Theoretical Considerations. 183 



structure of the animal : for the dividing plane conforms in all cases 

 exactly to the layers of the clay. Since clay forms the principal min- 

 eralizer, the outer circumference of these cylindrical and ovoid relics, 

 is often very indefinite : and where exposed to rains, they are often 

 entirely washed away. Along the central part, however, hydrate of 

 iron seems to have considerably indurated the clay ; and, indeed, this 

 iron often pervades the whole mass ; giving it a rusty aspect, and 

 hardness sufficient to resist the action of rains. In such cases the 

 the relics often stand out in relief upon the surface of the clay, which 

 has worn away around them. 



These remains I have found only in clay or loam. I have met 

 with them in three places in Deerfield, viz. a few rods east of the 

 Academy ; near the top of Bar's Long Hill ; and on the east side of 

 North Sugar Loaf. They are found also in Greenfield, on Green 

 River, southwest of the village : also in South Hadley, at the Canal 

 Falls ; and 100 rods south of them. I have found them likewise in 

 Charlestown, in clay ; and also on Nantucket in a stratum of ferru- 

 ginous sand. Among the specimens deposited in the collection, (Nos. 

 47, to 52,) I apprehend several species- may be made out. 



Theoretical Considerations. 



That these tertiary beds were deposited from a mechanical suspen- 

 sion of their materials in water, can admit of no question. But was 

 it from salt water or fresh ? On this question the only organic relic 

 that has been found in these beds, throws no light ; because its nature 

 is not understood. Still, I think we have some grounds for forming 

 an opinion in this case. For the horizontal position of the laminae 

 of clay in this formation, shows that no change of consequence has 

 taken place in the level of the surface, since th,ey were deposited. 

 Hence, unless the level of the sea has sunk greatly since that period, 

 it could not then have c'overed the regions occupied by these beds. 

 They were probably, therefore, deposited from fresh water lakes and 

 ponds. In the valley of the Connecticut, that river must have flowed 

 through the lake ; or rather, an expansion of the stream formed the 

 lake. The gradual sinking of its bed at length brought the tertiary 

 strata sub luce. 



The probable absence of marine relics from these beds, is another 

 argument in favor of this hypothesis ; and it would be complete, were 

 the fact certainly established. 



In respect to the tertiary strata in Charlestown and Cambridge, 



