1907.] 



on Seiches in tlie Lakes of Scotland. 



660 



The calculations for Earu were carried out by determining a 

 number of points on its normal curve * from the data of the Scottish 

 Lake Survey, and then fitting two parabolse to these points as nearly 





I 



,riJVtA/vi«A''^'~''^'>'^^\WW^^ 



|rv^AAa'WMyV\MA A A/WWIA 



5 



\'ii^ 





i,V\W^tvvWVVwWW/vy 



!• -? ?■! — r-^-j T — ^ r ?=r-T.= — r- 



FlG. 



as possible by the method of least squares. The fit is by no means 

 perfect, as you see by the diagram now shown ; yet the agreement 

 of the calculated with the observed periods is wonderfully good, as is 

 shown by the table on next page. 



Fig. 8. 



"We have not yet completed the determination of the nodes from 

 the Hmnographic observations, but the next slide (see Fig. 9) will show 

 you that our theoretical determinations must be faii'ly correct. The 

 upper limnogram shows a dicrote seiche at the eastern end of the lake ; 



* If the lake were of uniform breadth, the normal curve would be the 

 curve of longitudinal section. 



