Further analysis v;as made in testing several hypotheses by use of 

 chi-square analysis along the lines suggested by Snedecor.-^ The 

 follov;ing interpretations \vere made: 



1. For the two plantations, the percentages of trees falling in 

 each of the three condition classes were essentially the same, indicating 

 that the same extent and kinds of injury can be expected regardless of 

 location. 



2, The degree of feeding on trees under 3 feet in height is 

 significantly less than the degree of feeding on trees over 3 feet in 

 height. For the former 12,9 percent is an unbiased estimate of the 

 proportion fed on; for the latter 29.0 percent is an unbiased estimate 

 of the proportion fed on. 



3» The number of trees in each injury (or noninjury) category 

 is in direct proportion to the sample size. 



4. The ratio of undamaged to damaged to girdled trees follows 

 approximately an assumed 16: 4-: 1 ratio, respectively. 



From observation there are no sizable sections of either plantation 

 in which trees are girdled to such an extent that understocking now 

 exists. Since feeding by these rodents is continuing, hqi^vever, under- 

 stocking could become a reality in the near future. 



It is obviously desirable to minimize damage by large rodents in 

 young plantations like those described here. The two methods of 

 accomplishing this are direct and indirect control of the mammals 

 involved. In all areas requiring artificial regeneration, but especially 

 in large burns, direct control measures by poisoning are made difficult 

 by the presence of stock. Despite this it is still possible and perhaps 

 advisable to poison, even though the initial cost of plantation 

 establishment is increased. Indirect methods, though less obvious, but 

 certainly less expensive, are applied by encouraging natural enemies. 

 The open season in Idaho on all hawks and owls, the reduction in numbers 

 of cougar, and the organized poisoning of the coyote militate against 

 the natural control of these large rodents. An adjustment of the 

 regulation applying to birds v;ould be a limited but fonvard step in 

 reducing numbers of these undesirable rodents and protecting the 

 growing stock in Idaho plantations. 



1/ Snedecor, George , Statistical Methods. 4th Ed., 485 PP» Ames, 

 Iowa. 1946. 



-2- 



