proportionally on the ground, and each field sample is classified as needing or not needing man- 

 agement treatment by means of field measurements and observations . Nearly all plots needing 

 treatment are classified in only one or two of the 10 strata set up by the first photo scheme. In 

 the second scheme the plots needing treatment are found to be distributed through five or six of 

 the 10 strata used. The first scheme would be most valuable for use on aerial photos since it 

 eliminated eight of the 10 possible strata as a likely source of area needing management. 



Bartlett's test ranks these two schemes by means of the corrected sum of the squares ob- 

 tained from the numbers of plots needing treatment . The scheme with plots needing treatment 

 concentrated in the fewest strata will have the highest sum of the squares and consequently will 

 be ranked first. We recognize that this test offers no means of determining whether any of these 

 techniques is better than no stratification at all, and that the ranking obtained is relative and 

 not on a consistent mathematical scale. Also that the ranks are likely to be nonsignificant partic- 

 ularly in the lower ranks. Nevertheless, it does offer some indication of the better method and 

 has the unique advantage of comparable evaluation for both field- and photo-stratification schemes. 



Table 2. --Variance ratio, expected reduction in field plots, and rank of the 

 aerial photo and map stratification methods for volume estimating 





: Stratification method : 





Cubic volume 





: Saw log volume 





No. 





Number: 



_ _ . : Reduction: 

 Variance . , , 

 : in field : 



ratio 



: plots : 





] Variance ' 

 ratio 



Reduction: 





: Description : 



of : 

 strata : 



Rank 



in field : 

 plots : 



Rank 









- - - -Percent- - - - 





- - - -Percent- - - - 







None 





1 100 







MOO 







1 



Stand size and 



















crown cover 



9 



58 



42 



4 



66 



34 



5 



2 



Stand size 



3 



59 



41 



5 



65 



35 



4 



3 



Crown cover 



3 



99 



1 



11 



100 





11 



4 



Cubic volume- - 

 500-cubic-foot 



















class 



13 



34 



66 



2 



30 



70 



2 



5 



Cubic volume- - 

 1 ,000-cubic-foot 

 class and topo- 



















graphic site 



28 



31 



69 



I 



30 



70 



1 



6 



Cubic volume - - 

 1,000- cubic -foot 



















class 



7 



35 



65 



3 



31 



69 



3 



7 



Topographic site 



4 



95 



5 



10 



97 



3 



10 



8 



Species group 



6 



83 



17 



8 



77 



23 



8 



9 



Stand size and 



















density (map) 



9 



76 



24 



6 



76 



24 



6 



10 



Stand size (map) 



4 



77 



23 



7 



77 



23 



7 



11 



Density (map) 



3 



92 



8 



9 



97 



3 



9 



1 Total unstratified variance: Cubic, 2,122,100; Saw log, 69,451,100. 



NOTE: Map stratifications were taken from existing type map rather than directly from 



photo . 



