II. T. FULLER — DENUDATION AND DEFLORATION. 149 



which flows southward, and Sugar river, which flows northward, and are found yet 

 within a half mile of the glacial moraine which is the watershed between them. 



These terraces were in the early part of the present century all clothed with 

 forests. Some of them have been cleared and all of them cultivated within forty 

 or fifty years. Later they were given up to pasturage, and in the course of fifteen 

 years after I began to notice that the rounded slopes of these sugardoaf lulls on the 

 southwestern sides began to lose their green, and bare sand appeared. Then the 

 sand began to drift, generally toward the southeast, until in some spots acres were 

 denuded of vegetation and other acres were covered three, four, and five feet in 

 depth by the drifted sand. The work of destruction has continued until consider- 

 able parts of large farms are now worthless. The pi lenomena are confined to grassed 

 lands, either mown or pastured. They cannot be caused by the action of water 

 chiefly, because the beginnings of these changes are neither in ravines nor on the 

 sides of ravines, unless, perchance, a slope is toward the south or southwest, but 

 on the swells of the slopes. 



The process appears to be this: First, the pasture is fed oil' or the field mown 

 until the humus or organic matter in the soil, which is always thin, is exhausted or 

 nearly so. The roots of the herbage are feeble and shallow. By and by a dry 

 season occurs, and on the south-southwesterly slope, where the sun's rays strike 

 almost vertically in the hottest part of a summer day, the grass dries up root and 

 1 (ranch. The next spring these very spots lose more quickly than others the snow 

 as it melts under the sun. Then the winds that follow in the months of March and 

 April, generally in fair weather blowing from the Avest or northwest, cut out, as 

 they strike the surface at a very slight angle, the dry sand and transport it to the 

 nearest lower spot to the leeward. Sometimes the drift has gone across highways 

 or through double fences of open rails or boards ; sometimes, indeed, the sand has 

 blown over the higher crest of the ridge and been deposited on ground more elevated 

 than that whence it came. My observations of this denudation and defloration of 

 line silicious soils have covered the valleys of the Androscoggin ami >aco rivers in 

 .Maine and New Hampshire, of the Merrimack and Connecticut and their tributaries 

 in New Hampshire, Vermont, and Massachusetts. But I have noticed the begin- 

 nings, less marked, of the same kind of destruction of vegetation in southern New 

 York on the headwaters of the Alleghany river, and in northeastern Pennsylvania 

 on Oil creek, though these are comparatively newly cleared regions. For these 

 bared knolls one must look on the eastern or northern sides of valleys and for 

 the slopes thai dip a little west of south. The three causes, as I have already inti- 

 mated, are the shallowness or exhaustion of the soil, drought, and wind. I'nless 

 in some way counteracted, the "old fields" of the north may yet, if not in extent 

 yel iii desolation, vie with the "old fields " of the south. The only remedy is fer- 

 tilizing and sheltering the bared spots, planting trees to the windward, abandoning 

 grazing, and letting the forests again as of old occupy and reclaim and enrich in 

 nature'.- own way the areas which continued cropping lias exposed to waste by 

 drought and varied erosion. 



