Seiches in Some Lakes of Japan. 



13 



given lake and also to determine the periods and the positions of 

 the nodes, we constructed many models, filled them with water, 

 and set them in oscillation. Such experimental study is very 

 instructive, and as we learnt much from it, we shall now devote 

 some space to a description of it.* 



The models were made either of wood or of cement. The 

 scale for depth was always made greater than that for length, 

 otherwise the model lake would be too shallow and any motion 

 excited in it would Ije very much damped and soon die away. To 

 excite an oscillatory motion in the mass of water, it is necessar}- 

 to give it regularly timed impulses. A method for generating such 

 motion by means of a weighted spiral is described by ChrystaL 



Fiiï. 8. 



But we preferred the metliod of using a pendulum with a heavy 

 bob. The model was put on a wooden plate placed on a table 

 with two pieces of rubber tubing between the table and the plate 

 to serve as a sort of cushion. To one side of the plate, which 



* Honda, Teracla, Tosliicla and Ishitani. — Jonm. of Science College, Tokyo. Vol. XXIV. 



