226 VELD ESTIMATIOX. 



parative cliart of gTaplis for the figures for such a transect 

 shows one mass of points, or apices of graphs. The position 

 of any apex in a graph fixes the evhahitat in the transect of the 

 species concerned. 



This result was invaluable to me, as giving' me the key 

 factor to the plant which was proved to be the cause of cattle 

 pusliing-staggers. For its euhabitat in 1917-18, and eight 

 years previously from accounts, was found to be in a narrow 

 belt along stream and marsh edges, half in and half out of 

 water (namelv, anipliibious, not necessarilv semi-submerged). 



In 1918-19 and 1919-20 its zonal habitat, owing to 

 abnormal rain, was increased to belts 20 to 200 yards wide in 

 ridge to vlei transects of low angle, 3 degrees to 6 degrees. 

 The outbreak of this disease coincided with increase in spread 

 and growth of tliis plant. It was accordingly arranged to be 

 fed, and it produced the disease. 



5. As information concerning (liniate, rainfall, tempera- 

 ture, humidity, frosting, altitude, locality, etc., are recorded 

 concurrently with the estimation of a number of transects on 

 an area, it can well be seen that if a number of botanists, 

 using the same units, were engaged i]i such survey throughout 

 the country, and such surveys were collated and compared, 

 then the most valuable and exact data could be had as to — 



(a) The geographic distribution of species, and the causes 



of such distribution. 



(b) The practice of veld burning in the various floral 



regions of the Union, so that uniform laws could be 

 drafted regarding its advisability, etc. 



(c) The systematic valuation of the ranching and agTicul- 



tural possibilities of all the land in the Union (in 

 conjunction with the soil survey). 



(d) The causes of the disappearance of valuable timber. 



herbage and wild flowers could be ascertained, and 

 control of these baneful factors instituted. 



