234 MR. C. DATISOX OX THE 



resistance to a motion so minute ; and, secondly, because, on account 

 of their greater thickness, the changes of temperature might not 

 sufficient!}^ penetrate the stones in the short periods during which 

 such changes sometimes take place. The experiments were of two 

 kinds — the first qualitative, the second quantitative. I will now 

 give an example of each. 



Eccjoeriment 1. — The object of this experiment was to discover 

 if au}^ downward motion took place in a stone lying upon an in- 

 clined plane, and, if so, whether the motion were to be attributed to 

 the alternate expansions and contractions of the stone during changes 

 of temperature*. 



A brick (B) was fixed with its upper surface inclined at an angle 

 of 20° towards the south-west. On this was laid another brick (A) 

 whose movements, if any, were to be observed by means of a level, 

 resting at one end on the upper edge of the brick A, and at the other 

 on a support made of similar brick. If the brick A did not move 

 relatively to B, the vertical components of the expansion or con- 

 traction of the supports at either end of the level would be the 

 same, and the level would indicate no change of position. If, how- 

 ever, the brick A did move in the way anticipated, the nature and 

 manner of the movement should be in conformity with the theory 

 given by Moseley. 



The level was read frequently during the day, generally every 

 half- hour. The temperature was also observed at the same times. 

 As the experiment was intended to show the nature only of the 

 movement, I did not attempt to determine the value in angular 

 measurement of each division of the level. The following conclu- 

 sions were brought out very clearly by this experiment : — 



1. The upper end of the brick rises slightly with a rise of tempe- 

 rature, but descends beyond its first position with a corresponding 

 fall of temperature. 



2. The movements accompany, or take place a short time after, 

 the changes of temperature. 



3. The total downward movement is least on those days on which 

 the sky is clouded and the range of temperature small. 



A similar experiment might have been made to determine the 

 motion of the lower end of the brick, by resting the level on a 

 projection near the end and on a brick support near it. But the 

 above experiment, several times repeated, seemed to me to show 

 cqnclusively that stones resting freely on an inclined plane will 

 gradually creep down the plane owing to the expansions and con- 

 tractions of the stones, due to changes of temperature, taking place 

 mainly downwards under the influence of gravity. 



* [Since this paper was read, it has been pointed out to me that "in Canon 

 Moseley's experiments the lead which moved had an expansibility dif- 

 ferent from that of the wood on which it rested," and that if the two bricks in 

 Experiment 1 " had expanded equally no motion could have resulted." The 

 movement in this, and the following, experiment must therefore be attributed 

 to the unequal heating of the upper and lower stones. The distinction is im- 

 portant, though it does not affect the subsequent results. — Note added March 24, 

 1888.] 



