20 REPORT — 1864. 



the species of natural disturbances of the earth's crust. And if the same law of 

 increase continues up to very high temperatvires, steam would be powerful enough 

 to cause earthquakes and volcanos. The heat of a common fire is known to be 

 1141° F., dull red heat 980° (difference 161°) ; and he thought boiler explosions 

 were often caused by stirring the fire rapidly, and changing the latter temperatm-e 

 into the former, which would greatly increase the pressure. 



On the Properties of certain Stream-lines. 

 By Professor W. J. Macqxtorn Rankine, LL.D., F.R.8. 

 This paper was a summary of an investigation in continuation of one of which an 

 abstract was read to the British Association in 1863, and which has been published 

 in fuU in the 'Philosophical Transactions.' The new investigation consists of 

 three parts. The first part relates to certain exponential stream-lines, suitable for 

 the "buttock-lines" of ships, and resembling the lines introduced by Mr. Scott 

 Russell for that purpose. It also shows that, by the action of certain pressures on 

 the surfaces of water, waves may travel, which begin to break when the two slopes 

 of their crests meet at right angles*. The second part relates to Lissoneo'ids, that 

 is, to those forms of stream-lines which are the fullest, consistently with not pro- 

 ducing unnecessary disturbance in the water: it solves the problem in three 

 dimensions, which in the previous paper had been solved in two. The third part 

 relates to stream-line surfaces of revolution. (See Phil. Mag. October 1864: and 

 January 1865.) 



On a Mode of Determining the Velocity of Sound. By Dr. J. Stevellt. 



Suppose a piece of clock-work prepared, for instance, to strike single strokes 

 upon a beU each time the detent is set fi-ee ; the detent to be under the control of 

 an electro-magnet, which is instantly set in action by an observer, at a measured 

 distance from the bell or other orio-in of sound, depressing a key, and thus com- 

 pleting a galvanic circuit. The oDserver, being furnished with a chronometer, 

 depresses the key ; the instant he hears the stroke of the bell he again depresses 

 it ; hears a second sound, and so goes on for 100 or 1000 times, carefully noting 

 by the chronometer the instant at which he hears the last sound of the series. A 

 trained observer would not make a probable error of one-tenth of a second in 

 noting the whole time occupied by the whole series ; and to avoid all chance of 

 miscounting the number of soimds in the series, the clock may be readily made to 

 keep count of the number of strokes it makes. The whole time occupied by the 

 entn-e series is made up of the following portions : — 1st. The time consumed in the 

 mechanical work of the clock in producing the stroke, and of the key, from the 

 instant the observer touches it until it has completed the circuit. 2nd. The 

 personal equation of the obseiwer. 3rd. The time the soimd takes to travel 100 

 (or 1000) times the measm'ed distance of the origin of the soimd from the observer. 

 4th. The time the sound takes to travel 100 times (or 1000 times, as the case may 

 be) the measured distance. Now the first, second, and fourth of these portions 

 of time can be readily eliminated by repeating the same series of observations 

 exactly (the clock being wound up at the commencement of each series exactly to 

 the same extent) ; the observer, on the second occasion, placing himself at one- 

 half, or one-fourth, or at any determined part of his previous distance from the 

 origin of sound ; or by placing himself close up to it, using the same wires for the 

 galvanic cu'cuit On each occasion, in order to eliminate the fourth portion. The 

 author was not fully aware of the exact mechanism by which Professor Piazzi 

 Smyth discharges the cannons which he has introduced as time-signals, but he 

 had no doubt it could be adapted to this method, and thus determine experi- 

 mentally whether the velocity of sound is affected by the violence of its origina- 

 ting cause — a question which Mr. Earushaw has from theory decided in the 

 affirmative. It would, however, involve, the author supposed, the use of two 



* The author has since proved that the property of beginning to break when the two 

 slopes of the crest meet at right angles belongs to all waves in which " molecular rotation " 

 is nuU. 



