Salicicola Translations

A New System of the Genus Betula L.—the Birch

A. K. Skvortsov

Bull. Mosc. Natur. Soc. 2002, Otd. Biol., v. 107 (5): 73–76
The literature treating the systematics of the birches is quite extensive. However, most of it is dedicated to problems of species size, species delineation, or descriptions of new species and interspecific hybridization studies. The author of this article has been tackling the same range of problems for many years mostly on the material collected in northern Asia. These, however, represent only a part of the genus, that is, undoubtedly, its youngest part.
While studying material for the new Flora of China (1999), the authors noticed deficiency and inadequacy in the infrageneric systematics of the genus Betula. Being time restricted by the project deadlines, the co-authors (Li Peiqiong and the author of this article) decided to refrain from delineating any infrageneric groups. J. J. Furlow, the author of Betulaceae treatment in the Flora of North America (1997), did not provide any infrageneric grouping, either—and probably for the same reason. Here I am trying to fill out this gap while relying on my previous work.
In 1865 E. Regel proposed the division of the genus in 7 sections [1]:
1. Albae (B. alba L. sensu latissimo with 9 subspecies and B. microphylla)
2. Fruticosae (B. fruticosa, B. middendorffii)
3. Nanae (B. nana, B. michauxii, B. glandulosa, B. pumila, B. humilis)
4. Dahuricae (B. dahurica)
5. Costatae (B. schmidtii, B. nigra, B. ermanii, B. grossa, B. utilis, B. corylifolia)
6. Lentae (B. lenta, B. alleghaniensis)
7. Acuminatae (B. cylindrostachya, B. alnoides, B. maximowicziana) [2]
Here the Fruticosae/Nanae parts look particularly inadequate. Also, the section Costatae appears quite superficial, since the author dumped there all exotic species that were poorly known in Europe at the time. Regel outlined the sect. Lentae basing on the fact that B. lenta retains its pistillate aments intact into the winter season, which he considered somewhat reminiscent of Alnus.
The next important step was made by C. K. Schneider (1915, 1916). While working on Plantae Wilsonianae, Schneider had at his disposal vast newly collected material, particularly from East Asia. He divided the section Costatae in 7 subsections:
1. Nigrae (B. nigra alone)
2. Corylifoliae (Bcorylifolia alone)
3. Asperae (Bschmidtii and Bmedwediewi—with a question mark)
4. Ermannianae (Bermanii, Butilis, B. albo-sinensis, B. jacquemontii)
5. Grossae (B. grossa, B. fargesii, B. insignis, B. globispica, and—with a question mark—Bcostata)
6. Lentae (Blenta, Balleghaniensis)
7. Chinenses (B. potaninii, B. delavayi, B. chinensis)
However, while proposing his new system for the genus, Schneider at the same time confessed: "Eine wirklich zeitgemäße Bearbeitung der Gattung Betula fehlt": "There is no adequate contemporary treatment of the genus Betula" (1915: 312), which enables us to conclude that he was not really satisfied with his own system. Indeed, his subsection Grossae consisted of three or four components that were not closely related, while the obvious close relation between B. grossa and the American B. lenta was not depicted at all. Subsect. Chinenses also appeared heterogeneous; a close relationship between B. dahurica and B. nigra was not noticed; the peculiarity of B. corylifolia was underestimated as it was considered only in the rank of the subsection, and so on.
T. Nakai (1915) and V. N. Vasilyev (1969) proposed a few subgenera, but those were essentially rank-changing proposals, once again based only on material from northern Asia. These proposals did not constitute much progress.
Finally, a recent treatment by Z. D. Chen (1994) has been a kind of step back from Schneider's system. Chen proposed five sections (no subgenera, neither subsections). Of these five, three remained the same as those proposed by Schneider and other authors (Betulaster=Acuminatae, Betula, and Humiles); however, the majority of birch species are confined in two extremely heterogeneous sections: Costatae (B. medwediewi, B. raddeana, B. albo-sinensis, B. austro-sinensis, B. ermanii, B. costata, B. corylifolia, B. nigra, B. alleghaniensis, B. lenta) and Chinenses (B. insignis, B. delavayi, B. potaninii, B. calcicola, B. globispica, B. schmidtii, Bchinensis).
Therefore, Schneider's statement of 1915 about the absence of a truly contemporary system for the genus Betula remained valid as late as the end of the 20th century. Here I dare to bring to my colleagues' attention a system of mine. It is an honor for me to publish it in the same journal where E. L. Regel once published his fundamental works on the systematics of the birches.
I have been studying herbarium material in the following depositories: Moscow (MW, MHA), St. Petersburg (LE), Tomsk (TK), Vladivostok (VLA), Baku (BAK), Batumi (BAT), Yerevan (ERE), Tbilisi (TBI), Cambridge (A, GH), St. Louis (MO), New York (NY), Beijing (PE), Kunming (KUN), Dehra Dun (DD), Vienna (W), Kórnik (KOR). In 1960's we received a representative selection of Japanese samples from the Makino Herbarium (MAK) by exchange. I had opportunities to observe nearly all of the birch species native on the former USSR territory (except B. maximowicziana) as well as species of the Indian Himalaya and North America in their natural habitats. Observations of cultivated trees proved to be rather useful, particularly, those made at the Main Botanic Garden (GBS) in Moscow, the Central Botanic Garden in Kiev, the Arnold Arboretum in Boston, and the Dendrogarden in Pereslavl-Zalessky.

Genus Betula L.
Type species: Betula alba L. (syn. B. pubescens Ehrh.)
The type specimen of the species B. alba was cited in the literature many times. Recently it has been officially designated. This is # 1109–1 of the Linnaeus Herbarium (LINN) (Holub 1989: 40; more about B. alba typification in: Vasilyev 1964).
I propose to distinguish four subgenera in the genus, two of which, monotypic, are being first described here.
1. Betula subg. Sinobetula A. Skvortsov subg. nov.
[Latin Description:]
Plants appear to be without resinous glands. Leaf petioles very short (5–12 mm); leaf blades large (to 120 × 80 mm); lateral veins to 18 pairs, straight, conspicuously impressed; leaves below densely covered with fulvous-brown tomentum. Staminate aments unknown until now. Pistillate aments cylindric, in clusters, developing from terminal buds; bracts three-parted nearly to their base; bract lobes oblong to linear. Fruits small, round, their wings obsolete.
Typus (designated here): Bgynoterminalis Hsu et C. J. Wang 1983 in Acta Bot.Yunnan. 5(4): 331, cum fig.
So far there is the only one species known, which has been collected only once: Yunnan, Gongshan Xian, altitude 2,600 m, Sept. 26, 1956. P. Y. Mao N 00521. KUN(!)
This amazing, truly unique birch would fully deserve a rank of a separate genus in case there were anything unusual about its aments. But they appear to be just normal birch aments, though their location is very peculiar for a birch (which is depicted in the species name). Therefore, I find it optimal to assign a rank of the subgenus. The description is based on the single specimen so far available, which is preserved at the Botanic Institute in Kunming, where I've had an opportunity to examine it.
2. Betula subg. Nipponobetula A. Skv. subg. nov.
Syn. sect. Costatae subsect. Coryfoliae C. K. Schneider 1916 in Pl. Wilson. 2: 463.
[Latin Description:]
Trees. Leaves below glaucous, without resinous glands; lateral veins straight, directed towards denticle apices. Scales of staminate aments imbricate, their apices acute, more or less deflected, pilose. Pistillate catkins cylindric to ovoid, without glands; bracts three-lobed, but with a single fruit per bract. Fruit compressed, nutlet round to broadly elliptic, its apex grades into style. Style up to 3 mm long, with two stigmas.
Typus (designated here, the only species so far known): Bcorylifolia Regel et Maximowicz 1865 in Bull. Soc. Natur. Mosc. 38(2): 417 et tab. 8 fig. 1–3.
This birch is endemic for Honshu, though it is not so rare there (I've seen about 30 herbarium specimens total.) Though it had always been treated as a distinct species, that was Schneider who made a step ahead and promoted it to the rank of the subsection. Peculiar leaf shape and venation [3] different from the rest of the birches and—still more important—the fruit structure featuring rounded flattened nutlet with a pronounced style embraced by wings—these are the characters confirming the treatment of the species as a separate evolutionary line that has never developed any further.
3. Betula subg. Asperae Nakai 1915 in Fl. sylv. Koreana 2: 25 corrected (expanded) A. Skvortsov.
Typus (designated here): B. schmidtii Regel
While Nakai assigned just a single species B. schmidtii to this subgenus, I place the following three sections here: Asperae (Nakai) Kuzeneva (which corresponds to Nakai's subgenus); Chinenses (Nakai) Chen, and Lentae Regel emend. A. Skvortsov. The name Chinenses Nakai would be also appropriate for this subgenus, but I am choosing Asperae not only due to priority considerations, but also because the subgenus contains species other than Chinese. The subgenus differs from the subg. Betula (in the meaning proposed here) in two most important characters: lack of true-birch, layered bark as well as the structure of the fruit. We are going to discuss the latter in detail now.
Fig. 1. Birch fruits (sketches somewhat generalized). 1. - Betula corylifolia Regel et Maxim. 2. - Betula medwediewii Regel 3. - Betula pendula Roth. Lower row: cross-sections of the same fruits at about the middle of their length.Traditionally birch fruits have been depicted as nutlets with wings. It is only by the wing-to-nut width ratio that they differ from one another. Meanwhile, it is the very nature of wings that varies. The difference is particularly obvious when one compares a fruit of a birch belonging to the sect. Lentae and a fruit of one from sect. Betula or Acuminatae (Fig. 1). The former has its entire ovary compressed; its wings constitute the part of the ovary that is not filled by the nutlet. The wings in the latter are special outgrowths on both sides of the ovary. In the former case, the wings proceed onto the styles; in the latter, they do not reach the styles. This is a very minute, but constant and meaningful difference. In addition to that, the wings in Lentae are nearly concolorous to the nut and not transparent; usually they are rather concave, so that one can sometimes distinguish their adaxial and abaxial side. In the subg. Betula fruits are usually flat, so that one cannot identify adaxial and abaxial sides.
Unfortunately, the authors of morpho-anatomic treatments of the birch fruit, I. A. Korchagina (1974) and G. Natho (1976) did not pay attention to the details described here above.
Subg. Asperae has other diagnostic characters, though they may be more or less pronounced and sometimes appear within the subg. Betula:
— leaves on the elongated branches are smaller and narrower than those on the spurs (including the fruiting spurs);
— pistillate aments are nearly sessile or on short and stout (ca. 2 mm thick) stalks that are frequently densely pubescent;
— there are up to 3 fully developed leaves at proximal parts of fruiting branches; however, there are no buds in the axils of these leaves; branches that produce pistillate aments then die off;
— pistillate aments do not fall apart while disseminating seed, but rather after that, sometimes persisting until the next growing season;
— lateral lobes of bracts in the subg. Asperae are mostly the same width as the central part, nearly not reclining.
The composition of the subg. Asperae Nakai:
Sect. Asperae (Nakai) Kuzeneva 1936 in Flora URSS 5: 280
Typus: Bschmidtii Regel
Bfargesii Franch.
B. potaninii Batalin
B. calcicola (W. W. Smith) P. C. Li
Bchichibuensis Hara
Sect. Chinenses (Nakai) Z. D. Chen 1994 in Acta Phytotax. Sin. 32(2): 137
Typus: B. chinensis Maxim. (syn. B. liaotungensis Baranov)
Bdelavayi Franch.
Bglobispica Shirai
Sect. Lentae Regel 1865 in Bull. Soc. Natur. Moscou 38(2): 397
Typus: B. lenta L.
B. alleghaniensis Britt.
Bgrossa Sieb. et Zucc.
B. insignis Franch. (syn.: B. kwangsiensis Metcalf, B. kweichowensis Hu, B. austro-sinensis Chun)
B. medwediewi Regel (syn.: B. megrelica D. Sosnovsky)
4. Betula subg. Betula
Typus: Balba L.
Critical characters of the subgenus: presence of "the birch bark", which is most pronounced in the species belonging to sect. Costatae and Betula; pistillate aments on slender (about 1 mm), usually glabrous leafy stalks; axils of stalk leaves usually bear buds, which prolongs the life span of the fruiting branch. Upon seed ripening, the aments tend to completely disintegrate (though occasionally bracts may persist after seed ripening); fruits bear wings: outgrowths of the lateral ovary parts; these outgrouths don't reach to the proximal parts of styles; bract lateral lobes usually differ from the central part of the bract, often reclining sidewards.
The composition of the subg. Betula:
Sect. Acuminatae Regel 1865 in Bull. Soc. Natur. Moscou 38(2): 397
Syn. sect. Betulaster (Spach) Regel 1868 in D.C. Prodr. Syst. Natur. 16(2): 179
Typus: Balnoides D. Don
B. cylindrostachya Lindl. ex Wall.
B. luminifera H. Winkl.
B. maximowicziana Regel
B. alnoides is different from the rest of the birches by its autumn flowering time and distribution within tropical/subtropical area. However, its niche there is quite similar to that of our northern birches. In the eastern Indian Himalaya, near Kalimpong, I have watched its seedlings en masse conquering a fresh landslide, while mature trees were growing nearby forming a nearly pure, one-species white-trunk grove. This section is hardly represented on the territory of Russia (by Bmaximowicziana).
Sect. Dahuricae Regel 1865 in Bull. Soc. Natur. Moscou 38(2): 396
Typus: B. davurica Pallas
Bnigra L.
I strongly believe that the American river birch B. nigra is closely related to B. davurica. Both have strikingly similar trunks with exfoliating bark and bright yellow layers underneath these rags. Early fruit ripening that is typical of B. nigra is to some extent characteristic of B. davurica, as well; however, B. davurica appears not to be restricted to banks of water bodies the way it is the case with B. nigra.
Sect. Costatae Regel 1865 in Bull. Soc. Natur. Moscou 38(2): 396
Typus: B. costata Trautv. (syn. B. nikoënsis Koidz.)
B. ermanii Cham. s. l.
B. albo-sinensis Burk.
B. utilis D. Don
B. jacquemontii Spach
This section is fully confined to temperate East Asia, its representatives dominating forest plant communities at times.
Sect. Apterocaryon Spach 1841 in Ann. Sci. Natur. 15: 195
Syn. sect. Humiles W. D. Koch 1844 in Synopsis Fl. German., ed. 2: 761
Typus: B. michauxii Spach [4]
Bfruticosa Pall.
Bovalifolia Rupr.
Bglandulosa Mich.
Bnana L. s. l.
Bpumila L.
Here also belong all kinds of local infraspecific taxa as well as hybrids, which makes the section boundary vague. At any rate, this is a young section, which probably originates either from Costatae or Betula. Many authors have tried to divide it in two, grouping races around Bnana and Bfruticosa. Of course, round small leaves of Bnana are captivating. However, on a single Bpumila plant, one may find a transition from elliptic leaves that have denticles of the fruticosa-type, to small rounded ones with roundish denticles. Bpumila is, kind of, connecting both cycles in one entity.
Sect. Betula
Syn. sect. Pterocaryon Spach 1841 in Ann. Sci. Natur. 15: 185 [5]
Typus: Balba L.
These are our white-bark boreal birches. Even though reversemutants—specimens with dark bark—are not infrequent among them, the bark always remains layered. The group is a young one, phytosociologically very active, and rather polymorphous as regards characters. The amount of species attributed to this group is astounding. However, at a glance, one can distinguish three essential groups (which may be considered as series):
Fig. 2. Probable phylogenetic  relations within the genus Betula
warty birches
Bpendula Roth
Bpopulifolia Marsh.
pubescent birches
Balba L.
Bpapyrifera Marsh.
Boccidentalis Hook.
small-leaved birches
Braddeana Trautv.
Btianschanica Rupr.
Bmicrophylla Bunge
Relationships of the subgenera and sections might be approximately depicted as in Fig. 2. The subgenus Betula is by all means the youngest. If we try to derive it from another subgenus, then that other one should be Asperae. Or else it might have had some different ancestors that haven't survived.

The author is deeply grateful to all the colleagues from the institutions named in the text. A special thank-you to my long-time assistant M. V. Kostina for typing of the manuscript.
This article was completed with financial support of RFFI (Grant # 99–04–48202).
Chen, Z. D. Phylogeny and phytogeography of the Betulaceae // Acta Phytotaxon. Sin. 1994, 32(2): 101–153.
Furlow, J. J. Betulaceae // In: Flora of North America, Vol. 3: 507–538. N. Y., 1997.
Holub, J. Subspecies names from Berher, La flore des Vosges // Folia Geobot. Phytotax. 1989, 24: 410.
Hsu, Y. and Wang, C. A new species of Betula from Yunnan // Acta. Botan. Yunnan. 1983, 5(4): 381–382.
Koch, W. D. Synopsis florae Germanicae et Helveticae, Vol. 2. 1844.
Korchagina, I. A. [On the nature of the flower in Betulaceae] // Tr. MOIP 1974, 51: 50–74. In Russian.
Li, P. and Skvortsov, A. K. Betulaceae // In: Flora of China, Vol. 4: 286–313. Beijing, St. Louis, 1999.
Nakai, T. Flora sylvatica Koreana. Pars 2. Betulaceae. Government of Chosen. [Seoul], 1915.
Natho, G. Zur Fruchtmorphologie und Gliederung der Gattung Betula L. // Gleditschia 1976, 4: 9–21.
Regel, E. Bemerkungen über die Gattungen Betula und Alnus nebst Beschreibung einiger neuer Arten // Bull. Soc. Natur. Moscou 1865, 38(2): 388–434.
Regel, E. Betulaceae // In: De Candolle, Prodromus systematis naturalis regni veget., Vol. 16(2): 161–189. Parisiis, 1868.
Schneider, C. K. Bemerkungen zur Systematik der Gattung Betula // Österr. Bot. Zeitschr. 1915, 65(10–12): 305–312.
Schneider, C. K. Betulaceae // In: C. S. Sargent (ed.) Plantae Wilsonianae, Vol. 2: 423–508. Cambridge, 1916.
Spach, E. Revisio Betulacearum // Ann. Sci. Natur. 1841, Ser. 2(15): 182–212.
Vasilyev, V. N. [Betula pubescens Ehrh. and Bverrucosa Ehrh.] // Bot. zhurn. 1964, 49(12): 1784–1789. In Russian.
Vasilyev, V. N. [The birches of the Urals] // Tr. In-ta ekologii rast. i zhiv. 1969, 69: 59–140. In Russian.

[1]  E. Regel was not consistently using the term 'section': he mostly employed the German equivalent 'Abteilung'. However, from certain contexts (pp. 397, 405), one can conclude that what he meant by 'Abteilung' actually was 'section'. Hence we are to treat the divisions of the genus that Regel proposed in 1865 in the rank of sections. In his work of 1868, Regel preferred to label each of his divisions with a section sign ('§') instead of explicitly nominating their rank.
[2]  While listing species attributed to certain sections by different authors, we deliberately omit hybrids, vague species and also authors' names in order to make the lists more compact. Some species names have been replaced by currently accepted synonyms.
[3]  Regel and Maximowicz had the intention to depict the peculiarity of Bcorylifolia leaves in its name, but did not succeed in doing so: the leaves of this birch have hardly anything in common with the leaves of Corylus.
[4]  This was only Bmichauxii Spach, a peculiar endemic of the eastern Canadian Arctic, that E. Spach (1841) attributed to the sect. Apterocaryon. This species with its nearly or even completely wingless nutlets is nowadays usually considered to be the product of Bnana morphological reduction. The same kind of reduction has been observed in the Eurasian Arctic (in the birch described as Btundrarum Perf.). Even though this wing reduction is characteristic of only one species (or even a singleseries, according to Vasilyev, 1969: 69), whereas the rest of the section components retain nutlet wings, we still have to keep the name Apterocaryon for priority considerations.
[5]  The section Pterocaryon, as proposed by Spach (1841), embraced nearly all of the birches known by then (except Bmichauxii). There were Bnana, Bfruticosa, Bermanii, and Blenta. We can, accordingly, add this name as a synonym when naming other sections: 'Pterocaryon Spach pro parte'. However, the centerpiece of Spach's sect. Pterocaryon was certainly Balba L., so that retaining this name for any section other than Betula would be against any logics.

Translation I. Kadis
19 Dec 2004

Salicicola Last updated: