FLOW OF THE RIVRB TEV1OT. 187 



their vent on the very verge, or in the bed of the stream.* In severe 

 drought or in sharp frost, the open springs at any distance from the low- 

 est level never reach it, and the supply of the stream is altogether kept 

 up by the water rising within its own bed. In a lake, where there is no 

 outlet, the common laws of heat and cold influence its freezing, and deep 

 water never congeals until the whole mass is brought to the freezing 

 point, In our most northern settlements, Hudson's Bay for instance, 

 stagnant water is known to freeze to the depth of ten or twelve feet, and 

 all the fish are frozen along with it : not so in the rivers the surface is 

 frozen indeed, but the ice then becomes a protection to the running 

 water beneath, from the greater cold of the atmosphere. It is quite evi- 

 dent that in all running streams, the falling down of the colder, and the 

 rising of the warmer water cannot take place with any uniformity, for 

 the movements of the current prevent the regular action of heat and 

 cold, and the discharge of the springs into the channel of the river from 

 a greater depth than the frost penetrates, keeps the water still moving 

 freely below the surface of the ice. 



To explain the singular phenomenon of the stoppage of mountain 

 streams, it is necessary to consider how they usually flow. Where rocks 

 abound and the channel is alluvial gravel, every one, conversant with 

 Highland scenery, is aware how the rivulet meanders almost in circles, 

 forms a succession of streams and pools, and at every turn creates a na- 

 tural weir, extending from side to side at right angles. The pools 

 discharge themselves at their lower extremity, over these natural wears, 

 which again form the heads of the streams below, and the streams again 

 terminate in pools, and so on in succession, till the enlargement of the 

 stream renders the natural weir less observable. 



Let the wind come from any point of the compass, it must blow 

 directly in the face of a great number of these banks of sand and gravel, 

 will stop the descent of floating ice, and materially assist in freezing it 

 into a more solid mass. It will make the stoppage more easy in the na- 

 tural weirs below, by diminishing the current of water until it ceases to 

 flow altogether. To produce this, it is possible that frost alone may be 

 sufficient, but to do it more effectually, wind in addition would appear to 

 be requisite. Now this actually was the case during the night of Novem- 

 ber 26-7. The phenomenon which occurred at Ormiston cauld shews, in 

 a remarkable manner, how far the effect may be carried ; but the stoppage 

 of the rivers must be sought for in the natural weirs, where the frost, fix- 

 ing on the shallow bottom, forms a ground-ice, and then catching the 

 light floating grew, makes a chance of obstruction, while the stream being 

 stopped, there is time for the ice to form a covering on the surface of the 

 pools. After a time, which will be longer or shorter according to circum- 

 stances, the springs will force those barriers, and the flowing water will 



" The springs being universally directed towards the latter. 



