Dana and Chambers on Ancient Sea-Margins. 215 



alluvium might be carried off; where broad, some portions 

 would be left. 



4. The result would be the same, whether the rise were 

 gradual or abrupt. If the latter, the river would have more 

 rending power towards its mouth, and the deepening would 

 go on more rapidly. If a slow, gradual rise should com- 

 mence at any time, the river, through its increasing excava- 

 ting power, would begin to sink between its banks, and the 

 wear of the alluvial flat on either side by floods would also 

 commence. The terrace slope would also shew its first begin- 

 nings, as an outline to the river flats. Finally, the stream 

 would have a bank of its former height, — this height being a 

 constant quantity for the stream. It would have more or less 

 broad flats, which flats, or bottom lands, would commonly be 

 bounded by a slope ; and, if the former alluvium remained, 

 there would be a shelf or terrace above. 



5. From the conditions mentioned in the last paragraph, 

 it is evident that, during a gradual elevation (and as gra- 

 dual a sinking of the stream), the outline of the lower flat 

 might be varied, in consequence of a change in the bed or 

 banks of the stream, which should vary its direction, or the 

 direction of its principal current ; and that, therefore, in the 

 course of one and the same rising, a terrace of sixty feet 

 might form in one place, and in another, perhaps not far dis- 

 tant, one of twenty and another of forty ; or one of ten, and 

 another of thirty, and another of twenty, and so on. In still 

 other places, the upper alluvial plain might descend by a 

 very gradual inclination to the lower flat ; instead of forming 

 a proper terrace, or part-way, it might be gradually declin- 

 ing, and then fall off^ with the rapid slope that usually bounds 

 a terrace. These are all possible results of the causes men- 

 tioned, and are to be looked for in nature. 



6. The lakes of the country, where they could at once 

 empty themselves, would correspond with the rivers in the 

 results produced, and each terrace might indicate, in many 

 cases, a separate elevation. When they remained closed by 

 a barrier, they would either gradually become emptied, or by 

 abrupt steps at intervals ; and hence there might be several 



