nOUNDNESS OF THE SEA-BOTTOM. 23 



but more abrupt near the present shores of the con- 

 tinent, especially on the western side, which overlooks 

 the great ocean. 



The bed of the sea cannot, with strict accuracy, be 

 compared to the bed of a river. A section of the 

 Mississippi, at Plaquemines for example, resembles a 

 gutter. Neither does a lake of small extent present 

 the means of a satisfactory comparison, however deep 

 it may be. If we join the two opposite shores of a lake 

 by a straight line, that line will be above the bottom of 

 the lake, and will thus appear as a portion of the 

 surface. This is not true of a sea, if it be of any con- 

 siderable size. The earth fe roun4ed in form, the 

 free surface of the ocean is almost perfectly spherical; 

 and it is from that surface, as a starting-point, that 

 the depth of the sea must be estimated. Drive a 

 rectilinear tunnel through the earth from Paris to 

 Newfoundland, as sketched in the annexed diagram. 



Fig. 6. — Section of the Atlantic Ocean from Paris to .\n\ umiidlaiid. 



and it will be found that this tunnel nowhere encoun- 

 ters the ocean. It will, in fact, pass far beneath it. 



