124 H. F. CI eland— Formation of Natural Bridges. 



the bridge has been able to excavate, owing to a former water- 

 fall and the peculiar platy structure of the rhyolite, in which 

 curved layers of extremely different physical texture and fria- 

 bility offered a favorable site for attack by frost and water. 



The formation of lava bridges is usually explained as follows : 

 The surface of a lava flow cools and hardens while the interior 

 is still in a molten condition. As a result of this condition, if 

 the molten rock beneath continues to flow, a tunnel will result. 

 Such tunnels are of common occurrence on Mt. Vesuvius, the 

 volcanoes of the western states and in other volcanic regions. 

 From such a tunnel a bridge might be formed by the caving in 

 of the greater part of the roof. A study of the photograph 

 (tig. 3) showing the structure of the lava of which the Yellow- 

 stone Natural Bridge is formed shows that such an explanation 

 is untenable in this case at least, the rock being composed of 

 approximately vertical plates of lava of different degrees of 

 compactness. The writer has not made a study of other lava 

 bridges, but it seems probable that the mode of formation of 

 the Yellowstone bridge may be exceptional for bridges of this 

 character. 



In each of the cases cited the top of the bridge was formerly 

 a portion of the bed of the stream. If natural bridges were 

 formed as commonly supposed, it would be unusual to find that 

 a surface stream had once been superimposed upon the cavern 

 for its entire length. There is, for example, seldom any rela- 

 tion between the surface topography of a country and the 

 underground passages of extensive caves. 



Occasionally a small natural bridge occurs near the opening 

 of a cavern or where a spring flows from beneath a cliff. Such 

 a bridge is the sandstone arch spanning a spring which emerges 

 from beneath the sandstone capping of Lookout Mountain near 

 Chattanooga, Tenn. The bridge is formed by the widening of 

 a transverse joint, first by weathering alone and later by the 

 combined action of weathering and erosion, thus separating the 

 bridge from the cliff. The breadth of the span was increased 

 largely by weathering. 



The conclusion to which one is led by this study of natural 

 bridges from different parts of the United States and composed 

 of various kinds of rocks — marble, limestone, sandstone, and 

 lava — is that, although bridges may be formed, and undoubtedly 

 have occasionally been formed, by the partial falling in of the 

 roof of a long underground tunnel, the usual mode of forma- 

 tion is that described above. It should, however, be said that 

 examples exist concerning which it is difficult to say which 

 mode of formation was the more prominent. 



Williams College. 



