40 



PLANT SOCIOLOGY 



With increasing size of the test areas the highest-frequency classes 

 enlarge, while the lowest decrease. Kylin (1926, p. 148) and Gleason 

 (1929) have shown theoretically this displacement of classes. Figure 

 19 gives a practical example from the extremely homogeneous 

 Rhynchosporetum albae. 



re- 



Frequency classes 



76sq.m. 



Fig. 19. — /. Frequency diagrams of a homogeneous Rhynchospora alba association 

 at Zugerberg. A, from sample plots of 1 sq. m.; B, from sample plots of 4 sq. m. (10 

 samples) ; II. Species-number: area curve of the association. 



Frequency diagrams are comparable, therefore, only when they are 

 made from test areas of equal size. 



With this provision, diagrams with relatively large numbers of 

 species in the highest-frequency classes and fewer species in the lower 

 classes indicate patches of vegetation which are floristically homo- 

 geneous. Those with smaller number of species in the highest classes 



O-ZO 20-10 tO-60 60-80 80-100% 



I I m rr v " ^ 



0-20 20-HO HO-eO eo-80 80-100% 



Fig. 20. — A, Frequency diagram from a Car ex rostrata-Sphagnum lindbergii asso- 

 ciation in Norway; 20 samples of 1 sq. m. {After Nordhagen.) B, The same from a 

 Scirpus-Phragmites association near Montpellier, France; 10 samples of 1 sq. m. 



and larger in the lowest indicate non-homogeneous vegetation. Figure 

 205 shows a very homogeneous community; Fig. 20 A, one with medium 

 homogeneity. That these graphs cannot be taken to indicate the 

 maturity or the immaturity of a community is shown by the diagram of 

 the Scirpeto-Phragmitetum (Fig. 20B). This is from a wide ditch dug 

 about twenty years ago. Today it is covered for a long distance with a 

 dense and extremely uniform vegetation. 



Communities which are apparently extremely uniform in physiog- 

 nomy may actually show very little floristic homogeneity. The 



