ASSOCIATIONS OF RIVERS AND STREAMS 337 



(c) Winter period during which Ulothrix and Hormidium re- 

 appear and the Diatomaceae increase. 



The most important controlHng factor is apparently tempera- 

 ture whilst the chemistry of the water is also significant, although 

 local modifications of the flora may be brought about by changes of 

 light intensity and oxygen concentration. When compared with the 

 floras of streams from other areas it is interesting to note that the 

 same species often occur in widely different types of habitat, thus 

 providing a proof of the indifference of those plants towards 

 habitat. The algal communities could be divided into three groups, 

 those occupying a vertical substrate, e.g. waterfalls, those occurring 

 on a horizontal substrate and those which are free-living. 



A. Algal communities of vertical substrates : 



(i) Those attached to stones; eight communities were distin- 

 guished. 



(2) Epiphytic communities; four were distinguished, three of 

 which also occur in (i). 



(3) Three spray communities. 



B. Algal communities of horizontal substrates: 



(i) Those attached to stones, sand or mud; three communities. 

 (2) Six epiphytic communities. 



C. Free-living algal communities : two were distinguished. 



In a study of the encrusting algae of streams Fritsch (1929) 

 distinguished {a) filamentous algae, {b) algae of banks and (c) sub- 

 merged encrusting algae: in the particular stream there appeared 

 to be a brief succession terminating in a mat of Phormidium 

 autumnale. 



Interesting results have also been obtained from a study of 

 colonies of the blue-green alga, Rivularia haematites, growing in a 

 stream. It was found that the surface area of the thallus increases 

 greatly in proportion to the attachment area until finally the 

 force of the torrent becomes greater than the prehensile force 

 and the thallus is torn away. In fast streams the thalli are only 

 formed on big stones because the small stones together with the 

 colonies have been swept away. In such fast-flowing regions there 

 appears to be a relationship between size of thallus and size of 

 stone, but no such correlation can be demonstrated in quiet waters. 



C S A 22 



