THE DISTRIBUTIOX OF INSECTS IN NEW HAMPSHIRE. 337 



the heads of ravines descending from it, as the alpine region, and have 

 sometimes spoken of a small tract intemiediate between the two, but 

 still imperfectly characterized, as the sub-alpine region;" and this is the 

 most definite mention of a sub-alpine as distinct from an alpine zone yet 

 made by botanists. 



More than ten years ago, however, I pointed out * that two distinct 

 zones of life existed above the limits of forest growth in the White 

 Mountains, each of which was characterized by the presence of distinct 

 animals. So little has been added to these obsen'ations, that I have 

 incorporated them into the present essay. 



One feature of the White Mountain vegetation strikes the most casual 

 observer, viz., the abrupt limit of the forest growth upon these mountain 

 slopes, marking a very natural division into a wooded and a woodless dis- 

 trict. An observant eye will detect in the latter a further subdivision 

 into two regions, — a lower, where the dwarfed spruce, struggling upward, 

 conceals the gray rocks by a covering of uniform green, broken only by 

 the land-slips which have scarred the declivities with their lengthened 

 furrows, or, by the steeper faces of precipices, where trees obtain no foot- 

 hold; and an upper, much more restricted area, where the huge blocks 

 of lichen-covered stone lie piled in inextricable confusion, one upon 

 another, or have their interstices filled with sedges, which, on the more 

 level spots, occasionally form small fields like pasture-land, but full of pit- 

 falls and irregularities. 



These three zones (the forest district, the district of the dwarfed spruce, 

 and the rocky district) exhibit in a general way the limits of the moun- 

 tain, the sub-alpine, and the alpine regions ; and also corresjMnd, in the 

 characteristics of their inhabitants, to the Canadian, the Hudsonian, and 

 the sub-arctic or Labradorian faunas. They do not, however, correspond 

 to the divisions indicated by Tuckerman, for the "heads of ravines" and 

 all the surrounding districts belong to the sub-alpine region, while the 

 alpine is confined to the topmost areas of only the very highest jDeaks. 



The separation of the mountain from the sub-alpine region is well 

 marked by the limit of the forest growth, and this is so abrupt that a 

 narrow belt of a few rods is usually all that intervenes between the spruce 



* Bosl. Jouru. Nat. Hist., vii, 612-621 (1863). 



VOL. I. 45 



