310 STUDY OF HIGH ANTIQUITY. 



preserving regularly the same inclination. In ordinary times the torrent flows 

 along the central region of its cone ; there also it drops its largest boulders in 

 times of inundation, distributing the gravel and less coarse shingle over the 

 sides of the cone, for the relative volume of the drifted stones must diminish 

 with the force of propulsion of the water from the central region of the cone 

 towards its two edges. It is clear that a torrent, left to itself, cannot raise the 

 surface of its cone unevenly and create hollows and prominences. For if the 

 surface were raised on any one spot, the water would immediately flow round 

 and lill up the less elevated parts with drift. The action of water is essentially 

 levelling. The great number of torrential cones, or deltas, which the author 

 has had the opportunity, during the last 15 years, of examining in Switzerland 

 and in the Austrian Alps, have ahvays showed a regular surface. There may 

 be slight irregularities in the action of a torrent from one year to another, but 

 proceeding chiefly from meteorological variations, they become insensible, when 

 the cone is considered in its totality ; even at a given spot they will rapidly 

 become levelled and effaced by the continued action of the torrent itself. We 

 must also consider that the alluvium of a torrent is furnished by a slow degra- 

 dation of its hydrographical basin, which can only yield the drift matter grad- 

 ually, a circumstance necessarily contributing to regulate the growth of the 

 cone. Thus, when in July, 1S48, the torrents in Corinthia (eastern Alps) 

 were swollen and brought down a disastrous quantity of drift, the author heard 

 the country people ascribe the damage, in good part, to the circumstance that 

 the upper region of the water-courses had been more than usually encumbered 

 by loose matter. 



The torrent of the Timbre, where it flows into the lake of Geneva, at 

 Villeneuve, (Switzerland,) forms a cone or delta, such as we have just described. 

 This cone has an inclination of 4 degrees, and its opening, or the angle at the 

 toj) of the delta, measures about 700 degrees ; its radius, taken as a minimum, 

 being 900 Swiss feet, (one Swiss foot equal to 0.3 metres, is divided into 10 

 inches.) 



Modern embankments, in the shape of solid walls, having forced the torrent 

 somewhat towards its right bank, on the northern side of the cone, the alluvium 

 has since been accumulating more on that side, and has raised the surface here, 

 whilst the southernmost part of the cone ceased to increase. Documents pre- 

 served in the parish archives of Villeneuve mention these embankments as 

 having been built in the year 1710, and their recent origin is confirmed by the 

 scanty covering of vegetable mould on that part of the cone which was protected 

 by them. Here, where the ground had not been cultivated, there was only 

 from 2 to 3 inches (6 to 9 centimetres) of earth, inclusive of the space occupied 

 by the roots of the grass. The railroad has cut transversely through this cone 

 perpendicularly to its axis, the cutting measuring 1,000 feet in length, and 

 reaching in its central part, Avliere the cone is highest, to 32.^- feet above the 

 definitive lev^el of the rails. The section thus obtained (see page 316) may be rep- 

 resented by the segment of a circle, rising to 32.^ feet above its chord of 1,000 feet. 

 Happily for science, the works for the railway have been carried on very 

 slowly at this spot; they were begun in 1856, and are not entirely finished 

 now, (March, 1863.) The author followed them attentively from the beginning. 



The cone's interior structure, brought to light by this beautiful artificial sec- 

 tion, was found to be most regular. In the central region the rounded boulders 

 attained a diameter of 3 feet, as in the actual bed of the torrent. From thence 

 the drifted matter gradually diminished in size along the two halves of the cone 

 towards the two extremities of the cutting. There was an exception for the 

 alluvium formed since the embankments of 1710, for here the drift was naturally 

 coarser than in the underlying part. The waters of a torrent are not apt to 

 produce a marked stratification, of which but slight traces were to be seen, and 



