68 Expedition to the 



consumed in efforts to get her afloat. A military post was 

 established at Bellefontain, under the direction of the gov- 

 ernment of the United States, by general Wilkinson, in 

 1803; but the soil on which his works were erected has dis- 

 appeared, the place being now occupied by the bed of the 

 river. A few fruit trees only, which stood in the end of his 

 garden, are yet standing, nbut are now on the brink of the 

 river. The first bank is here ten or twelve feet high, rising 

 perpendicularly from the water. Near its base are the 

 trunks of several trees with one end imbedded, and the other 

 projecting horizontally over the surface of the water, afford- 

 ing an evidence of the recent deposition of the soil of the 

 low plains, and an admonition of the uncertainty of tenure, 

 on the first bank of the river. One of these projecting trunks 

 is still in good preservation. It is about three feet in dia- 

 meter, and from its direction must pass immediately under 

 the roots of two large trees, now occupying the surface of 

 the soil.* Similar appearances are frequent along the Mis- 

 sissippi and Missouri, and furnish abundant evidence that 

 these rivers are constantly changing their bed, and, from the 

 great rapidity of the stream, as well as from the appearances 

 presented, we must suppose these changes are not very slow- 

 ly produced ; but their range is confined to the valley, with- 

 in the second banks, which are here raised about seventy feet. 

 On this second bank, in the rear of the site of the former 

 works, the buildings belonging to the present military esta- 

 blishment have been erected. They were commenced in 1810. 

 The houses are of one story, constructed of logs, based upon 

 masonry, and united in the form of a hollow square. At the 



* In a section of forty feet perpendicular, of the alluvion of the Mis- 

 sissippi near iNew Madrid, Mr. Shultz found seven hundred and ninety-eight 

 layers, indicating an equal number of inundations, in the time of their 

 deposition. Supposing these inundations to have happened yearly, we have 

 an easy method of forming an estimate, of the rapidity of the elevation of 

 the bed of the Missisippi. These layers were found to vary in thickness, 

 from one fourth of an inch to three inches. See Shultz's Travels, vol. 2. 

 p. 90. 



