Panarctic Flora

Browse

3701 Papaver L.

GBIF

Notes: . Notes are based on discussions among Elven, Murray, Petrovsky, and Solstad and on the works of Solstad et al. (2003), Solstad (2009), and Solstad (unpubl., ITS and plastid sequences and low-copy genes).

All species of Papaver that reach the Arctic belong to sect. Meconella Spach, characterized by being perennial with all leaves in basal rosettes and single, one-flowered scapes from the rosettes. Section Meconella is a monophyletic group (Carolan et al. 2006) with more than 100 published species names probably representing 50-60 species of which 30-40 reach the Arctic (Solstad 2009). Molecular evidence (Carolan et al. 2006) suggests that the section is misplaced in Papaver. Its sister-group is the Asian part of the genus Meconopsis Vig., i.e., excluding its western European type species, M. cambrica Vig., which is a true Papaver. This affinity is also supported by Solstad (unpubl.). The nomenclatural consequences have not yet been addressed. Recognition of sect. Meconella at rank of genus may be unavoidable, but then under another name as the section name is occupied at generic level by Meconella Nutt. (Papaveraceae) with three western and southwestern North American species (Hannan 1997).

There are numerous studies of sect. Meconella that include the arctic plants, foremost among them Knaben (1958, 1959a, 1959b, 1979 - North Atlantic regions and partly North America, early experimental); Rändel (1974 - worldwide, monographic; 1975, 1977a, 1977b - regional studies); Tolmachev (1975a - arctic Russia, monographic); Bezdeleva (1987 - Russian Far East, monographic); Kadereit (1993 - Europe, monographic); Peschkova (1994a - Siberia, monographic); Murray (1995 - North America, nomenclature); Kiger and Murray (1997 - North America, monographic); Petrovsky (1999 - arctic Russia, summary); Nilsson (2001b - northwestern Europe, monographic); Selin and Prentice (1988) and Selin (2000 - northwestern Europe, morphometrics); Solstad (1998) and Solstad et al. (1999, 2003 - North Atlantic regions, molecular), Solstad (2007) and Solstad and Elven (in prep., northern Canada, morphology), Solstad (2009 - global, molecular and morphology); and Schönswetter et al. (2009 - central Europe, molecular and morphology). A valuable summary of the cytotaxonomy in northeastern Asian Papaver is Zhukova and Petrovsky (1985a).

In spite of all these studies, parts of Papaver sect. Meconella is still messy, especially in the Asian and Beringian regions. Previous treatments have often been contradictory for many of the northern entities. We suspect that the problems are due to a Pleistocene or perhaps even later evolution of the majority of the northern species, by hybridization and polyploidization (i.e., reticulation) from diploid and low-ploid temperate species, some extant and some probably extinct or not yet discovered. The majority of the high polyploids seem to be stabilized, but not all. Support for this hypothesis is found in the wide span of ploidy levels (2x to 12x or more, 2n = 14-105). Also within morphologically defined species, it is the rule rather than the exception to find two or more ploidy levels. Autopolyploidy (in addition to allopolyploidy) is indicated (Solstad 2009).

Section Meconella has a nearly continuous distribution throughout the arctic regions with outliers in the Cordilleras south to southwestern U.S.A. (P. uintaense and P. pygmaeum, see D. Löve and Freedman 1956; D. Löve 1969; Welsh et al. 2003a), in coastal eastern Asia south to Japan (P. miyabeanum), in Siberia and central Asia at least south to Mongolia, China, and Pamir (the P. croceum, P. nudicaule, and P. rubro-aurantiacum aggregates and possibly more, see Peschkova 1994a), even to the northwestern Himalayas (P. himalayanum), and in the southwestern, central and southeastern European mountains (the P. alpinum aggregate, see Schönswetter et al. 2009). The majority of the southern plants are alpine, but temperate lowland steppe, steppe-forest, and river shore species are found in southern Siberia and central and eastern Asia. Diploid species are known from the temperate mountains and steppes (the P. alpinum aggregate, P. croceum or perhaps its aggregate, parts of the P. nudicaule and P. rubro-aurantiacum aggregates, P. anomalum, P. stanovense, P. pygmaeum, and perhaps some of the several species never investigated for ploidy level), but also in northern regions (the P. microcarpum aggregate, P. walpolei). The majority of the assumed and known diploid species we have investigated also have tetraploids in lower frequency. More regularly tetraploid and hexaploid species occur both in temperate mountains and steppes and in the north, whereas the known high polyploids (8x and above) all are northern. There is at present no strong support for any specific hypothesis about the parental partaking of the diploids in the evolution of the major arctic high polyploid species, although several affinities are suggested by AFLP markers and morphology (Solstad 2009).

The reproductive system varies from predominant outcrossing (e.g., in P. croceum, P. nudicaule, and in parts of P. alpinum s. lat.) to predominant inbreeding (e.g., in P. dahlianum and P. radicatum). The plants have no means of vegetative propagation but instead show high seed production and rapid turnover in often small and fluctuating populations in unstable situations (Solstad 1998). In the Arctic two or more species are often found in mixed populations or close together. This could promote rapid evolution by genetic drift and comparatively frequent hybridization accompanied by polyploidization. Knaben (1959a, 1959b) demonstated a gradual decrease in pollen fertility and seed set with geographic and 'taxonomic' distance in artificial crosses within and between species and subspecies in the North Atlantic and North American regions, but rarely full sterility, in spite of ploidy differences. This suggests that some of the major northern species probably share major parts of their genomes. It should, however, be noted that there is little evidence for hybridization in the field.

Few of the species of Papaver that reach the Arctic have been unambiguous as to circumscription and none of the geographically wide-ranging high polyploids. (a) The Russian, North American, and northwestern European treatments have in some cases recognized taxa in different ways. For instance, workers concerned with the variation around the North Atlantic have emphasized, more than others, ploidy differences as limits between species. This is due to few entities and only two or three high polyploid levels (8x, 10x, 12x) documented for northwestern Europe, Greenland, and northeastern North America. Such a correspondence is not equally evident in the high diversity regions of northeastern Asia and northwestern North America. The occurrence of two or more ploidy levels within a morphologically defined species is quite common (Solstad 2009). (b) The meaning of some names of widely distributed species - P. dahlianum-polare, P. lapponicum, and P. radicatum - have until recently been ambiguously defined and partly contradictory among workers due to lack of proper typification. The circumscriptions of these species are still disputed. (c) Experimental and molecular studies have until recently been undertaken only in the low diversity North Atlantic regions. (d) The only global monograph of the group - Rändel (1974, see also 1975, 1977a, 1977b) - is now 40-45 years old and based on few and plastic characters without experimental work. Rändel's (1974) conclusions are not supported by subsequent studies in the North Atlantic regions. Her conclusions should therefore be viewed critically. The most recent monographic treatments for Europe (Kadereit 1993) and North America (Kiger and Murray 1997) were largely based on Rändel's works. (e) No efficient comparison has been made between the variation and names in the two high diversity regions of northeastern Asia and northwestern North America. And (f) molecular data suggest that there are several yet undescribed entities, making many of the currently named species heterogeneous and impossible to circumscribe meaningfully.

Petrovsky (1999) surveyed the species of the genus in arctic Russia, also with a view towards non-Russian regions. He grouped the species into seven aggregates and some odd species. During the period since 1999, we have studied the Papaver collections in Ottawa (CAN, DAO), Fairbanks (ALA), Tomsk (TK), Novosibirsk (NS, NSK), Moscow (MHA, MW), St. Petersburg (LE), Stockholm (S), and Oslo (O). We have also undertaken field work in northern Europe, northern and northeastern Asia, Alaska, Canada, and Greenland and performed molecular studies and ploidy level determinations. We have thereby reached many tentative conclusions that modify Petrovsky's framework, especially his concept of what he described as the aggregates centered on P. dahlianum, P. lapponicum, and P. radicatum. He assigned many Beringian species to these aggregates, but Solstad (2009) found no molecular support for these species or any very close relatives being present in Beringia (except for one taxon related to P. dahlianum). This means that 11 of the 27 species Petrovsky accepted within these three aggregates probably belong in other relationships. There is, however, not sufficient experimental support for a critical evaluation of all the species accepted by Petrovsky. Sequences studied until now (ITS, cpDNA; Solstad unpubl.) have not varied enough or consistently enough to support a phylogeny resolving more than a very small part of sect. Meconella. This must be the result of the assumed recent diversification of this group. Works on AFLP markers (Solstad 2009) and on low copy genes (Solstad unpubl.) are more promising. Variation in AFLP markers makes possible a rough division into major branches and numerous clusters that largely correspond with current taxonomy for the more distinctive species and that subdivide the less distinctive species in a morphologically and geographically meaningful way. Results from these AFLP studies combined with a morphological analysis (Solstad 2009) are used for most of the solutions proposed below.

Higher Taxa

Lower Taxa (Show all)