Stroganowia leventii

German, Dmitry A., 2014, Taxonomic remarks on some Asian Lepidium s. l. (Brassicaceae), Phytotaxa 186 (2), pp. 97-105 : 101-102

3. Stroganowia leventii and S. kazachstanica

Revision of the status of these two recently established species was necessitated by the fact that they were described as representatives of Stroganowia , a genus proven to be non-monophyletic with all clades nested within Lepidium ( Mummenhoff et al. 2009) , and none of the two has been yet transferred to the latter genus. A study of available material demonstrated that only one of them is a distinct species distributed surprisingly much wider than previously considered while the other is recognized as a new synonym of another species of Lepidium .

Lepidium leventii (V.I. Dorof.) D.A. German , comb. nov. ≡ Stroganowia leventii Dorofeyev in Dorofeyev et al. (2004: 72). Type:— TURKEY. Erzurum Prov., ad meridiem 42 km ab Köprüköy, in parte superiore vallii Araks , declivum herbaceum, altitudo supra mare 1400 m, 39°44’ lat., 41°43’ long., 18 June 2002. V.I. Dorofeyev 999 (holotype LE!, isotypes LE (4×)!).

= Stroganowia leventii var. pilosa Dorofeyev (2006: 129) . Type:— TURKEY. Erzurum Prov., ad meridiem 42 km ab Köprüköy, in parte superiore vallii Araks , declivum herbaceum, altitudo supra mare 1400 m, 39°44’ lat., 41°43’ long., 23 June 2003. V.I. Dorofeyev 1771 (holotype LE!, isotypes LE (6×)!).

Stroganowia leventii V.I. Dorof. (subsequently Lepidium leventii ) has been described as a local endemic of northeastern Turkey known from the single population ( Dorofeyev et al. 2004, Dorofeyev 2006). It was compared with SE Iranian/Afghanian L. affghanum (Boissier 1867: 358) [as S. affghana (Boiss.) Pavlov (1933: 369) ] and NW Iranian L. buschianum Al-Shehbaz in Al-Shehbaz et al. (2002: 8) [as homotypic S. persica Busch (1926: 225) ] and distinguished from the prior by “2–3 times wider basal leaves” and from the latter by “basal leaves twice narrower, with elliptic vs. ovate blade” ( Dorofeyev et al. 2004). Comparison with these and other species demonstrates that L. leventii is most close morphologically to also geographically closest L. buschianum sharing with the latter the habit, leaf texture and nervation, long divaricate-ascending pedicels, fruits of similar shape (predominantly broadly elliptic) and size (4–7 × 3–4 mm) with stort but distinct style, presence of long crisped trichomes on leaves (at least on leaf margins proximally and usually petioles), pedicels, and sepals, etc. The basic difference among the two species is morphology of basal, lower and middle cauline leaves which are ovate to ovate-elliptic, suddenly narrowed to a petiole (base of some lowermost leaves can be even slightly cordate), rounded at the apex and always integerrime in L. buschianum vs. narrowly elliptic to broadly lanceolate or rarely lanceolate, gradually narrowed to a petiole, usually acute (rarely some round) at the apex, biggest basal ones in the distal part often with shallowly crenate margin.

Revision of material stored under the name S. persica resulted in a finding that relevant specimens are distinct taxonomically from the type of L. buschianum but morphologically best fit L. leventii . Thus, the latter is not a local endemic of NE Turkey but also occurs in central western Iran where the main part of its distribution area is situated. The whole range of the species appears highly fragmented with several localities in eastern Lorestân (Luristan) and probably adjacent westmost Esfahân approximately 1000–1050 km to the south-east from the locus classicus and the two localities (Galli Sur and Kuh Rang) in Chahar Mahal va Bakhtiari province almost 1200 km apart from the Turkish population. Despite such a huge gap, I failed to find any features separating Turkish and Iranian plants; therefore, they are assinged to the same species, L. leventii . In contrast, based on the above findings, L. buschianum appears a locally distributed species still known (at least based on the material of all studied herbaria) only from the type garhering [ IRAN. Persia borealis, prov. Tabris (Atropatania), in montibus Sher-Dara inter st. viae ferr. Sofian et Sejvan, 1430– 1500 m. 18 VI 1924, A. Grossheim s.n.; lectotype designated by Botschantzev 1984: 76, as type ( LE!, isolectotypes BP 136088!, GH 00105286, M!)]. This locality is situated in between the Turkish and Iranian populations of L. leventii and its formation via bottleneck effect based on an isolated population of the latter species or their common ancestor can be assumed which agrees with apparently relic character of the distribution of L. leventii . Molecular study would be desirable for testing this scenario, the monophyly of L. leventii as delimited herein and elucidating the evolutionary history of relevant species group.

Only two gathering of “ S. persica ” (Dow Rud, W. Koelz 15776, W!, E: http://data.rbge.org.uk/herb/E00040397 and Galli Sur, W. Koelz 18101, W!) were reported in Flora Iranica and, as long as neither the type material nor original publication could be studied while preparing that treatment, Hedge (1968: 121) correctly noted that relevant determinations can only be treated as provisional. With more data available it is corrected here to L. leventii .

Additional Iranian specimens examined (all sub nom. Stroganowia persica ):— IRAN, [Border of Lorestân and Esfahân]. Persia: W: Luristan. In convallibus borealibus montium Khali Kuh 50–60 km ab Aligudarz meridiem versus, substr. calc., 2300–2800 m. 12–14 June 1974, K.H. Rechinger 47895 (B: http://herbarium.bgbm.org/object/ B100184691!, G!, M!, W!, WU, photo!) & 48000 (G!, W!); [Lorestân]. Persia: W: Luristan. Oshtoran Kuh, substr. calc. 16–17 June 1974, J. Renz 48210 (G!, M!, W!); [Chahar Mahal va Bakhtiari] 12 km SE tunnel Kuh-Rang, pente calcaire à buisson. 7 July 1959, alt. 2360 m. H. Pabot 2197 (G!) .

Dorofeyev (2006) recognized a separate variety, S. leventii var. pilosa V.I. Dorof. characterized by moderately pubescent (instead of glabrous) stem and leaves. Indeed, one can distinguish two groups of plants though based predominantly on the qualitative character of indumentum rather than its intensity. More precisely, margins and, in young leaves, also surfaces of var. leventii are long crisped pilose (like in L. buschianum and in Iranian specimens of L. leventii ) while in var. pilosa such trichomes are restricted to margins but surfaces are mainly covered with shorter, straight to slightly flexuous, often subappressed ones which are also present on margins. On the other hand, sparce short straight trichomes can be found on both stems and leaves of some specimens corresponding to var. leventii and thus there is no clear limit between the two entities. Therefore, recognition of var. pilosa does not look reasonable and it is not accepted here.

Lepidium cardiophyllum ( Pavlov 1933: 364) Al-Shehbaz in Al-Shehbaz et al. (2002: 8). Type :— KAZAKHSTAN. [South Kazakhstanian prov.]. Mts. Talas-Alatau, stony meadow slope on the pass Dau-baba, 1200 m. 31 August 1931. N.V. Pavlov 1226 (holotype MW 236552 !, isotype LE!).

= Stroganowia kazachstanica Karpova (2006: 131) , syn. nov. Type :— KAZAKHSTAN. [South Kazakhstanian prov.]. Mts. Duona-tau between Seslavino and Tyukulbas. 9 August 1982. M.G. Pimenov, M.G. Vasilyeva & S.S. Daushkevich 51 (holotype MW!).

Lepidium cardiophyllum (Pavlov) Al-Shehbaz is confined to a rather limited part of NW Tian Shan of southern Kazakhstan and adjacent Uzbekistan ( Botschantzev 1984, German et al. 2013). Among all Middle Asian species previously assigned to Stroganowia this one is peculiar for its basal and lower cauline leaves with cordate bases and profoundly crenate margins. Other stem leaves are gradually diminishing and narrowing upward, their margins become entire and bases cuneate, gradually narrowed to a short petiole (uppermost leaves are subsessile).

Stroganowia kazachstanica A. Karpova was initially compared with similar but apparently not very closely related W Pamir Alai endemic L. minor ( Botschantzev & Vvedensky 1941: 19) Al-Shehbaz in Al-Shehbaz et al. (2002: 9) [as S. minor Botsch. & Vved. ] ( Karpova 2006). Later, based on more detailed morphological and anatomical studies ( Karpova & Yakovleva 2012a, b), it was found closest to L. cardiophyllum [as S. cardiophylla ] and these two species were assigned to a separate group ( Karpova & Yakovleva 2012b) recognized as a not effectively published “ Stroganowia sect. Stroganowia subsect. Cardiophyllae A. Karpova ” ( Karpova 2008).

Revision of available material on S. kazachstanica and potentially related taxa lead me to a similar conclusion. The species is only known from the type gathering represented by the single plant lacking lowermost leaves. This looks the main reason why it was not identified as L. cardiophyllum because in the rest of morphological characters (leathery leaves, relatively short, to 1 cm, only slightly divaricate pedicels, elliptic to oblong-obovate fruits with a style 0.5–0.8 mm long, pink petals, pubescence of petioles and pedicels with short straight trichomes and sepals covered with very thin crisped trichomes) it is indistinguishable from the latter species. Therefore, S. kazachstanica is treated here conspecific with L. cardiophyllum . It can be added that the type localities of both taxa are situated ca. 20 km apart from each other in similar ecological and biogeographical conditions.