forests, at least when not cut off from sea-winds by interposed chains of equal altitude. Trees such mountains will have. The only and the real wonder is, that the Sierra Nevada should rear such immense trees!

Moreover, we shall see, that this forest is rich and superb only in one line; that, beyond one favorite tribe, it is meagre enough. Such for situation, and extent, and surrounding conditions, are the two forests-the Atlantic and Pacific-which are to be compared.

In order to come to this comparison, I must refrain from all account of the intervening forest of the Rocky Mountains,— only saying, that it is comparatively poor in the size of its trees and the number of species; that few of its species are peculiar, and those mostly in the southern part, and of the Mexican plateau type; that they are common to the mountainchains which lie between, stretched north and south en echelon, all through that arid or desert region of Utah and Nevada, of which the larger part belongs to the great basin between the Rocky Mountains and the Sierra Nevada: that most of the Rocky Mountain trees are identical in species with those of the Pacific forest, except far north, where a few of our eastern ones are intermingled. I may add that the Rocky Mountains proper get from twelve to twenty inches of rain in the year, mostly in winter snow, some in summer showers.

But the interior mountains get little, and the plains or valleys between them less; the Sierra arresting nearly all the moisture coming from the Pacific, the Rocky Mountains all coming from the Atlantic side.

Forests being my subject, I must not tarry on the woodless plain-on an average 500 miles wide-which lies between what forest there is in the Rocky Mountains and the western border of our eastern wooded region. Why this great sloping plain should be woodless-except where some cotton-woods and their like mark the course of the traversing rivers-is, on the whole evident enough. Great interior plains in temperate latitudes are always woodless, even when not very arid. This of ours is not arid to the degree that the corresponding regions west of the Rocky Mountains are. The moisture from the Pacific which those would otherwise share, is-as we have seen-arrested on or near the western border, by the coastranges and again by the Sierra Nevada; and so the interior (except for the mountains), is all but desert.

On the eastern side of the continent, the moisture supplied by the Atlantic and the Gulf of Mexico meets no such obstruction. So the diminution of rain-fall is gradual instead of abrupt. But this moisture is spread over a vast surface, and it is naturally bestowed, first and most on the seaboard district,

and least on the remote interior. From the lower Mississippi eastward and northward, including the Ohio River basin, and so to the coast, and up to Nova Scotia, there is an average of forty-seven inches of rain in the year. This diminishes rather steadily westward, especially north westward, and the western border of the ultra-Mississippian plain gets less than twenty inches.

Indeed, from the great prevalence of westerly and southerly winds, what precipitation of moisture there is on our western plains is not from Atlantic sources, nor much from the Gulf. The rain-chart plainly shows that the water raised from the heated Gulf is mainly carried northward and eastward. It is this which has given us the Atlantic forest region; and it is the limitation of this which bounds that forest at the west. The line on the rain-chart indicating twenty-four inches of annual rain is not far from the line of the western limit of trees, except far north, beyond the Great Lakes, where, in the coolness of high latitudes, as in the coolness of mountains, a less amount of rain-fall suffices for forest growth.

We see, then, why our great plains grow bare as we proceed from the Mississippi westward; though we wonder why this should take place so soon and so abruptly as it does. But, as already stated, the general course of the wind-bearing rains from the Gulf and beyond is such as to water well the Mississippi valley and all eastward, but not the district west of it.

It does not altogether follow that, because rain or its equivalent is needed for forest, therefore wherever there is rain enough, forest must needs cover the ground. At least there are some curious exceptions to such a general rule, exceptions both ways. In the Sierra Nevada we are confronted with a stately forest along with a scanty rain-fall, with rain only in the three winter months. All summer long, under those lofty trees, if you stir up the soil you may be choked with dust. On the other hand, the prairies of Iowa and Illinois, which form deep bays or great islands in our own forest-region, are spread under skies which drop more rain than probably ever falls on the slopes of the Sierra Nevada, and give it at all seasons. Under the lesser and brief rains we have the loftiest trees we know: under the more copious and well-dispersed rain, we have prairies, without forest at all.

There is little more to say about the first part of this paradox; and I have not much to say about the other. The cause or origin of our prairies-of the unwooded districts this side of the Mississippi and Missouri-has been much discussed, and a whole hour would be needed to give a fair account of the different views taken upon this knotty question. The only settled thing about it, is, that the prairies are not directly

owing to a deficiency of rain. That, the rain-charts settle, as Professor Whitney well insists.

The prairies which indent or are enclosed in our Atlantic forest-region, and the plains beyond this region, are different things. But, as the one borders-and in Iowa and Nebraska passes into the other, it may be supposed that common causes have influenced both together, perhaps more than Professor Whitney allows.

He thinks that the extreme fineness and depth of the usual prairie soil will account for the absence of trees; and Mr. Lesquereux equally explains it by the nature of the soil, in a different way. These, and other excellent observers, scout the idea that immemorial burnings, in autumn and spring, have had any effect. Professor Shaler, from his observations in the border land of Kentucky, thinks that they have, that there are indications there of comparatively recent conversion of oak-openings into prairie, and now-since the burnings are over of the reconversion of prairie into woodland.

I am disposed, on general considerations, to think that the line of demarcation between our woods and our plains is not where it was drawn by nature. Here, when no physical barrier is interposed between the ground that receives rain enough for forest, and that which receives too little, there must be a debateable border, where comparatively slight causes will turn the scale either way. Difference in soil and difference in exposure will here tell decisively. And along this border, annual burnings for the purpose of increasing and improving buffalofeed-practiced for hundreds of years by our nomade predecessors, may have had a very marked effect. I suspect that the irregular border line may have in this way been rendered more irregular, and have been carried farther eastward wherever nature of soil or circumstances of exposure predisposed to it.

It does not follow that trees would re-occupy the land when the operation that destroyed them, or kept them down, ceased. The established turf or other occupation of the soil, and the sweeping winds, might prevent that. The difficulty of reforesting bleak New England coasts, which were originally well wooded, is well known. It is equally, but probably not more difficult to establish forest on an Iowa prairie, with proper selection of trees.

[To be continued.]

ART. X.-On the Structure and Origin of Mountains, with special reference to recent objections to the "Contractional Theory;" by JOSEPH LECONTE.

(Read before the National Academy of Sciences, April 19, 1878.)

In order to write intelligibly on this subject it is necessary first of all to define clearly in what sense we shall use the word mountain. In popular and even in scientific language this word is used to express every considerable elevation above the general level of the earth surface, whatever be its extent or its mode of origin. It is applied equally to a complex system of ranges formed at different times, such as the Andes, the North American Cordilleras, or the Appalachian; or to each one of the components of such a system, as for example the Coast Range, the Sierra or the Wahsatch or to each one of the component parts of these components, or even to separate isolated peaks upon these last whether formed by erosion or by volcanic ejections. In this paper I shall call an aggregate of ranges formed at different times a mountain system or chain. Each monogenetic component of such a system, such as the Sierra I shall call a Range. The components of these again, whether formed by erosion or by foldings, or by fissures and slips, I shall call Ridges. Isolated peaks, whether of erosion or of volcanic ejections, are so obviously distinct in their mode of origin, that they need not trouble us in this discussion.

Now, a theory of mountain-formation is essentially a theory of the formation of Ranges. For, on the one hand a mountainsystem is but an aggregate of ranges more or less parallel to each other in the same general region but formed at different times, and the addition of range to range in the formation of such a polygenetic system, even though there may be a general bulging of the whole region, introduces no new element in the discussion; and on the other hand the subsequent formation of ridges and peaks by erosion belong not to the category of mountain formation at all, but to that of mountain-sculpture. Sometimes, it is true, ridges are formed also by faulting, but in all cases they are subsequent to the formation of the mountain proper-in all cases they belong to the category of mountain decay, not to that of mountain-origin. It seems to me that much of the refined classifications, and of the minute divisions and subdivisions of types of mountains, in which some recent writers have indulged is the result of an imperfect recognition of the distinctions enforced above. Limited, as above, mountain ranges are, I believe, always formed by horizontal pressure *A well chosen word often helps greatly to clear up a subject. We are indebted for this one to Professor Dana.

crushing the strata together and thus producing foldings and thickening and consequent elevation. In what I shall say of mountain-structure I shall be compelled for the sake of clearness to assume this. I hope to justify this assumption in what I shall subsequently say on mountain-origin.

I. Mountain-Structure.

Mountain ranges may be conveniently divided into two general classes which, however, graduate completely into each other, viz: those which are composed of a single anticlinal fold, and those which are composed of a number of folds alternately anticlinal and synclinal, either open, as in the Jura, or closely appressed, as in the Appalachian, the Coast Range, the Alps, and many others. The one kind is formed where the earth crust is more rigid, the other where it yields more readily to the horizontal pressure. Both kinds are greatly modified, sometimes by metamorphism, sometimes by faulting, sometimes by volcanic outbursts, and always by subsequent erosion.

1. Mountains of a single fold.-The simplest conceivable mountain range consists of a single anticlinal fold of a series of strata. In such cases the deeper strata of the series are thickened and swelled upward by the horizontal pressure, while the upper strata are raised into a vault with little or no thickening, or may even be thinned and broken by tension. Nearly always the yielding is greater on one side than on the other; so that the vault is unsymmetrical. In such cases a great fissure and slip is apt to occur on the steeper side. The following figure (fig. 1) is an ideal section of such a mountain

1.

2.

before erosion had modified its form, or rather (since upswelling and erosion goes on together pari passu) as it would be if restored. Now it is evident that in the formation of such a vault, fissures would almost certainly be formed; and if beneath the vault there should exist a mass of fused or semi-fused matter (sub-mountain molten matter), formed either by the invasion of the deeper sediments with their included waters, by the interior heat of the earth during the preparatory process of sedimentation, or by the heat evolved by crushing in the act of formation itself of the mountain, dislocations would be apt to occur and further, both the fissures and the faults would be most apt to occur just where the bending of the strata is