W. F. Hume — Notes on Russian Geology. 551 



which may be over one-fourth of an inch in diameter, far exceeding 



the general average of their fellows. 



In a brickyard to the S. of the town a further point of interest 

 is displayed, a point indeed to which Prof. Artnachevsky attaches 

 great importance. The higher part of the quarry is formed of a 

 very fine white sand, supposed to be of Miocene age. Its surface 

 had evidently undergone considerable denudation before the super- 

 jacent deposits were laid down upon it. These consist of a thick 

 layer of Loess, which is markedly stratified at the base, the stratifi- 

 cation lines agreeing with the slope of the denuded sand-surface 

 beneath. Prof. Armachevsky regards this as a strong proof of the 

 action of running water, and has pointed out numerous similar cases 

 in his work on the Tchernigov Government, as at pp. 12, 14. In- 

 deed, in almost every exposure mentioned by him in the N. E. of 

 that government the same stratified condition of the base of the 

 Loess is noted, pointing to a different process of deposition during 

 the earlier portion of the period in which it was forming. 



At the back of the town, this sandy clay is undergoing rapid 

 denudation, giving rise to very broken ground, the many-branched 

 ravines, which are one of its peculiarities, being clearly marked. 

 This feature, though already noticeable in the upper parts of the 

 Dnieper basin, attains an enormous development on the Steppe 

 plateaux which lie between the town of Ekaterinoslav and the 

 Krivoi Rog hematite mines. 



In these districts deep gorges are formed under the influence of 

 atmospheric waters, percolating through the extremely porous soil. 

 Possibly owing to the capillary texture produced in it by the roots 

 of plants penetrating perpendicularly through it, the Loess tends to 

 split in a vertical manner. This occurs during the hot summer 

 period, then, when the rains come on, the water finds its way down 

 the cracks, and large pieces are flaked off, leaving almost per- 

 pendicular walls. These precipitous ravines, piercing far into the 

 table-lands, give to the sceneiy (paradoxical though it may appear 

 to say so) an almost mountainous chai'acter, the general effect being 

 enhanced by the vastness of the setting. To so alarming an extent 

 has this action spread, that the railway joining Ekaterinoslav and 

 the above mines is now, in many places, only a few yards from the 

 edge of the abyss, and special precautions have to be taken to 

 divert the drainage from points which are in most imminent peril. 

 Sometimes the wearing away begins from below, where the Loess 

 meets the river, the ravine then slowly works its way into the 

 plateau branching out into wider and wider ramifications. 



At other points, far away from the drainage areas, the Loess 

 blends with the other surface deposits in giving rise to broad 

 continuous Steppe plains, devoted in large measure to the cultivation 

 of cereals, and extending in almost unbroken sequence to the 

 foot-hills of the Carpathians. On the virgin soil not occupied by 

 the agriculturist, the abundance of Euphorbiaceous plants was 

 especially noteworthy. 



Having thus briefly dealt with the N.W. and W. parts of the 



