Table 13 . --Analysis of the effect of elevation of seed origin and test site on height 



and growth of trees in the vigor- quality study 









1 t; o L oJ-LC dllU 



bum ut; ui- 



variation 









: DCEF : 



PRPP 





Combined 









: Elevation : 



Elevation 



1 eb L 



: Elevation 



o ite A 



measured 



: zone^ : 



zone^ : 



site^ 



: zone-^ 



elevation^ 



















4 



6. 



56* 



4.77* 



<1 



7. 



31* 



<1 



8 



1. 



93 



4.84* 



<1 



7. 



64* 



1.86 



1 n 

 i u 



<1 





i . 4 o 



*~ i 



2. 



06 



1.11 



1 9 



i z 



1. 



96 



7 Q 





3. 



46 



^ 1 



14 



4. 



12 



6. 16* 



7.14 



8. 



84* 



<1 



16 



8. 



98** 



7.03* 



12.3* 



13. 



4** 



<1 



19 



8. 



15 



4.98 



31.0** 



10. 



2** 



<1 



Growth between 

















ages : 

















12 S 16 



17. 



6** 



14.1** 



35.4** 



30. 



0** 



<1 



16 5 19 



5. 



04* 



<1 



109.7** 



4. 



04 



<1 



■"-"F" value resulting from the comparison 



2"F" value resulting from the comparison 



^"F" value resulting from the comparison 

 ^* and ** indicate significance at the 5 



Elevation zone mean square 



Trees within elevation zone mean square 



Test site mean square 



Rep within site mean square 



Site X elevation mean square 



Rep within site X elevation mean square 



percent and 1 percent level, respectively. 



Although the differences among families within elevational zones were large (the 

 range between tallest and shortest varied from 10 percent to 35 percent of the mean at 

 different ages), the differences were seldom significant because of the small numbers 

 of families tested and large amounts of within-family variability. The variance attrib- 

 utable to family-by-site interactions was small and nonsignificant except at age 8. 

 The question of family-by-site interactions is even more confusing if one also includes 

 year of planting. A few families were planted in each of 2 or 3 years. For some of 

 these families the seedlings planted in 1955 may have grown faster at PREF than DCEF 

 but those planted in 1957 grew faster at DCEF or vice versa. 



DISCUSSION AND CONCLUSIONS 



To date all of our studies with western white pine indicate that the majority of 

 the variation which might be utilized in a tree improvement program is related to the 

 differences among individual trees within stands rather than to differences among stands, 

 elevational zones, or geographic areas. When stand or area differences were significant, 

 there did not appear to be a pattern to the variation. In most of the tests, seedlings 

 from high elevation parent trees (generally 1,375 m or above) grew slower than those 

 from lower elevation parents. The differences were significant in only about half of 

 the tests and the average growth of seedlings from some high stands was good. However, 



17 



