260 



situations : heat and shelter being its cure or preventive, it 

 is? seldom seen in close plantations. 



For such exceptions to general success, it is not easy to 

 account, especially in cases, where our subjects are in the 

 full possession of the protecting properties, particularly that 

 of competent and healthy roots : but it is certain, that such 

 exceptions occasionally occur, with almost every sort of tree, 

 even the hardiest, and those the least sensitive of drought. 

 It is, moreover, a remarkable fact, that I have sometimes 

 found those plants of the late or aboriginal oak (the hardiest 

 of all that genus) thrive the best, which had the worst roots, 

 that is, the fewest in proportion to the top ! The truth is, that 

 vegetable physiology is in itself an obscure subject ; and 

 although many important facts respecting it are known and 

 ascertained, by microscopical observation, and careful study, 

 yet probably so many still remain to be discovered, that it is 

 impossible to suggest a satisfactory theory for every anomaly 

 that occurs in practice. But in the particular instance above 

 mentioned, namely, occasional backwardness in the growth 

 of trees, when least expected, if no theory should be found 

 to account for it, there seems yet a certain remedy for the 

 evil, which experience has pointed out. 



If a tree appearing backward, be from five-and-tvventy to 

 thirty feet high, and of girth in proportion, let four cartloads 

 of tolerable earth be taken, of a quality rather opposite than 

 similar to that, on which it has been planted, that is, for 

 hght soil, of the aluminous, and for heavy, of the silicious 

 character ; to which let a cartload of coal-ashes be added, 

 with the rough cinders carefully riddled out. Let the 

 whole be laid round the tree, and very intimately mixed in 

 the manner above pointed out for extra earth and compost, 

 and particularly when made up of manure of this species.* 

 Then, let the composition be spread on the surface, from the 



♦ See pages 217, 218. nnleh. 



