242 RIKA RD STERNER 



many things that indicate that se\eral species belonging to this group are still continu- 

 ing to spread — e. g. Senecio palustris, Petasites spurius and Scolochloa festucacea. 



Of the contijienially Boreal species a great number have a very extensive 

 distribution in the north of luirope and in the greater part of Middle I'-urope. 

 The often abundant distribution of these species in Central Europe is caused by 

 the topography: they are found in the mountain woodlands of hilly districts. 

 They reach their westerly limits in eastern or central France and in south-eastern 

 Belgium. It should be observed that Boreal species that occur in Great Britain 

 also have in these districts temporary western boundaries. Engler (1879) has 

 sought to explain this circumstance by the theory that the first-named species 

 immigrated from the east and did not get as far as Great Britain before the 

 English Channel broke through the land-connection which is supposed to have 

 existed in early post-glacial times. 



The Boreal species that scarcely extend beyond East Europe form a group 

 that is comparatively few in numbers, but is interesting from several points of view. 



In the first place should be noticed the Sarmatian psamviopliilons species. The 

 majority of these belong to critical genera which have other species very much 

 akin to the Sarmatian ones distributed in adjacent district such as Koeleria 

 grandis (Bess.) Domin, several Dianthus-species, some species belonging to the 

 I'otentilla collina-group, Pulsatilla j^ratensis L. {Hayek's sense), probably several 

 P'estuca ovina-species. Perhaps in the Sarmatian herbaceous sand-grass heaths 

 or pine-forest heaths (on this point see also Chapt. viii) we have special psammo- 

 philous types of vegetation, which are not yet fully differentiated as regards the 

 species which are typical of them. This might also be connected with the fact 

 that the Sarmatian region is not sharply divided from surrounding tracts in its 

 climatological and orographical character. 



As regards the distribution of other Sarmatian or Subarctically Sarmatian species 

 the hypothesis may be put forward, to some extent in accordance with Engler 

 (1. c.) that these species, unlike the species that are also distributed through 

 Central Europe, were able, at the close of the Ice Age, to spread only from one 

 or two refuges in the cast of Europe, but not from any refuge tvest of the Alps. 

 Here then would be found an explanation why the species are lacking in Central 

 and Western Europe. 



The two Hast Eziropean refuges would be the Carpathians (possibly the Eastern 

 Alps) and the Urals. 



Vierhapper (191 1) has devoted a careful study of the distribution and mode 

 of occurrence of certain species falling under this group. 



Several remarks of interest about the distribution of continentally — Boreal spe- 

 cies are to be found in a recently published work of Wangcrin, treating the 

 appearance of a »montan» flora in East and West Prussia (Wangerin 1920). 



