express our impressions, so to say, in terms of years, we are baffled by the complexity of the problem, and can but confess our ignorance. Occasionally indeed we obtain a faint glimmer of light, but the result is only to show us obscurely a long vista, without enabling us to define any well-marked points of time. Thus in Denmark we found three periods of arborescent vegetation, corresponding to the three epochs of human development, and we know that the extermination of one species of forest tree and its replacement by another is not the work of a day. The Swiss archæologists, however, have attempted to make an estimate somewhat more definite than this.
The torrent of the Tinière[1] at the point where it falls into the Lake of Geneva, near Villeneuve, has gradually built up a cone of gravel and alluvium. In the formation of the railway this cone has been bisected for a length of one thousand feet, and to a depth in the central part, of about thirty-two feet six inches above the level of the rails. The section of the cone thus obtained shows a very regular structure, which proves that its formation was gradual. It is composed of the same materials (sand, gravel, and larger blocks) as are even now brought down by the stream. The detritus does indeed differ slightly from year to year, but in the long run the differences compensate for one another, so that when considering long periods and the structure of the whole mass, the influences of these temporary variations, which arise from meteorological causes, altogether disappear, and need not therefore be taken into account. Documents preserved in the archives of Villeneuve show that in the year 1710 the stream was dammed up and its course a little altered, which makes the present cone slightly irregular. That the change was not of any great antiquity is also shown by the fact that on the side where the cone was protected by the dykes, the vegetable soil, where it has been affected by cultivation, does not exceed two to three inches in thickness. On this side, thus protected by the dykes, the railway cutting has exposed three layers of vegetable soil, each of which must, at one time, have formed the surface of the cone. They are regularly intercalated among the gravel, and exactly parallel to one another, as well as to the present surface of the cone, which itself follows a very regular curve. The first of these ancient surfaces was followed on the south side of the cone, over a surface of 15,000 square feet; it had a thickness of four to six inches, and occurred at a depth of about four feet (1.14 metre measured to the base of the layer) below the present surface of the cone. This layer belonged to the Roman period, and contained Roman tiles, and also a coin.
The second layer was followed over a surface of 25,000 square feet; it was six inches in thickness and lay at a depth of 10 feet (2.97 metres, also measured to the bottom or the layer). In it have been found several fragments of unvarnished pottery, and a pair of tweezers in bronze, which to judge from the style belonged to the
- ↑ See Morlot, Leçon d'Ouverture, &c.
