Panarctic Flora

800201 Lagotis glauca Gaertn.

• Gaertn., Novi Comment. Acad. Sci. Imp. Petrop. 14: 534 (1770). Described from Kamtchatka (the Russian Far East).

2n= 22 (2x). - Asia (Himalaya), Far East (Japan). - Vasudevan (1975); Sakai (1934).

Geography: European (NE) - Asian (N) - amphi-Pacific-Beringian.

Notes: Murray, Elven, and Petrovsky: The mainly Asian genus Lagotis is present in the Arctic and near-arctic areas with two to four assumedly closely related taxa, variously treated as from one to four species. At least six species names have been applied through times for northern parts of this species or species aggregate, under the genus names Lagotis and Gynandra: "glauca", "gmelinii", "hultenii", "minor", "ovata", and "stelleri".

The priority species name is Lagotis glauca Gaertn. 1770 based on plants from Kamtchatka. This name belongs to a comparatively coarse plant distributed in the oceanic North Pacific regions, with broad, crenate blades where the transition between blade and petiole is abrupt, and with conspicuously large bracts. Hultén (1949a) also noted short filaments and large anthers, confirmed by Murray & Elven (2010a). This plant is diploid (2n = 22). The names Gymnandra ovata Willd. 1811 and G. gmelinii Cham. & Schltdl. 1827 probably belong to this taxon, perhaps also some of the others.

More northernly distributed plants are present in a broad range from northeastern European Russia throughout Siberia and the northern Russian Far East to American Beringia. They are less coarse, with narrow, often more sharp-toothed blades with cuneate base and gradual transition to petiole, with small bracts, and according to Hultén (1949a) and Murray & Elven (2010a) with long filaments and small anthers. The illustrations of Hultén (1968a: 804-805) are characteristic for the southern and northern plants. Even if the differences between L. glauca s. str. and the northern plants are rather pronounced, these two are connected by some intermediates in the coastal North Pacific regions.

The variation among the northern plants is intricate and has been treated in several ways. The morphological differences are small and might not alone justify any racial segregation. However, there are two ploidy levels. At least four independent reports support that the plants from northeastern European Russia east to northern Yakutia are tetraploids (2n = 44), whereas at least nine independent reports support that the Beringian plants (Chukotka and northwestern North America) are diploids (2n = 22).

The four species names assigned to the northern plants are all based on Eurasian plants: Gymnandra minor Willd. 1811 described from northeastern Siberia, probably the lower Lena River area (within the range of the tetraploids), G. stelleri Ledeb. ex Spreng. 1825 and G. stelleri Cham. & Schltdl. 1827 both based on a Steller specimen described to have been collected in "Sibiria inter Lenam et Oceanum the Okhotsk Sea" (probably within the range of the diploids), and Lagotis hultenii Polunin 1951 described from the northern Urals (within the range of the tetraploids). Note that G. stelleri Cham. & Schltdl. is an illegitimate homonym. There are four proposals for handling this variation:

(A) The plants in the northern Urals and surroundings as "hultenii" vs. those from Taimyr east to the Mackenzie River as "minor". Hultén (1968a, 1968b) accepted L. hultenii as a relatively local species in the northern Urals (northeastern European Russia and northwestern Siberia) and assigned the material from Taimyr to the Mackenzie River to L. glauca subsp. minor. This implies that he drew the dividing line in the middle of the range of the tetraploids but we should remember that very few chromosome counts had been made at his time. His proposal for names is, however, unacceptable. He misapplied the name L. hultenii as this name was not based on plants from the Urals but from Taimyr (Jenisei), i.e., outside the range he reported for it but within the range he reported for L. glauca subsp. minor. He synonymized Gymnandra stelleri Cham. & Schltdl. with L. glauca subsp. minor.

(B) The tetraploids from the northern Urals to northern Yakutia as "minor" (incl. "hultenii") vs. the diploids in Chukotka and northwestern North America as another race (by Gjærevoll 1967 as L. glauca subsp. stelleri). This proposal has support from the chromosome counts and was implied in the treatment by Porsild and Cody (1980: 530) when they named the North American plant L. stelleri and stated it to be amphi-Beringian.

(C) The non-Beringian Eurasian plants as one taxon (to which all the names "hultenii", "minor", and "stelleri" belong) vs. a nameless Chukotkan and northwestern North American plant. This rather strange treatment is the one implied in the treatment of Hultén and Fries (1986: 1121, map 1639). These authors mapped and commented on L. minor as exclusively Russian-Siberian. The map indicated a very disjunct distribution with one locality group in the northern Urals (previously as L. hultenii, replaced southwards in the Urals by L. uralensis Schischk.), two groups in the Lena River drainage, and one group in the Kolyma River drainage. This treatment implies that they synonymized L. hultenii with L. minor but that they explisitly excluded from L. minor the massive occurrence in Chukotka and in northwestern North America, i.e., in Beringia. With a European viewpoint, Webb (1972) treated the variation in the same way. He accepted L. minor throughout northern Eurasia from northeastern European Russia (thereby including L. hultenii) to northern Siberia, but not into North America. He synonymized the name L. stelleri with L. minor, i.e., he implied also this name not to be relevant for the Beringian plant(s). Webb accepted a L. uralensis in the central and southern Urals. Vydrina (1996) followed the same approach for Siberia. Neither Hultén and Fries (1986) nor Webb (1972) gave any reason for why they excluded the Chukotkan and northwestern North American plants from L. minor.

(D) All the northern plants as one species (L. minor) or as one race (L. glauca subsp. minor). The most extensive treatment after Hultén (1968a, 1968b) is that of Petrovsky (1980b). He accepted, besides a Pacific L. glauca subsp. glauca, a very widely distributed subsp. minor from the Urals to the Mackenzie River with "stelleri" and "hultenii" as synonyms, and he mapped it with a nearly continuous and well documented range from the Urals to the Bering Strait. Petrovsky's treatment and map was available six years before Hultén and Fries (1986) stated this plant to be "disjunct, incompletely known ... in N. Russia and Siberia", throwing some doubts on the knowledge behind the Hultén/Fries treatment. Ivanina (1991) followed the same approach in principle but accepted two species for the Russian Far East, L. glauca and L. minor, the latter as the major northern plant. Cody (1996) treated the northern plant as L. glauca subsp. minor with the combinations with "stelleri" as synonyms.

For the Checklist, we follow proposal (B) for the following reasons. We have compared material (ALA, O, S; field) from throughout the "diploid" range in Chukotka, Alaska, and the Yukon Territory and find no morphological differences whatsoever. We have compared material throughout the "tetraploid" range in northeastern European Russia and northern Siberia (specimens seen from Novaya Zemlya, Gydan, and the lower Jenisei and Lena rivers) and find it morphologically quite uniform. We have also compared material from the "diploid" and "tetraploid" ranges and find the plants fairly similar morphologically but perhaps separable in a few minor characters. They are, however, much more similar than are these plants compared with the Pacific L. glauca s. str. For morphological reasons, we cannot accept the tetraploid L. hultenii as a species as long as the diploid Beringian and Pacific plants are treated as two subspecies, in spite of the ploidy difference. The origin of the tetraploid in the northern Urals to northern Siberia is unknown, i.e., whether it is an autotetraploid or an allotetraploid. Southwards in the Urals, it is replaced by L. uralensis, but the ploidy level of that species is not known (Vydrina 1996) and its partaking in an allotetraploid origin purely hypothetical. There seems to be two morphologically slightly different races, one tetraploid in northeastern European Russia and northern Siberia, and one diploid in northeastern Asia and northwestern North America. We provisionally accept them as two subspecies on the basis of the ploidy difference but a merger is the solution if no better morphological distinctions are found. The western tetraploid is subsp. minor (with L. hultenii as a synonym), the eastern diploid has been problematic to name (see subsp. lanceolata).

We also find the relations between subsp. glauca and the northern diploid in the Bering Strait area to be different from that reported by Hultén (1968a, 1968b). He stated that "transitions between the two races by him as subsp. [glauca and subsp. minor] occur frequently where the ranges overlap". Based on a survey of specimens (ALA, O, S; the last-mentioned includes Hultén's specimens), we found that we could fully accept two races but circumscribed these differently from Hultén (and from Petrovsky 1980b and Sekretareva 1999). In our opinion, subsp. glauca is very limited in Alaska, only documented by vouchers (ALA) from the southwesternmost parts north to the Pribilof Islands (St. George Island, non-arctic), St. Matthew Island (arctic) and Cape Newenham (borderline arctic). Transitions towards the northern diploid occur as more widespread in south-central and southwestern Alaska north to the Seward and perhaps Chukchi peninsulas (still well south of the northern limit of subsp. glauca as mapped by Hultén) and surprisingly also in Coronation Island in southeastern Alaska. The number of intermediate specimens convinced us that rank as major races (subspecies) is more appropriate than as species.