to plants having the following diagnostic features: Leaves undivided, usually spatulate-lanceolate; outer phyllaries appressed, narrowly lanceolate to ± lanceolate (not ovate-lanceolate), variably but usually bordered; achene without developed cone, sparsely minutely squamulose above, light stramineous-brownish (not grey); beak relatively short. Most importantly, the other species have outer ligules pink-purple and striped outside and paler, lilac to pale yellow inside, while T. lilacinum has uniformly deep pink florets, the outer ones almost not striped outside (see Kirschner & Štěpánek 1993, Plate 19a, plants from the Gol-Ukok locality); it is a polliniferous plant, as a rule.

Ecology and distribution: — Taraxacum lilacinum is confined to wet alpine open slopes, sites with fine gravelly soils, vicinity of rocks, probably also alpine meadows; usually between 3000 and 3800 m. It is reported from a number of regions in Middle Asia but specimens studied come from Kyrgyzstan (only specimens studied are mapped on Fig. 5 ), but reliable literature records exist from Xinjiang part of the Tian Shan (see Ge et al. 2011, Zhai 1999). The protologue record from Ketmentau ( Kazakhstan / Xinjiang border region) is considered as uncertain.

Specimens examined: — KYRGYZSTAN: “Flora Iliensis. Taraxacum lilacinum Krassn. Prope Sary Jassy. Leg. Krassnow “ (LE).—Narynskiy rayon, bassein r. Naryna [Naryn District, basin of the Naryn River], khrebet Baidulla-tau, reka Karachyr [the Baidulla-tau Range, Karachyr River], 29 Jul 1926, M. Sovetkina & M. Uspenskaya 1434 (TASH, no. det. 15825).—Tonskiy r-n, syrty, ruslo reki Karagoman, v verkhovyakh [Tong District, river bed of upper Karagoman River], 18 Jul 1937, E. Sarkisova & F. Shamsutdinova (TASH, no. det. 15854).—W. Terskey Alatau, Gol-Ukok valley, near Kochkorka township, 3350 m, Aug 1988, R. Businský (PRA, no. det. 28204, 28205).

Figures

FIGURE 2. Protologue of T. glabrum var. lilacinum Krasn.

FIGURE 4. A, T. lilacinum (PRA, no. det. 28204), cultivated. For the general habit of plants from natural habitats, see Kirschner & Štěpánek (1993, Plate 19a). B, T. sublilacinum (PRA, no. det. 28194). Scale bar equals 3 cm.

FIGURE 5. Distribution ranges of Taraxacum chionomelas (square), T. corvinum (triangle), T. lilacinum (circle), T. pseudoatratum (pentagon), and T. sublilacinum (hexagon).

FIGURE 10. Achenes of the members of T. sect. Atrata: A, T. alpigenum (PRA, no. det. 28132); B, T. atrochlorinum (PRA, #72); C, T. corvinum (PRA, #114/1); D, T. goloskokovii (PRA, #105/2); E, T. goloskokovii (LE, no. det. 6526); F, T. lilacinum (PRA, no. det. 28204). Scale bar = 1 mm.