438 SCIENTIFIC RECORD FOR 1884. 



limb of wbich is about S*"" in diameter and somewhat longer tluiu the 

 other, which is muclS larger, and closed at the top by means of a movable 

 piece of tube of the same size, carrying a stoi)-cock and attached by a 

 rubber band. The'small tube is graduated into cubic centimeters above 

 a point on the same level as one of several lines drawn on the larger tube. 

 To make an observation the tube is filled up with water to one of these 

 lines and the substance whose specific gravity is required is dropped in 

 the larger limb. Of course if it sinks it displaces its own volume of 

 water, and by blowing through the stop-cock the level of the water is 

 depressed below the original mark. It is then allowed to rise again 

 slowly until the surface reaches the line at which it stood before the 

 solid was introduced. By reading the bulk of the water displaced in 

 the smaller tube, which may be done with much accuracy, and by di- 

 viding the weight of the body previously ascertained by this number, 

 the specific gravity is obtained. {Phil. Alag., V, xvii, 459, June, 1884.) 



Belli has observed that if two glass plates be made to adhere by ap- 

 plying them exactly to one another, and are then left so that gravity 

 acts upon them, they always separate, no matter how small may be the 

 weight of the one united to the other by adhesion. Obviously molecular 

 attraction cannot be the cause of this phenomenon. The long duration 

 of the adhesion arises simply from the slowness with which air pene- 

 trates between the plates. Stefan has come to the same conclusion 

 from a mathematical investigation. {II Nuovo Gimento^ xiii, 34: J. 

 Phys., December, 1884, II, iii, 552.) 



Turpin and Warrington have repeated Bottomley's experiment of cut- 

 ting through a block of ice by means of a wire having weights attached 

 to its ends, with a view of noting the effect of varying conditions. Bot- 

 tomley had observed that string would not cut through the ice, and ex- 

 plained the fact by supposing that the string is not a sufficiently good 

 conductor to relieve itself of the cold in front and pass it back to the 

 water behind. To test the effect of conductivity, wires of silver, copper, 

 brass, iron, and German silver, 0.67™™ diameter, were employed, each 

 carrying 2.5 pounds. The results show that the time increases as the 

 conductivity decreases. In a second experiment the wire was of iron, 

 and was loaded in the first case with 5 pounds, in the second with 7^, 

 and in the third with 12^ pounds. It was found that the time taken 

 was approximately inversely proportional to the load. Pfaff's experi- 

 ment of placing a tube upright on a block of ice, pressing it down by a 

 weighted lever, while maintaining it at 0° by surrounding it with snow, 

 was then repeated, using tubes of copper, brass, lead, and glass, each 

 about a foot long and of three-eighths inch bore, the weight being 

 2 pounds. In four hours the copper tube had penetrated 100""", the brass 

 tube 35"™, the lead tube 7""" and the glass tube 3"™; showing con- 

 clusively that when the temperature is not lower than 0° C. the chief 

 factor is the lowering of the freezing point by i^ressure, and not the plas- 

 ticity of ice. If the result takes place below 0°, as Pfaff states, the 



