NATURAL nilLOSOPIIT. 157 



acids, the ingredients may be regained by evaporation and re- 

 crystallization of the salts, and therefore they are much less 

 expensive than the solutions in acids. 



FORMATION OF ICE UNDER WATER. 



In the 16th annual report of the Detroit Ijoard of Water Com- 

 missoners, for 1867, are detailed the efforts of the comnussioners 

 in devising some way of preventing the ice from chokuig up tlie 

 main inlet pipe. Tliis pipe .extends 150 feet into the river, and 

 terminates in a bell-shaped mouth elbow, 3 feet in diameter, 

 turned upward, in water 25 feet deep. Covering the end of the 

 pipe is a boiler-plate strainer, perforated with half-inch holes, 14'J: 

 to the square foot. Inside the shell of the strainer is a diaphragm 

 plate with similar holes, and below this the strainer shell has 

 4-inch holes, to allow the sand to pass through, so as not to bank 

 upon the outside of the strainer. When the engine is pumping, 

 the water is required to pass through the strainer holes at the rate 

 of 120 barrels per minute. This is the full supply, but in extreme 

 cold Aveather, under certain circumstances, it is with great diffi- 

 culty any water can be obtained, in consequence of the accumu- 

 lation of ice. The circumstances under which the difficulty oc- 

 curs are, when the weather is cold and ice is forming in the lake 

 above, and on the shores of the river, and the river is free from 

 ice over the strainer. But when the river is covered with ice over 

 the strainer, the difficulty does not occur at any degree of cold. 

 The great difficulty occurs when the thermometer ranges from T^ 

 or 8*^ to 18*^ or 20°. above zero ; but when the mercury rises above 

 20" the difficulty soon ceases. The greatest number of deten- 

 tions, it has been observed, occurs at night, and when the sun is 

 obscured by clouds, but, when the sun is unclouded, no difficulty 

 is ever experienced. 



With the rapidly increasing consumption of water, the commis- 

 sioners foresaw that the time would very soon arrive when it 

 would not be safe to permit any detention to the pumping engines, 

 and that this remarkable phenomenon must be solved and the 

 difficulty overcome. 



As no experiments had ever been previously made, and the 

 theory was so strongly presented that the trouble was wholly from 

 anchor-ice forming on the strainer, an opening was cut through 

 the down-stream side of the strainer, and a self-acting door was 

 hung; but this and the plan of suspending a line of booms so as 

 to retain a coverino^ of ice over it when the rest of the river was 

 not covered, both failed to accomplish the object sought. The 

 theory that the coverino: of the entire surface of the river bv ice 

 prevented radiation, and by that means the ice did not form on 

 the strainer, Avas strongly urged; but, if so, any cov^ering over 

 the strainer would answer the same purpose. To test it, last 

 summer submarine divers built a submerged platform of planks 

 immediately over the strainer ; but this proved of no avail, for the 

 stoppages occurred at a higher temperature than before. 

 14 



