Joty—On the Conservation of Mass. 51 
The foregoing fourteen experiments reveal certain effects on the motion of the 
beam apparently traceable to some force developed under the conditions of 
experiment. If a force isto be ascribed in any way to the motion of the Earth 
through a medium, it must exhibit a change of direction as regards east and west 
from midnight to midday. There is some indication that the noon-day experi- 
ments are attended with a more decisively westerly perturbation or acceleration 
than the midnight observations. ‘The fourteen experiments may be analysed as 
follows, using the sign ?? to indicate a quite uncertain result, and the sign ? 
following the compass-direction to indicate considerable doubt as to the inter- 
pretation placed on the curve. 
(1) Midnight: easterly (?). 
(2) Noon: easterly (?). 
(3) Noon: westerly. 
(4) Noon: (??). 
(5) Noon: westerly. 
(6) Noon: westerly. 
(7) Noon: westerly (?). 
(8) Noon: westerly. 
(9) Midnight : 
(10) Midnight : 
(11) Noon: 
(12) Midnight: 
(13) Noon: 
(14) Noon: 
westerly (?). 
easterly (?). 
westerly. 
easterly. 
westerly. 
westerly. 
There are thus seven results where the perturbation is most probably ascribed 
to a force directed westward; and these are all noon-day experiments. None of 
the midnight results is westerly with the same degree of probability. There are, 
again, no good easterly perturbations obtained at midday ; while there is one very 
good one obtained at midnight. With reference to error from passage of heat 
through the cork-jacket, it is to be noticed that in all cases the final result of the 
experiment is indicated clearly in the second or early in the third minute from the 
moment of contact. It seems safe to conclude that an appreciable acceleration 
of the load could not arise from this source so soon after the mixing of the 
substances. The experiments, therefore, point on the whole to a positive result. 
Further experiments are desirable, and more especially observations made 
at midnight. I do not think that with the present apparatus any final conclusion 
can be arrived at beyond that of showing that the force to be expected—if 
existent at all—is of very small value. It is evident that where such uncertainty 
as to the form of the vibration and such variation in acceleration enter the 
experiments, the changes of velocity noticed must be liable in many cases to 
misinterpretation. The conclusion may be drawn, however, that no inertia effect 
corresponding to the least weight-change measurable by the most sensitive balance arises in 
K 
TRANS. ROY. DUB. SO0., N.S., VOL. VIII., PART II. 
