6S MEASUREMENT OF WHOLE WOODS. 



(c.) Several tiualiLies exist, Nvliich change gradually from 

 one to the other. In this case, the sample plot may 

 take the shape of a strip, ^Yhich runs through the 

 whole wood, so as to include a due proportion of each 

 quality (Fig. 27). As this is difficult to accomplish, 

 it is generally hetter to follow the method given 

 under (/>), to divide the wood into several parts and 

 to take a sample plot in each. 



(<^.) Several qualities prevail irregularly over the whole 

 wood. Here a sample plot of average stocking must 

 be selected, a matter frequently beset by great 

 difficulties. 



3. Extent and Shape of Sample Plots. 



The sample plot must be of sufficient extent, to contain the 

 different classes of trees in the same proportion as the wood. 

 Hence, its size depends on the degree of regularity of the 

 stocking; the more uniform this is, the smaller ma}^ be the 

 sample plot. It follows, that they may be made smaller in 

 young fully stocked, than in old irregularly stocked woods. 



Very small sample plots have the disadvantage that propor- 

 tionately too many trees fall into the boundary lines. 



The absolute extent of the sample plot depends on the 

 desired degree of accuracy. In mature woods, it should not be 

 less than 5 per cent, of the whole area, but in young \YOods it 

 may be much less. 



The best shape would be that which includes the greatest 

 area as compared with the boundary, in other words, a circle. 

 As this is impracticable, it is usual to give to the sample plot 

 the shape of a square, or of a rectangle approaching a square. 



4. JSIcasuremeiit of Valuvte on Sample Plots. 



This can be done according to any one of the methods 

 described above. As here a conclusion is drawn from the 

 volume of a small area to that of the whole wood, it is desirable 



