SZA.B<5. SOUTH-EAST OT EUROPE. 5 



cation may be followed uninterruptedly for several miles. The 

 Nummulites at one locality, not far from Gran, form a layer nearly 

 1 foot thick ; at another place the true Loess-shells are mixed with 

 Melanojpsis, Cerithium, and other fossils of the Congeria-beds. The 

 Loess strata are nearly horizontal. 



According to the evidence of the organic remains, the Loess may 

 be divided into two portions. The great Mammalia occur in the 

 lower, while the Loess-shells are only exceptionally found in that 

 portion. The upper division abounds with Loess-shells, but the 

 remains of Pacliydermata are absent, and even the Cervas, Bos, and 

 Equus are only occasionally found. This distinction refers only to 

 districts where the formation is well developed, the thickness being 

 not less than 100 feet ; but in other cases, both portions sometimes 

 appear to be confusedly brought together. 



The Surface in the Pleistocene Period. — As regards the deep channels 

 of the rivers, it is quite certain that where the banks are more or less 

 vertical, and the beds on each side correspond, the place occupied 

 now by the river was originally filled up by a continuation of the 

 same strata which now constitute the banks, and that the uninter- 

 rupted plain thus formed was furrowed posteriorly. The restoration 

 must be effected on a much greater scale where two or more great 

 rivers joined. An example of this kind occurs at the southern end 

 of the boundary-line between the Danube and the Theiss, where, in 

 the midst of an extensive area formed by alluvium (of both ages), a 

 small egg-shaped plateau, consisting of typical Loess, and having 

 vertically broken cliff-like walls, remains like a column of earth to 

 mark the extent of former denudation. 



Having thus restored to one continued plain the Pleistocene 

 deposits in Hungary and in the Danubian Principalities, the author 

 proceeds to consider the similar phenomena in the neighbourhood of 

 the Black Sea ; and he states that the Loess does not thin out towards 

 the sea, but, on the contrary, becomes more largely developed, and 

 that the shore is formed by abruptly broken cliffs. 



The method of restoration hitherto pursued by Dr. Szabo reclaims 

 from the sea a large portion of the areas now occupied by it, and 

 proves that, at some period after the deposition of the Loess, that 

 part of the Euxine which is necessary for the restoration of the Loess- 

 plateau was not covered by water, but was so much elevated that 

 the Loess-strata now wanting touched each of the corresponding ones 

 which now form the shore. Now, if we suppose that water has 

 covered the whole area over which, at the same low level, the Loess 

 would be found as an uninterrupted mass, we obtain a lake, the 

 water of which covers, besides the Pontic lowland, probably the 

 whole Arabio-Caspian depression, as suggested by Pallas, Humboldt, 

 and all the naturalists who have travelled there. 



But such a sea was possible, in the Pleistocene period, only upon 

 the supposition that the present Black Sea did not then exist as such, 

 but that its present bottom was elevated to such an extent that the 

 Loess plateau could exist in its original completeness. A natural 

 consequence of this condition is, that it was not in connexion with 



