of this energy which we observe, and for the length of time during which it must have been given off according to the evidences of geology.
There is no chemical reaction which is not hastened or retarded by a change in temperature. In general, the velocity of a chemical reaction is increased by an elevation of the temperature and diminished by a reduction of the temperature. But radium compounds emit their rays undisturbed, at an even, unaltered rate, whether they be heated to a high temperature or cooled by immersion in liquid hydrogen and, what is perhaps equally striking, whether they are in the solid state or dissolved in some solvent.
In view of such facts as these, it is idle to suppose that radium is an unstable compound decomposing into its elements, using the terms compound and element in their usual sense. Conflict as it may with preconceived opinions, we seem forced to concede, not only that the transmutation of the elements is possible, but also that these transmutations are going on under our very eyes.
As has already been pointed out, this does not mean that we shall shortly be able to convert our elements into each other. Far from it, up to the present time we have not the slightest idea how to initiate such a process nor how to stop it. We can not, by any means known to us, even alter the rate at which it proceeds.
Now how shall we fit all these new facts and ideas in with our old ones regarding the elements and atoms, and how many of the old ideas must be discarded? Brief consideration is enough to convince us that very few of the old ideas, in fact none of value, need be sacrificed. We must indeed grant that Dalton's fundamental assumption is false, that the atom, in spite of its name, is divisible, and consequently that our elements are not our simplest substances, but decidedly complicated complexes. But all the facts included in the laws of definite and multiple proportions remain fixed and reliable, as indeed must all facts, expressions of actual experimental results, no matter what else varies. And there is not the least necessity for altering the methods of using atomic weights in calculations, nor for ceasing to use structural diagrams and models for molecules. We must merely modify our ideas and definition of an atom, and this modification is in the direction of an advance. We know more about an atom, or think we do.
Assume the inferences from the evidence just reviewed to be correct, and how do they affect our conception of the atom? First of all, our smallest, lightest, simplest atom, that of hydrogen, becomes an aggregation of about eight hundred smaller particles or corpuscles, and the atoms of other elements become aggregations of as many corpuscles as are obtained by multiplying the atomic weight of the element by eight hundred. Thus the atom of mercury must be thought of as containing 800 times 200, or 160,000, corpuscles. Next, the methods by which we believe we can calculate the approximate size of atoms and corpuscles
