III. RADIALS 



SELECTION 



An essential part of this study of climate and trees has been the 

 laboratory work on the rings, by which the actual wood from the tree 

 is placed under microscope and measuring-machine. In this way 

 specimens from different trees may be compared together and an 

 accuracy reached which would be hopeless in work on the stump. 

 By laboratory means, cross-identification and correct dating are 

 obtained before measuring and the measuring can be done to any 

 desired accuracy which the rings permit. Hence it is essential to secure 

 ring specimens which represent the tree, to get them to the laboratory 

 without injury, and then preserve them in such a way that they can 

 be used over again or referred to subsequently for any desired purpose. 



Definitions — It is obvious that such ring specimens must be cut 

 across the rings in order to display the proper sequence. The ideal 

 form, therefore, is a radius of the tree, carrying an unbroken series 

 of rings over all parts of the tree's history which are worth while. 

 Such pieces are here referred to as tree-samples, ring records, radial 

 pieces, or simply radials. Of course, they may take different forms, 

 depending on various conditions of collection; for example, whether 

 they come from living trees, fallen trees, or stumps. 



LIVING TREES 



The main point in sampling living trees is to get a short radial 

 sequence of rings without injury to the tree. The best instrument for 

 this is the Swedish increment borer, which will be more fully described 

 in a subsequent chapter on instruments. These borers will not go 

 into hard woods nor even into junipers, but they work well in pines. 



Direction of boring — If the tree is on a steep hillside, it is usually 

 more convenient and customary to bore on the up-hill side. Theo- 

 retically there could be a difference in the rings between the up-hill 

 and down-hill side of a tree, but no such difference has been noticed. 

 Other things equal, it is well to eliminate the possibility by being 

 consistent throughout a group. If the ground is generally free from 

 steep inclination, one should adopt a certain compass direction and 

 use that consistently in the group. Early investigation showed about 

 Flagstaff a slight average increase of growth on the north or northeast 

 side of a tree, due to lingering of snow in the shade of the tree, but this 

 is probably of little or no importance in radial selection. 



Height above ground — Height from the ground, if well below the 

 branches, has not been found to introduce error. So far as observed, 



17 



