130 NATURAL HISTORY OF SCOTCH FIR 



the species of pimis sylvestris, a comparison of wliich shows most 

 manifestly the existence of local varieties and the necessity of dis- 

 tinguishing them. A trunk, thick and knotty, and thick-set, covered 

 with a bark, coarse and much cleft, brown in the lower part, and grey 

 on the rest of the tree ; the crowns horizontal, very much drawn to- 

 gether, lining the tree from its base, composed of strong branches 

 often flexuous, which famish and at times annihilate the trunk, and 

 the ensemhle of which forms an enlarged and diffused head ; such are, 

 with about five or six exceptions, among the thirty individuals of 

 which this lot is composed, the pins sylvestres of the High Alps. 



" It may be seen from this that they are amongst the lowest rank 

 of the species to which they belong, and that in the formation of pine 

 woods they should be avoided. But this exclusion ought not to 

 extend to all possible cases ; there is one in which the Briangon pine 

 may not only be useful, but better than all others of the species ; it 

 may be so where plantations are formed on the brows of mountains, 

 or on their plateaux exposed to the violence of winds. The little dis- 

 position manifested by it to attain height, the thickness of its base, 

 the strength of its lower branches, which are persistent for a length 

 of years, almost carpeting the ground, make it more suitable than 

 any other of the pwis sylvestres to hold on and grow in such situ- 

 ations ; also when one is re-clothing mountains with woods, the Brian- 

 9on pine may be employed with great advantage conjointly with the 

 Mugho — the former to garnish the middle region of the declivity, the 

 Mugho for the higher zone — for these are the respective places which 

 Nature has assigned them on the brows of mountains." 



In writing of the fivefold series, which he had formed, M. Vilmorin 

 says : 



" These five divisions are not by any means equally good, or nearly 

 BO. The first and the last (the elongated pyramidal Riga pine and 

 the pine of Briangon) present the only two types absolutely distinct ; 

 the others have characters much less decided. This could not be 

 avoided, and pertains to the very nature of the species in which the 

 individuals as well as the local masses differ — one may say indefinitely. 

 And thus between the two opposite types which I have just named, 

 the lots in the school form an unbroken chain, which binds together 

 these two extremes. The sub-divisions, as I have said, were necessary, 

 as may be seen indeed by the definition of the five series adopted. 



" The outline, such as I have been able to give it, does not yet 

 suffice for a convenient classification of all the lots. A certain num- 



