248 PHYSICS. 



graph cousisted of a revolving cylinder of silver covered with lampblack, 

 connected with one terminal of the secondary coil. The other terminal 

 was a fine wire brought very near the surface of the cylinder. When by 

 the motion of the membranes the primary circuit was broken, a spark 

 passed to the cylinder and made a fine dot in the lampblack. By refer- 

 ence to a sinuous curve of times traced simultaneously on the cylinder 

 by an electrically vibrated fork, the time could be measured to .00001 

 second. When the air was quiet the gun was fired and the arrival of 

 the pulse at each successive membrane was recorded. By measuring 

 the distances the velocities are readily calculated. It was found that 

 immediately in the rear of the cannon the velocity was less than at a 

 distance, but that going farther and farther away from the cannon the 

 velocity oi sound rose to a maximum considerably above the ordinary 

 velocity and then fell gradually to about the velocity usually received. 

 Hence the author concludes : 1st, that the velocity of sound is a function 

 of its intensity ; and 2d, that experiments uj^on the velocity of sound in 

 which a cannon is used contain an error i^robably due to the bodily 

 motion of the air near the cannon. For a correct determination of the 

 velocity of sound, therefore, he thinks that a musical sound of low in- 

 tensity must be used. — {Am. J. Sci., Ill, xvii, 116, February, 1879.) 

 Ihlseng under tlie direction of Rood has measured the velocity of 

 sound in various kinds of wood. In his first experiments he used the 

 method of Kundt, vibrating a rod of the given wood, 1 to 2 meters long, 

 longitudinally, while one end of it, armed with a cork, entered the end 

 of a glass tube a meter in length, and containing a fine powder. To the 

 other end of the tube a i)iston was fitted so as to make the length used 

 a multiple of the half wave length in air. In this way rods of cedar, 

 white- wood, yellow and white pine, hickory, M^hite ash, holly, mahogany, 

 black walnut, cherry, and white oak were experimented with. In a sec- 

 ond series of experiments a method was adopted by which the rod reg- 

 istered its own vibrations on a smoked-glass plate drawn rapidly across 

 the vibrating end of the rod in a horizontal direction by means of a 

 weight ; while at the same time a tuning-fork registered the time. The 

 conclusions given are : 1st, it is possible to measure the velocity of sound 

 in rods with considerable accuracy by the graphic method ; 2d, Kundt's 

 method gives results which are slightly higher than those by the graphic 

 method; and 3d, the graphic method demonstrates the existence of 

 transverse along with longitudinal vibration and gives their ratios. — 

 {Am. J. ScL, III, xvii, 125, February, 1879.) 



Kohlrausch has given a method of studying the condition of the air 

 in a vibrating column, and of repeating the experiments of Savart and 

 Seebeck upon the position of the fixed nodes and loops x>roduced by the 

 combination of the direct and the reflected sonorous waves. For this 

 purpose a caoutchouc membrane, stretched upon the extremity of an 

 open tube, is lowered into the organ-pipe which is sounding, by means 

 of a string passing through the cork closing the pipe. In this case the 



