PRATT ON SHELL-BEDS. 1 59 



attain such dimensions, and consequently the spot in such cases must 

 have been occupied for a considerable period. In many instances there 

 is higher ground near at hand, places much drier and more suitable for 

 huts or camping grounds, and where that was not the case, we know that 

 the earlier races not unfrequently raised " tenement floors" of earth to 

 secure those advantages, but these shell beds are never found upon such 

 elevations, either natural or artificial, but always on the low flats, close 

 by the water. 



Second. The great scarcity of relics of human handiwork, as there 

 are certainly not more than might w^ell be expected to be accidentally 

 dropped and lost in the streams and along the banks, w^hile the undoubted 

 " kitchen heaps" invariably contain an abundance of such remains, and 

 the fragments of them. Also, the absence of a dirty and trodden floor, 

 and of any dirt mixed with the shells, as the soil which fills the spaces 

 between the shells, and which not unfrequently constitutes by far the 

 greater portion of the whole mass— for the shells are in many cases 

 very few and far between— is exactly the same in character as that imme- 

 diately surrounding the place, and fragments of limestone which often 

 abound there are sharp, angular and unworn, like those still lying on the 

 slope below. 



Third. The species of shells are about in the same proportion as might 

 be expected in a natural deposit, while, if the moUusks were used as food, 

 there would surely be evidences of selection. Even in the heaps of shells 

 collected by musk-rats, we find them in very different proportions, the 

 hardest and toughest classes are scarcely at all collected by the rats. The 

 Uniones, rectus, gibbosus, verrucosus, tuberculatus, plicutus, and ligamen- 

 tinus are scarcely ever to be found there, nor are the aquatic univalves, 

 nor the land snails, while in our shell beds all of these are present in full 

 proportion. If, then, these were the remains of the daily meals of our 

 human predecessors, they must have selected with less taste and less intel- 

 ligently than the musk-rats of later days. In hundreds of cases, also, the 

 shells are found in pairs, closed, having apparently never been opened.* 



Fourth. The almost total absence of remains of all kinds of food. 

 Surely, we cannot suppose that a people subsisted entirely upon clams ! 

 But, though close by the water, there are almost no fishbones, and appar- 



*1 would refer, also, in this connection, to the shell heaps, as they are called, at the edge of the 

 high sand bank by the river, about a mile below New Boston, Ills. They are situated on the 

 very highest part of the sand ridge, which is many feet higher than the prairie surface a 

 short distanee back from the river, and have the appearance of a row of large heaps of shells. 

 Upon examination, however, I found that they are but the remaining portions of a horizontal or 

 somewhat undulating shell-bed, fitteen or twenty centimeters in thickness As the sand washes 

 away and the shells fall down the slope on all sides, it gives the appearance of a shell-mound. 

 All the shells, however, have fallen from the top, and are only on the surface of the 

 sand. The layer remains at the top, a level surface of a meter or two in extent, and growing nar- 

 rower every year, and of the depth above named, and no shells nor fragments are to be found in the 

 sand immediately beneath it. The species found here exhibit the same indications of selection 

 as those of the musk-rat heaps before mentioned, the same species being absent in both cases. 



Relics of flint and pottery are also numerous among these shells, as should be expected in a 

 refuse heap, and the sand beneath the shells is much discolored, black and dirty. 



