340 RIKARD STERNER 



the layer of soil on the limestone pavement is not sufficiently thick and moist 

 to permit closed forests to arise. On Oland they consist of low foliferous forests, 

 mainly composed of hazel and with sporadic low trees, such as oak, ash and 

 Sorbus suecica. Between the hazel thickets the ground is covered by meadows 

 rich in herbs, greatly reminding one of certain Alvar meadows. On Gotland the 

 pine is the dominant tree (cf. Hesselman 1908, pp. 83 ff.). These types of vege- 

 tation would seem to show great resemblances in composition to those in 

 corresponding localities in Central Europe. Special mention may be made of 

 their wealth of orchids. 



On hill slopes with a southerly exposure and with a nutritive soil there are 

 found throughout South Sweden forest types which we can call »oak slopes» or 

 »grove slopes» (Swed. y>ckbackar-» and -^lundbackary^). They are most fully deve- 

 loped, and occur in the greatest abundance, in the south-east of Sweden, espe- 

 cially in calcareous districts. In these forest types we have something that corre- 

 sponds exactly to the thinner oak forests of Central Europe. Many of the species 

 distinctive of South Sweden are distributed throughout the oak-forest zone of 

 Middle Europe: e. g., Agrimonia eupatoria. Clinopodium vulgare, Geranium san- 

 guineum, Lathyrus niger, Melampyrum cristatum, Origanum vulgare, Serratula 

 tinctoria. Here too we tind many species which have a continental distribution. 



The flora of these forest slopes have many representatives in the conglomeration 

 of species from wholly different plant communities, which forms the flora on 

 tree-clad rocky slopes with a southerly exposure. Here they meet with represen- 

 tatives of clifi" vegetation proper, of the flora of herbaceous slopes, the flora of 

 coniferous forests, ploughland weeds etc. As so far scarcely any analyses of this 

 vegetation have been published with regard to South Sweden, I have considered 

 it proper to give some such analyses from diff"erent districts in the south-east of 

 Sweden (See table 4, Appendix II; cf. Erdtman 1922). 



These lists of species may be of interest from several points of view. In this 

 place I shall only call attention to a peculiarity with regard to the dispersal 

 biology of the species, which is of some consequence for its bearing on an 

 argument in the sequel. The flora includes what is assuredly an unexpectedly 

 small number of species which are especially equipped for wind dispersal. It 

 should further be observed that the species which have such equipment occur 

 abundantly in other types of vegetation in the surrounding district, in coniferous 

 forests or on ploughlands. An exception to this, however, is provided by 

 Cynanchum vincetoxicum. This is all the more peculiar because the equipment 

 of the species for wind dispersal ought to be extremely effective — namely a 

 winged, flattened, light seed provided with a powerful pappus. As will be shown 

 below, this circumstance may be connected with the peculiar distribution of the 

 species in South Sweden. 



