THE CONICAL GROWTH OF TREES. 63 



only nine lines annually ; and the cross section of the branch 

 actually shows the three inner rings or woody layers, formed 

 by the leaves of the first three years, to be much broader 

 than the four outer rings, the leaf deposits of the last four 

 years. 



The same relation between the breadth of the wood-rings 

 annually formed, and the extent to which the main axis is 

 developed, will still continue to subsist even after the side axes 

 have grown to some considerable extent, provided their growth 

 is accelerated or falls back together with that of the principal 

 axis. If the reader will refer to Figure 2, he will see that the 

 primary axis made a considerable growth the first three years, 

 or between 1852 and 1855, and that from 1855 to 1856, the 

 growth of the axis received a remarkable check, the vegetation 

 of the side axes being retarded at the same time. Now, let him 

 look at the rings or breadths of the respective cones annually 

 formed, which are represented in the cross-section of the axis 

 at the bottom of the diagram, and he will see that the wood 

 ring or conical stratum of woody matter deposited the fourth 

 year, is much narrower than the other rings. 



It is not, however, always the case in a system of axes, that 

 the growth of the primary and secondary axes advances or falls 

 back together. It not unfrequently happens that the growth 

 of the primary axis is retarded, whilst at the same time some of 

 the secondary axes make considerable headway. Hence, when 

 the primary axis puts forth one or more generations of side 

 shoots, the growth of each must be taken into consideration in 

 estimating the amount of wood formed by the leaves. A Beech 

 branch, for example, eighteen inches long and nineteen years 

 old, shows on the transverse section of its wood, from within to 

 without, sixteen narrow and then three broad rings. These 

 breadths of the wood-rings do not correspond with the succes- 

 sive annual growths in length made by the principal axis ; for 

 in this case quite another result is obtained, that axis having 

 hardly grown at all after the sixteenth year. The three broad 

 wood-rings last formed must therefore have derived their matter 

 from the branches, and we find on examining the side axes, 

 which are thirty-three in number, that eleven of them, formed 



