1919] Hall-Goodspecd: Chrysil 197 



slowly crowded out by a new generation of the more aggressive Sage- 

 brush. The result is that the gray forms are much more scattered 

 than the green ones and almost never form pure stands of any great 

 extent. Thej^ are sufficiently abundant in some places to yield a 

 supplementary supply of rubber in case the green forms are harvested 

 from near-b}- areas, but they should not be considered in making esti- 

 mates of the total amount of native rubber available. Their chief 

 interest lies in the fact that they are very erratic as rubber producers 

 and might therefore furnish starting points in case experiments were 

 undertaken in the breeding or selection of superior strains. 



VII. ESTIMATES BY DISTRICTS OF THE AMOUNT OF 

 CHRYSIL AVAILABLE IN WESTERN NORTH AMERICA 



Although field studies on the occurrence and distribution of rubber- 

 bearing shrubs have been carried on in all of the western states, actual 

 estimates have been made for only a few of the areas upon which they 

 are known to grow. The sudden termination of the war has removed 

 the immediate need of a complete survey, but if an emergencj^ call 

 is ever made the work thus far completed will furnish a starting 

 point for a definite estimate of the total amount obtainable as well 

 as indicate the localities where it may probably be found in the largest 

 quantities. Somewhat detailed estimates for the areas examined are 

 therefore presented by districts. 



This phase of the work has been carried on chiefly by the senior 

 author, who has himself visited a majority of the areas reported upon 

 and who assumes full responsibility for the estimates here given. His 

 studies have been supplemented by those of a number of other botanists 

 as mentioned on page 186. 



The estimates were prepared in the following manner. Districts 

 where the brush grows abundantly were visited and traversed when 

 possible in several directions. Data thus accumulated, supplemented 

 by such information as could be obtained from reliable residents, 

 furnished rough estimates of the acreage covered, the percentage of 

 vacant land within the area, and the ratio of Rabbit-brush to other 

 shrubs. Small quadrats were then selected and the number of plants 

 per acre determined. Average samples were weighed or at least 

 estimated as to weight. From these data the total tonnage of shrub 

 has been calculated. A sufficient number of samples has been analyzed 



