« AnkstesnisTęsti »
ance of the opinion that the loess has settled on the dry surface of the continent beneath the atmosphere. So soon as I was in a good position to observe, and while on my first journey through loess regions I conceived this idea. But the difficulties of its application grew daily greater, just in the same measure as I, by observing, came to know more of the vast extent of this formation.
Nowhere has anyone observed deposits on so grand a scale (except in the case of masses of matter shot out of volcanoes), in which one could have supposed a subaërial origin. They were in fact taken for a formation of an inferior kind and in vain would anyone search in Handbooks of Geology for agencies, which could explain the origin of such enormous accumulations of the finest earth without the help of flowing waters. The proofs, however, were altogether beyond refutation to begin with, and beside these negative arguments already given, there are a great number of positive arguments, of which I will now speak.
The first is founded on the way in which the shells of the landsnail come before us. I have mentioned already that these make their appearance exclusively as fresh water snails; that they are strewed everywhere through the whole immensity of the deposits, sometimes sparingly, sometimes heaped together in masses, and that the bleached shells, although they are fine and delicate, are nearly without exception well preserved. We must, therefore, suppose that each of the little creatures died on the spot where we find its shell and further that this shell was not exposed to the influence of any destructive agencies. It is possible to have another explanation for the appearance of the snail-shells. It is to be remembered that a number of different kinds of snails are accustomed to hide during winter several yards beneath the surface of the earth and to make their appearance again in spring. It is believed that many die there and leave their shells in their hiding places. This is without doubt the origin of a great many of those shells which are found in the loess near the surface. This fact, however, is not sufficient to explain the appearance of great accumulations of shells at depths of several hundred feet where I myself have found them in places at the bottom of deep cuttings in the loess where caves, to be used for human dwellings, have been newly hollowed out some 60 or 80 feet in width. The snail shells thus located lead to the acceptance of the idea that the loess grew very slowly in the period of its origination, that its surface at any given period contained the necessary dampness for the success in life of the little creatures then living there, and for the plants which were their food, but that the climate was
still so dry, that it favored the preservation of their chalk shells whether the animals died underground or at the surface of the earth.
To similar conclusions would the bones of land mammals lead us if we could discover their presence with greater certainty. These animals, as well as the snails, must have died in the proximity of the place where their remains are found and consequently suggest that they lived on dry land. The third proof is formed from the traces of vegetation, which do not consist of actual remains of plants, but of millions of empty spaces which have preserved the form and ramifica tion of plant-roots. Perhaps there arise doubts as to the justice of this explanation of their origin. But if we mark now the canals on the walls of cuttings in the loess, which the roots of living plants dig for themselves and which they leave behind when they die, we find that they resemble exactly those which we find in all parts of the same locality; each of the channels once enclosed one of the root fibres of a plant which grew on the surface. If therefore we see before our eyes a loess-bank of a thickness of several hundred feet. in perpendicular cutting, we must suppose that each minute division of an inch from the foot to the surface marks the spot where in former times the surface of the deposit happened at the time to be, and that, as long as the slow growth of the soil continued, the surface for the time being was covered with vegetation.
It remains still to find the agencies which procured the material for the gradual elevation of the soil in the loess deposits. At the first a slight observation leads to three different kinds. The first kind of agency is the rain-water, which drizzles down from the higher to the lower parts and which on the decomposition of the stony surface of the adjacent mountain ranges washes away those constituent parts which have become loose.
The second is the wind, the extraordinary effects of which, in the accumulation of hard material broken up into the form of dust, one has continually opportunity to observe in those countries.
The third of these agencies lies in the mineral parts, which the roots of innumerable grass plants, thanks to the moisture diffusing itself through them, draw from the soil and which they leave behind them when they decay. These various constituent materials finely divided are kept from dispersion by a cover of vegetation and it is only an unimportant quantity of them that is carried away by the wind. On the basis of this explanation I have called the loess in my first notice on this subject (Letter on the provinces of Honan and Shansi, 1870) a vast burial ground of the corpses of innumerable generations of grass plants. But at that time proof was still want
ing. It was quite clear that we could only expect its confirmation in some locality where a gradual growth of the soil still takes place in the above mentioned way. I supposed that this must be the case in Mongolia. The desire to prove the loess theory as then stated was that which in particular induced me to go to the steppes, and I had the gratification to obtain there the establishment of that theory by distinct proofs. Instead of finding a uniform undulating plateau, as this part of Mongolia is often in books described to be, I was surprised to find the same forms of the surface, which are so characteristic of the loess if one subtract from it the systems of the washed out ravines which are formed at a later stage. One sees those same troughs which extend from ridge to ridge, and which, looked at cross ways, resemble a rope loosely stretched between two persons, or which either trend down from three high sides to a fourth which remains open or form a shallow basin closed all round. And these trough-shaped hollows may be recognised as characteristic of the form of the surface in Central Asia generally.
Here are given all those conditions the presence of which at the time of the subaërial formations of the loess is seen to be needed by simply observing the loess as it now exists. The rocky skeleton of the ridges forming the sides of the basins so far as it is still uncovered underlies the detritus caused by moisture and warmth through the influence of vegetation in summer and the formation of ice in winter. Rain washes or pushes the larger fragments, without injuring their edges, gradually down on that side of the basin which slopes the most, and so the filling up goes on. In case of periodical torrents of rain, the fragments will be washed away to a much greater distance, smaller pieces being found towards the middle of the basin, where they will ultimately be covered by fine earthy formations during dry periods. This is entirely analogous to the above-mentioned flat strips of land which separate two loess banks. At each renewed attack the finer parts would be washed further away and they would be found only towards the middle of each of the greater basins. We can see these processes taking place before our eyes. Add to this the appearance of salt on the surface, which is drawn from the depths below by force of the rain and which proves another of our suppositions. The third proof shows itself in the frightful dust-storms which leave a deposit destined in each separate locality, after years or centuries, to prove a not unimportant factor in the elevation of the soil. In other countries the dust is mostly washed away. In these basins without drainage the greatest part remains in the place where it fell, detained there firmly by the vegetation which springs up, helping in this way
the growth of the steppes. But, as I have already pointed out, the salt steppe does not show its formation, because. its curvature from the sides to the central salt lake has exactly the right form to cause the water in shallow beds to flow towards the central salt lake. If ever a sudden event should cause a deepening of the area of drainage, and thereby work a deeper cutting of the brooks in their under course, yet in course of time there would take place a change from depth to shallowness through the addition of new sediment leading to the restoration of the natural angle of inclination. If the middle part should be elevated so much that its soil should become horizontal far above the borders of the lake, the deposition on the slopes would continue in force until the curve took again its earlier form.
It is for this reason that the nature and mode of the filling up of the steppe hollows remains to our eyes not perceptible as long as we continue in any of those localities where there is no outward drainage. The analogy of the formation of the plateau surface with that of the basins of the loess, the resemblance in the way in which the basins in both cases are divided and the circumstance that the observation of the loess shows us the process of formation, as it actually takes place on the salt plateau, all this, I say, makes it in the highest degree probable, that the ravines in the loess basins represent distinctly the nature of the inner structure of the basins of the plateau. Still we can also get a proof of absolute identity. In the loess district in the North of China if a traveler ascend toward the steppe, he comes to the last springs of various rivers, at places where it is irresistibly clear, that here is a district which a short time ago had no outflow, but has now so changed that it sends its waters either through the Yellow River or through the Peiho to the sea. Such localities are found on the other side of the Great Wall in parts of Shansi and Chili along the border of the grass land. It is probable that in each case of this kind, the point of the level of the standing water, whether above the surface or under it was so far heightened in a depressed locality during a rainy period, that it reached a rift in the rock or arrived at the lowest point in a hollow bend and found there an outlet. As soon as this was done and a canal dug, which drew off the salt lake, the flowing away of water would continue even though the climate became somewhat drier. As then the newly dug canal where the water continued to flow off, became deeper, it would in course of time naturally follow that the water would cut its way deeper into the soft parts of the plateau. The same took place with the principal canal, as well as with all the affluents which discharged themselves in former times
into the lake and those which formed themselves anew. In this manner originated the endlessly ramified water rifts which laid the loess open to view. It can at the present time be seen everywhere near the springs which burst forth along the borders of the plateau and even if the cuttings are only one foot deep, the loess character of the formation is still not to be mistaken. We arrive in this way with sufficient certainty at the two conclusions which here follow:
1. The contents of the plateau basins of Mongolia, with the exception of the Han-hai, where at least the deposit is of marine origin, are composed principally of loess, which is towards the edges of the basins full of deposits of sharp cornered fragments, but is toward the middle formed of finer materials.
2. Each loess basin was formerly a salt plateau basin without an outlet. We get at last a clear picture of the way in which the levelling of the inequalities of Central Asia was accomplished and how its mountains have obtained their covering. But still we want an analogy by which it may be absolutely shown that the assumption of the identity of the originating process in both cases is right. I mention this in the last place, because it is only after the establishment of all other points that it can be made clear. I myself also was led through observations in Mongolia, to seek for it. The salt lake in the centre of each single basin naturally ought to have left in loess countries proof of its former existence. In what way this occurred is easily to be recognized, for if a lake lies in the deepest part of a depression without an opening, and material is carried to it from the sides of the basin, in the way described above, a part of it settles down on dry land as loess, but another portion is carried into the lake and is here deposited on a system of horizontal stratification. While the surface of the whole cavity gradually rises, the horizontal deposits will always be limited to the middle and those with vertical structure to the sides. In this way there will be a stratified nucleus surrounded by masses without stratification, and according to the abundance of water which might prevail at different times the former will encroach on the latter, farther in some localities and in others not so far. If in later times there takes place a drainage of the basin, the whole structure must be laid open, but as a rule the stratified nucleus will be found more destroyed than the surrounding parts on the sides because the flowing away of water takes place principally through the layers in the lake and many of the rifts of the sides of the basin discharge themselves into the lake from the time of their first origination. Nevertheless in nearly every loess basin it can be observed. I first saw the lake stratifica