H. N. Barber.

Evolution in the Genus Paeonia

Nature vol. 148 No. 3747. August 23, 1941. Pp. 227-228



The genus Paeonia has three main centres of distribution and diversity : in the Mediterranean and Black Sea basins, in the Far East from Tibet to Japan, and in North America. From my own and previous chromosome studies (1, 2, 3), I find that in these separate regions it has undergone three different methods of species formation, all the time working with the same haploid set of five chromosomes.

In the Mediterranean area are a group of small-range diploids (2n = 10). Most of these diploids, apparently by simple doubling, have given rise to large-range tetraploids (2n = 20) usually lying to the north of their progenitors. Three of these large-range tetraploids, namely, arietina and peregrina in the Balkans and Asia Minor, and coriacea in southern Spain and Morocco, have no nearly related diploids ; apparently their ancestors have failed to survive (see accompanying table).



Species in Paeonia

 

2x

4x

Europe

 

 

 

Mlokosewitschii (Caucasus)

Witmanniana and varieties (Caucasus

 

daurica (Crimea etc.)

mascula (scattered)

 

Clusii (Crete)

officinalis, humilis and (N. Mediterranean) vars.

 

Cambessedesii (Baleares)

Russii and vars. (Western islands of Mediterranean)

 

2 other localized species in Mediterranean; 1 widespread in Ukraine

3 other widespread species unrelated to any surviving diploids

Asia

 

 

 

japonica(Japan)

obovata and varieties (E. Asia)

 

8 other widespread diploids

No other tetraploids

N. America

 

 

 

2 diploids

No tetraploids



It thus seems that the advancing ice had driven the diploid species into Mediterranean peninsulas and islands, and from these isolated fragments of pre-glacial species Europe was afterwards recolonized by more vigorous, or perhaps more adaptable, tetraploids as the ice retreated. The kind of adaptability is shown in some cases to consist in stronger tuber development.

In Asia, on the other hand, there has been no such impassable barrier to movement back and forth, and we find a number of diploid species—some covering a large range, for example, P. lactiflora and P. anomala—and only one tetraploid, P. obovata. This species again is in the north (Manchuria and eastern Siberia), but it seems that the diploids, never having been restricted in population and variation, have themselves been able to meet the opportunity of colonization and to follow the retreating ice northwards.

In the smaller territory of California and Oregon a smaller population shows yet another system of variation. Only two species are recognized, and both of them are true-breeding diploid hybrids of the kind known in Oenothera. Instead of having five bivalents at meiosis they have rings of six, eight or ten chromosomes (3).

Thus in the three regions Nature seems to have carried out an experiment in variation and selection which helps us to understand how different conditions, partly external and partly perhaps internal, can lead to different methods of evolution in members of one genus.

I wish to thank Major F. C. Stern for much of the material on which this investigation was made. All the details of the geographical distribution will be found in his forthcoming monograph (4).

H. N. Barber.

John Innes Horticultural Institution,

Merton Park,

London, S.W.I 9. July 23.

(1) Dark, S. 0. S., J. Genet., 32, 353 (1936).

(2) Sinoto, Y., Cytologia, 9, 254 (1938).

(3) Stebbins, J. L., and Ellerton, S., J. Genet., 38, 134 (1939).

(4) Stern, F. C., "The Genus Paeonia", Roy. Hort. Soc. Monographs (in the press).