ANTHKLIA : AN ARCTIC-ALPINE PLANT ASSOCIATION 87 



aespitosa (arctic-alpine form), and Polygonum viviparuin are 

 widely distributed over other arctic-alpine associations. 



The various authors point out that the conspicuous 

 flowering species of the snow-flush vary from place to 

 place. Here PolytricJmm may form a carpet, there Salix 

 herbacea a mat, or Gnaphaliiini in white tufts. Riibel gives 

 prominence as tone-imparting species to Polytrichum, Salix 

 herbacea, Sibbaldia, Alchemilla, Gnaphalitnn, and Ligusticum 

 Mutellina, and he demonstrates that the association presents 

 sub-types according to the dominant plant ; thus Arenaria 

 biflora is given as much more abundant when Polytrichum 

 is the conspicuous plant ; on the other hand, Gnaphalimn is 

 fairly constant in all the sub-types of snow-flush. This 

 patchy occurrence of various plants in separate stations or 

 within the same snow-flush leads Brockmann to the conclu- 

 sion that accident or chance plays an important part in 

 the constitution of the plant-covering. In other words, the 

 snow-flush is an open association into which species from 

 neighbouring plant communities migrate. The existence of 

 comparatively large pure patches of a single species, 

 frequently observed in the snow-flush, is attributed to the 

 capacity of the plants for vegetative propagation. It seems 

 to be essential for existence in the snow-flush habitat that 

 the species should form low matted tufts sufficiently com- 

 pact to resist periodic flooding, and along with this the 

 power to extend laterally by short horizontal shoots. 

 Gnaphalium supinuni forms mats from which the short 

 flowering stems arise, and short horizontal shoots extend 

 along the ground to give off" new leafy shoots which flower 

 in some future year. Cerastium, Sibbaldia, Alchemilla, and 

 Veronica alpina extend by lateral branches rooting in the 

 underlying carpet of Hepaticas or Moss. Salix herbacea 

 forms a loose mat extending comparatively rapidly by 

 underground horizontal branches which root and may form 

 new plants. That this habit of growth leads to consider- 

 able stability may be experienced when one collects 

 specimens ; either one brings away much soil attached, or 

 the shoots break off" short, leaving most of the plant 

 behind. 



Another interesting adaptation is noted by Brockmann- 



