get the succulent green grass, and. not throug.h -ny special preference.

Paunch, anZalysis seems the most reliable method of gettting at the amount

of young grass taken. 
 
                                Palatabilit.- 
 
          t _.estimate of the palatability of the.var-ious pelat species

presents some difficulty, :since apparent palatability of foraýge
must 
be correlated with its availa-bility. The principal for,-ge plants have 
been arranged. bypreference shown, in wvhat at this time seems their 
logical classes. It is believed that choice forage mray include plants 
whose degree of use is considerably less than 80% of the tota-l current 
growth. The.smue is true of the lesser degrees of use within the 
class ification. The following table is based on winter. and early spring

observations only: 
 
  Choice            Good.          Medium           Low          Inferior

 
Bitterbrush    Gray Chryso thaam.us Sage         C1ryso thaimus  Dwarf 
                                                    .(geen)       map ple

Ceanothus      Douglas Fir         ..Forsellesia Serviceberry   Bearberry

Mt. Mah.ogany Balsamroot           Willow       Wild currant   Dogwood.-.

Wheat-rass)Spring                   Syringa      N ine bark     Wild. Rose

Fescue     )   Cheatgras s (Spr ing) Elder        Choke chery r  Spireea

                                    Sedges       Oregon grape   Eriogonum

                                                 Yellow pine 
 
           It is thought that a study of paunech analysis mazy furnish, the-

most reliable data on palatability. The value of paunch analysis would 
be greatly increased if -paunmches could be taken at regular intervals 
from selected areas where the composition of vegetative cover is known. 
 
                          Mechnical Influences. 
 
          Formation of contour trails is very evident in deep snow. The 
 forrmation of such trails on bare ground does not occur with sufficient

 frequency to destroy mach vegetation. On the lower ranges it is diffi- 
 cult to judge- the influence of deer on contouring,,! since doraes tic stock

 have ply.ey.a irportnt -art. 
 
           Other than the ees exa'dined last fall in Bernard. and Sheep 
 Creeks, no serious erosion has been observed. 
 
                          Census and Distribution 
 
           Portions of practicrally every creek have been counted at some

 time during the winter. 
 
           Beginning with the middle of January, monthly, or neaxr monthly

 counts were made on portions of the following creeks, some being counted

 twice only: Vicinity of Range Creeh including Sunflower Creek, head of