THE CONTINENTAL FLORA OF SOUTH SWEDEN 307 



torum,) Crepis praemorsa, Phleum Boehmeri (Plate 5), Potentilla arenaria, leuco- 

 politana (with several newly discovered occurrences) and rupestris (p. 326), Tri- 

 folium montanum (p. 301), Veronica spicata (p. 310), and Viola rupestris (p. 319). 

 Add to these the continental wooded-hillside species: Dracocephalum Ruyschiana 

 and Pulmonaria angustifolia; and the following species, not continental, yet re- 

 markable with regard to their distribution: Hyjiericum montanum (p. 346), Lathyrus 

 heterophyllus (Plate 4), Oxytropis campestris, Potentilla Tabernaemontani, Scle- 

 ranthus perennis, Thesium alpinum (Plate 4), and Thymus serpyllum. 



In this connection must be mentioned the importance of the oses in eastern 

 Smaland for the migratory history of many xerothermous and heliophilous spe- 

 cies in South Sweden. The immigration and first dis])ersa] of such species 

 should to a great extent be assigned to a period with a dry and warm climate 

 (the Boreal and Sub-boreal period). In this case the southerly exposed slopes 

 might have formed suitable immigration routes to a much greater extent than 

 would be the case in a natural vegetation with the present climate. Thanks to 

 them the species have been able to make their way into or through the South 

 Swedish highland, which must, as to the rest, have been like a desert to them. 

 And by this route they may perhaps have reached other parts of South Sweden. 

 The Hogsby os should be especially noted. I should like to put forward the ques- 

 tion whether this os may not have formed a link across South Sweden from 

 Oland-Gotland and south-eastern Smaland to Lake Vattern and Falbygden in 

 Vastergotland. The very peculiar distribution-type of a few species points to 

 the existence of such a link and to the determination by it of the distribution 

 of the species. The species are distributed in the eastern part of the province 

 of Jonkoping, especially in and about the Ema valley and in parts of Vastergot- 

 land, especially Falbygden. In the country to the south-east of the southern end 

 of Lake Vattern there may be occurrences that to some extent connect these 

 centres of distribution. Such is the distribution of Dracocephalum Ruyschiana 

 (Plate 4) and Potentilla rupestris (p. 326) and, though not so decidedly, Pul- 

 monaria angustifolia (Plate 4). Even the easterly Central European Lathyrus 

 heterophyllus (Plate 4) and Thesium alpinum (Plate 4) may be ranged with 

 this group. The absence or scanty distribution of these species in Ostergotland 

 is especially remarkable. As to the direction of the migration we can hardK- 

 draw any justifiable conclusions from the present distribution. The two first- 

 named at least may be supposed to have passed from Vastergotland to Sma- 

 land. The wide distribution of Dracocephalum in South-Eastern Norway and 

 the species being lacking on Oland and Gotland may justify such a supposition 

 about this species. 



In western Smaland the glacio-fluvial deposits are not, as a rule, developed 

 into ridges. They are placed in old erosion valleys, parth" filling them. But if 



