Flora of the Canadian Arctic Archipelago
Rosaceae, Rose family.
Published in Sp. Pl. 495. 1753.
Vegetative morphology. Plants 1–25(–35) cm high; perennial herbs; caespitose. Taproot present. Caudex present. Aerial stems branching from a tap at or near ground level into two or more branches; erect. Leaves mainly basal; alternate; dying annually and non-persistent, or marcescent. Stipules present; 4–20 mm long; 1–6 mm wide; green, or pink or reddish; hairy; pilose, or villous, or long-silky; glandular, or without glands; apex acuminate, or acute, or obtuse. Petioles (2–)5–60(–100) mm long; with sessile glands, or without sessile glands; hairy; pubescent, or pilose, or villous, or woolly, or long-silky. Petiole hairs shorter than the diameter of the petiole, or longer than the diameter of the petiole; appressed, or spreading, or erect; straight, or floccose, or wavy, or crispate; smooth, or rough. Leaf blades compound. Blades 5–25(–50) mm long, 5–30(–50) mm wide, veins palmate. Blade adaxial surface with sessile glands or without sessile glands, glabrous or glabrescent or hairy, hairs pilose or villous or long-silky, hairs simple, hairs sparse or moderately dense or dense, hairs white, or translucent. Blade abaxial surface hairy, hairs pilose or tomentose or woolly or long-silky, hairs sparse or moderately dense or very dense, hairs white or a mixture of white and yellow, hairs straight or wavy, hairs appressed or spreading. Blade margins entire or crenate or dentate or deeply divided, with non-glandular hairs, with 1–10 teeth on each side of the blade, with teeth all around the blade or toward the apex; degree of incision 5–90%; apices obtuse, or rounded, or retuse. Leaflet arrangement palmate, or pinnate, or digitate. Leaflets 3–7; (3–)6–20(–35) mm long; (2–)5–15(–25) mm wide; elliptic, or ovate, or obovate, or obtriangular; veins conspicuous. Apical leaflet base distinctly stipitate, or not distinctly stipitate; stipe 0–7 mm long.
Reproductive morphology. Plants bisexual, or agamospermic. Flowering stems shorter than the leaves, or about as high as the leaves, or conspicuously taller than the leaves; with leaves, or without leaves in the upper half. Flowering stems hairy. Flowering stems pubescent, or pilose, or villous, or tomentose, or woolly, or long-silky. Flowering stem hairs simple; shorter than the diameter of the flowering stem, or longer than the diameter of the flowering stem; white or translucent. Flowers solitary, or in inflorescences. Inflorescences cymose; lateral; dense, or diffuse. Pedicels present, or absent. Flowers per inflorescence 1–5(–15); small, or medium-sized, or large. Sepals conventional. Epicalyx present. Epicalyx segments 2–9 mm long. Epicalyx segments (0.4–)1–3(–4) mm wide. Epicalyx segments shorter than the calyx segments, or equal in length to the calyx segments. Epicalyx segments narrower than the calyx segments, or equal in width to the calyx segments. Sepals 5; free; (0.8–)2–3.5 mm long; (2–)3–8(–11) mm wide; green, or brown. Calyx with sessile glands; hairy. Calyx hairs pilose, or villous, or long-silky; white or translucent. Calyx margins ciliate, or margins without cilia. Petals conventional; free; shorter than the calyx, or longer than the calyx; 5; yellow; with contrasting markings, or without contrasting markings; elliptic, or ovate, or obovate; unlobed, or slightly lobed or undulating; 2–10(–12) mm long; (3–)4–8(–9) mm wide. Stamens 15–30; stamen filaments glabrous. Anthers yellow; ellipsoid, or triangular; 0.2–0.8 mm long. Ovary superior; carpels 25–45; apocarpous. Styles 0.5–1.8 mm long; conical, or straight; basal portion smooth, or covered with short papillae, less than 0.1 mm high, or covered with long papillae, 0.1 mm high or higher. Ovules per ovary 1. Fruit sessile; with calyx persisting; dry; an aggregate of achenes; ovoid; green at maturity, or straw-coloured; (0.5–)0.8–1.8 mm long; (0.2–)0.5–1(–1.2) mm wide; surface venation reticulate, or appearing veinless, or venation ribbed; indehiscent.
Chromosome information. 2n = 14–77.
General notes. Potentilla species known from the Arctic islands fall into three groups that may be recognised as either genera or subgenera. Two of these are taxonomically uncomplicated; genus Comarum L. (subgenus Comarum (L.) Syme with Comarum palustre) and Argentina Lam. (subgenus Chenopotentilla (Focke) Juz. with Argentina egedii). These are not very closely related to the third group (or to each other). Their species are most probably sexual and do not, as far as known, form hybrids with species of Potentilla s.s. The third group Potentilla L. s.s. has three sections in the Arctic: sect. Aureae (Th. Wolf) Juz. with P. crantzii and P. hyparctica, sect. Multifidae (Rydberg) Juz. with P. pulchella, and sect. Niveae (Rydberg) Juz. with at least three species of the P. nivea complex and two of the P. uniflora complex. All these seem to be facultative to nearly obligate agamosperms, even if the evidence is scanty for some of them (Asker and Jerling 1992, Eriksen 1996). Pseudogamy seems to be the rule, and one of the most common outward signs of agamospermy, nearly perfect fruit/seed-set, is absent in the agamospermous species. Pseudogamy requires pollination for endosperm development, but the pollen does not fertilise the ovule. Pseudogamy may occur with pollen from the same species or from a related species, which is often more efficient (Nyléhn, personal communication, 2000). Predominance of agamospermy has recently been shown in the three Svalbard taxa of the P. nivea complex: P. nivea, P. arenosa subsp. chamissonis, and the local P. insularis Soják, which is possibly conspecific with P. rubricaulis s.l. The variation pattern found in P. pulchella (see Hansen et al. 2000) and in other parts of sect. Multifidae also indicates at least some degree of agamospermy.
Both species concept and species delimitation are problematic in agamospermous plant groups. Soják (1985, 1986, 1989) interpreted the complicated morphological variation found in these sections of Potentilla to be the result of extensive hybridisation between 'primary' species both within and between sections. Hybrids could then propagate, form populations, and even attain ranges, by agamospermy. Entities, assumed to be such stabilised agamospermic hybrids, have been treated as species by Soják (1985, 1986, 1989). This approach was also adapted by Yurtsev (1984, in draft for the Panarctic Flora checklist, 1999, Elven et al. 2003) for the arctic flora and has resulted in about 60 species being named from the arctic or near arctic areas. More than half of these species are considered to have resulted from hybridisation. One of the basic assumptions in the approach of Soják and Yurtsev is that digitate and subpinnate leaves in these groups are a result of hybridisation between 'primary' species with ternate (i.e., sect. Niveae and parts of Aureae) and truly pinnate leaves (sect. Multifidae).
Four hybrid species have been reported by Soják and Yurtsev from the arctic areas surrounding the North Atlantic. In three of these hybrid species, Soják's hybrid origin hypotheses have been shown to be improbable:
1. Potentilla scandica Soják was interpreted by Soják as a hybrid P. crantzii × P. gelida C.A. Meyer. It represents rather normal variation in leaflet number in P. crantzii (Nyléhn, personal communication, 2000).
2. Potentilla insularis was interpreted by Soják as a hybrid P. arenosa subsp. chamissonis × P. lyngei. It is only found in areas where P. lyngei (from sect. Multifidae) is absent and does not include genetic markers from P. pulchella, which is the Multifidae species present in the area (Svalbard, see Hansen et al. 2000, Hamre 2000). It might be better interpreted as a locally deviating part of a more widely delimited species or species group, which also includes the western hemisphere arctic expression of P. rubricaulis s.l.
3. Potentilla protea Soják, interpreted as P. crantzii × P. hyparctica. It seems to be single, first-generation hybrid plants, from the material investigated from northwestern Russia, Svalbard, Fran Mayen Greenland, and Canada (Elven, observation, 2000).
A few specimens in material investigated from the Arctic Archipelago (CAN, DAO, O) seem to represent several hybrid combinations:
P. arenosa coll. ×hyparctica (1–2 sites on Baffin, one site on each of Ellesmere, Victoria, and Banks islands),
P. arenosa subsp. chamissonis ×nivea (two sites on Baffin),
P. arenosa coll. ×pulchella (one site on each of Devon, Ellesmere, and Banks),
P. arenosa s.s. ×rubricaulis s.l. (two sites on Ellesmere, one on Banks),
P. hyparctica ×nivea (one site on Baffin),
P. hyparctica ×pulchella (one site on each of Baffin, Devon, and Melville),
P. hyparctica ×rubricaulis s.l. (possibly one site on Baffin),
P. nivea ×rubricaulis s.l. (one site on Baffin),
P. pulchella ×rubricaulis s.l. (two sites on Banks),
P. vahliana and another parent (two specimens from Baffin, one from each of Axel Heiberg and Melville).
Some of these hybrid products may form local populations, as there are several individual plants collected from some of the sites, but none of them seem to be more than local phenomena that do not deserve special rank.
The complicated taxonomy and nomenclature of Soják and Yurtsev is based on a specific hybridisation model. This model has proved insufficient for explaining the investigated North Atlantic supposedly 'hybrid' taxa. It remains unproved that any of the generally accepted and widespread taxa in the Arctic islands have a hybrid origin, and the naming of single hybrids, hybrid populations, or even population groups should be dissuaded.
Hairs and glands in Potentilla. The kinds of hairs and their location have been, and still are, decisive for species definition and delimitation in many parts of the genus. This is specially the case within the complicated sect. Niveae and related groups. The following categories are found and used in the descriptions where relevant. Some further discussion of hairs and their importance is found in Eriksen and Yurtsev (1999), and the concepts and descriptions below are based on their proposals.
Floccose hairs are flat, irregularly twisted or felted and more or less appressed to the surface. The hairs are usually so long and intertwined that it is impossible to tell where a single hair begins or ends. They often form a dense tomentum on the lower leaf surface and petiole but may also occur scattered on the peduncle and inflorescences axes. Among our species, such hairs characterise P. nivea and its hybrids. They are easily discernible with a strong lens or a dissection microscope.
Crispate hairs are unicellular and less than 1 mm long. The more or less individual hairs are terete and wavy often in a corkscrew-like fashion. They may form a tomentum on the lower leaf surface or the petiole, and sometimes they make the peduncle appear villous. Among our species, such hairs characterise P. pulchella, P. rubricaulis s.l., P. uniflora, and P. vahliana.
Straight or nearly straight hairs appear in several shapes and may be smooth or verrucose ('rough'). To decide unambiguously between smooth and verrucose one needs a very strong magnification (SEM, Scanning Electron Microscopy), but it is possible to get an impression, with a little training, using a strong dissecting microscope. The surface of smooth hairs appears even and often shiny, that of verrucose hairs slightly irregular and a little dull, owing to impeded translucence.
Straight, smooth, long, and sub-appressed to spreading hairs form a silky indumentum ('long-silky') that characterises petioles, leaf surfaces, and partly peduncles in P. uniflora, P. vahliana, and P. pulchella. Similar hairs, which are verrucose, characterise P. rubricaulis s.l. This type of indumentum appears soft, as opposed to the next.
Strongly spreading to patent, long, straight hairs characterise P. crantzii, P. hyparctica, and P. arenosa s.l. Those of P. crantzii are smooth, those of P. hyparctica are smooth or verrucose, and those of P. arenosa s.l. are verrucose and especially stiff ('bristly').
Glands occur in most of the species, both as sessile glands on most plant parts and as glandular hairs on petioles, peduncles, and leaves. Potentilla crantzii and P. hyparctica are heavily glandular, the others less so. Thus, the number and distribution of glands seem to be potentially of less taxonomic importance than the hairs (see Eriksen and Yurtsev 1999).
This publication is available on the internet (posted May 2011) and on CD-ROM (published in 2007). These versions are identical in content, except that the errata page for CD-ROM is accessible on the main index page of the web version.
Recommended citation for the web-based version of this publication: Aiken, S.G., Dallwitz, M.J., Consaul, L.L., McJannet, C.L., Boles, R.L., Argus, G.W., Gillett, J.M., Scott, P.J., Elven, R., LeBlanc, M.C., Gillespie, L.J., Brysting, A.K., Solstad, H., and Harris, J.G. 2007. Flora of the Canadian Arctic Archipelago: Descriptions, Illustrations, Identification, and Information Retrieval. NRC Research Press, National Research Council of Canada, Ottawa. http://nature.ca/aaflora/data, accessed on DATE.
Recommended citation for the CD-ROM version of this publication: Aiken, S.G., Dallwitz, M.J., Consaul, L.L., McJannet, C.L., Boles, R.L., Argus, G.W., Gillett, J.M., Scott, P.J., Elven, R., LeBlanc, M.C., Gillespie, L.J., Brysting, A.K., Solstad, H., and Harris, J.G. 2007. Flora of the Canadian Arctic Archipelago: Descriptions, Illustrations, Identification, and Information Retrieval. [CD-ROM] NRC Research Press, National Research Council of Canada, Ottawa..