Let us, then, suppose there is some such agent filling the spaces between the atoms of the gas and think of what it must be. Were we to ask the question we have a strong suspicion the first impulse of many people would be to reply--With gas of course. But this reply could not satisfy us. We should immediately be led to think of that gas also consisting of atoms with vacant spaces between them filled with something--some more gas; and were we to follow up that thought through a sufficient number of stages, it is easy to see that in the end the whole space occupied by any gas would come to be filled up with its own solid atoms, without any empty spaces between them through which they could move; and so rendered quite incapable of pushing each other into a vacuum formed by any pump that might be applied to extract them from any vessel of any kind; or we must suppose that each particle would fly of its own good will into the vacuum made by the pump--as it were on the wings of the morning. But we recall to memory that the wings of the morning do not always carry us to rest, and we see that filling the spaces with gas would only end in choking up the vessel altogether. It might be said: Nobody imagines that the molecules of the gas in the spaces would be sufficient to fill them up altogether; and then we have only to ask, What then would there be in the spaces between the molecules of these successive gases to prevent the whole of them from gravitating to the bottom of the vessel? And to add that there would still be empty spaces left, absolutely empty, that would have nothing in them to help in any way to force the molecules or atoms of any gas or vapour into a vacuum anywhere. It is clear then that a gas, such as Dr. Crookes has described a gas to be, could only end in filling the spaces left between its molecules or atoms. It would be an obstruction to their collisions and bombardments which form an essential part of the description or definition. We must, therefore, have recourse to something else for filling up the spaces between the molecules of a gas, and the only thing we can lay hold of is our _limited liability_ agent the ether, which we allow to do all we want it to do and nothing more. Vapours of solid or liquid matter would be of no use, for they would only condense into solid or liquid matter; unless always maintained at their temperatures of evaporation or ebullition, and that would at the best be only another form of a gas--nobody would use a liquid to assist in pumping air out of a vessel--and, besides, we should still have to show what keeps their particles apart, what fills the spaces between them, which would force us to appeal to the ether as the only source, just as before. If there are no spaces between the particles there can be no vapours.