particles must be smaller than any we can recognisefiner than the tiny constituent parts of wood-smoke which reflect the short-waved blue light to our eyes rather than the long-waved red, and therefore must be comparable in dimensions with the wave-length of light with which they have selective dealings; finer than the thinnest film of beaten gold leaf which maintains its colour. But how is it possible to get beneath, and find some distance of which we can say: this is certainly less than the average distance between the centres of the molecules ?

Among the debts which science owes to Lord Kelvin is the earliest solution of this difficult problem. Let us consider one of his methods. The soap-bubble is still the plaything of many a nursery and schoolroom, but the genius of Lord Kelvin has also given it an honoured place in the lecture-theatre and the laboratory. In the thin film of tepid soapy water expanding and contracting under the action of the human breath he recognised a means of attacking one of the deepest of molecular problems. In stretching a soap-film, we do work against the elastic force of the film which tends to contract it; and heat energy, another form of work, must also be supplied to keep the film from cooling as we expand it. But, however much work we do on the film in drawing it out and maintaining its temperature, the sum of our efforts cannot exceed the energy which would be needed to evaporate the film into steam, and thus to destroy all the cohesion between its molecules when they exist in the liquid state. If the film is to remain a film, less work than that needed to produce this change of state must be put into it. Now the energy needed to change water into steam is well known, since a simple measurement of the heat required gives it to us. Thus we can calculate the quantities concerned, and show that, if a film could be stretched until its thickness was reduced to about 10-8 centimetre, that is to about the one-hundred-millionth part of a centimetre, or the one two-hundred-and-fiftymillionth part of an inch, it would cease to have the properties of a film.

Here, then, we have our lower limit. Before we get to this excessive thinness, which would correspond to the beginning of a process of conversion into steam, the opposite sides of our film must be coming within

molecular range of each other, where a change of condition is possible and the character of the molecular forces begins to alter. We have thus found something which can be compared with, and shown to be smaller than, the dimensions of molecular structure. And other more modern estimates agree. From various sources of experimental knowledge the distance between the centres of the molecules of solids and liquids is found to be a little greater than 10 centimetre. Something less than one hundred million, or 108, molecules placed in a row would reach from one end of a centimetre to the other, while something less than two hundred and fifty millions would be needed to stretch over an inch. The actual size of each molecule cannot be estimated on these lineswe do not even know that the expression size of a molecule' has any meaning. All that is possible is to determine the number of molecules in a given space.

The corresponding figures for gases may be deduced from these results, but an independent investigation is possible. While we restrict ourselves to the consideration of pressure effects alone, the kinetic theory gives no indication of the number of molecules in a cubic centimetre. But when we pass to other properties, such as viscosity or conductivity for heat, our theory must take account of collisions between molecules, and then we can deduce an estimate of the average distance a molecule moves between two collisions, and of the distance over which one molecule can affect another. By such means it is calculated that about 2 x 1019 molecules exist in one cubic centimetre. The cube root of this number is about 3 x 10°, so that about three million gas molecules in a row correspond to one centimetre length, a number which agrees well with the estimate which we should obtain by comparing the value for a liquid or solid.

Here, then, was definite proof that the molecular structure of matter, which it was necessary to conceive in order to construct a consistent model of nature, was of far too fine a texture ever to be visible in our microscopes, or to affect any of the instruments then known to science. It seemed that the individual atom was for ever beyond our perception; that we could but deal with it ideally and conceptually as the indistinguishable unit in a vast crowd, known to us in bulk. By assuming the

existence of molecules, we could deduce the behaviour of the crowd and compare it with that observed, but, while we might thus prove that the atomic and molecular hypothesis was one possible explanation, we could never convince ourselves that some other unknown possibility did not lie hid in the obscurity of matter. And for some half century the subject rested at this point.

Nevertheless, the invisibility which veiled its foundations did not check the usefulness of the atomic and molecular theory. By its help, most of the physical and chemical researches of that half century were undertaken. In particular, the marvellous development of electrical science, the explanation of the conduction of electricity through liquids in terms of moving particles, or 'ions,' electrically charged, and the consequent application of similar concepts to the electric phenomena of gases, was undertaken and prosecuted by molecular and atomic conceptions; while the amazing revelations of spectrum analysis showed that the vibrations of the individual atoms were in use as tremors in the web of communication across the star-set space. And to a further development of this same branch of molecular theory we owe Sir J. J. Thomson's discovery of corpuscles with some thousandth part of the mass of the hydrogen atom, identical in all types of matter, and Rutherford's convincing explanation of the marvels of radio-activity as due to the explosive disintegration of atom after atom of the radio-active element, and the ejection of successive sub-atomic corpuscles.

It was indeed the phenomena of radio-activity that opened the new chapter in science we have now to study, and revealed the individual atom as a perceptual entity to the senses of man. The rays which proceed from radium are of three main types, to which the names a, ß and y have been assigned. Of these, both the a and the Brays at any rate consist of minute particles projected from the atoms of the radium salt with immense velocities. By passing them through regions where they are subjected to magnetic and electric forces of known intensity, it is possible to measure their velocity and their mass. While the mass of the particles is identical with that of Thomson's sub-atomic corpuscles, the mass of the a particles

was found at once to be considerably greater, and proved to be of the order of that of the lighter chemical atoms. More refined measurements, undertaken with a view to ascertaining the atomic weight, suggested that they were atoms of helium carrying positive electric charges equal to twice the negative charges on Thomson's corpuscles; and this conclusion was soon confirmed by the demonstration through spectrum analysis that the gas called helium was evolved in measurable quantity by all the salts of radium.

By this time the main lines of radio-active theory had been laid down, chiefly by Rutherford. A mass of radioactive material, such as radium in one of its salts, is subject to a process of atomic disintegration which, in the case of radium, will destroy half of it in about 1760 years. Large quantities of energy are thus liberated, drawn from the internal intrinsic energy of the atomic structures, which some people have likened to Lilliputian models of the solar system. Radium has an atomic

weight (see above, p. 111) of 226. When, in the changes and chances of its molecular life, a radium atom for some unknown reason becomes unstable and explodes, it emits an atom of helium, projected with a velocity of some twelve thousand miles a second. Through the resistance of atmospheric air at the ordinary pressure, equal to that of a column of some 30 inches of mercury, such an atom of helium will travel about two centimetres, or rather less than an inch. The radium residue after this shattering of its constitution has an atomic weight of 222, and forms a molecule of the radio-active gas known as radium emanation. When this molecule in turn breaks up, another a helium particle is projected, and a solid substance with an atomic weight of 218 is deposited on surfaces in contact with the emanation. Further similar processes of disintegration may be traced, until the substance finally seems to settle down again into quiescence, after its period of amazing activity.

In three separate ways these atomic a projectiles have been revealed to the human eye as single independent entities. Soon after the discovery of radium, Sir William Crookes devised a pretty means of investigating the effects of illumination which the rays from radioactive bodies produce on screens of zinc sulphide and

similar sensitive phosphorescent materials. A tiny speck of a radium salt was placed at a distance of a few millimetres from a zinc sulphide screen. When the screen was examined microscopically in a dark room, brilliant little scintillations or flashes of light were visible coming and going on its surface. When the nature of the rays was made clear, it was seen to be possible that each atomic projectile might produce its own scintillation, but it was also possible that a scintillation was caused only when a number of projectiles happened to hit the same spot, or that one projectile might start many scintillations by an indirect effect on the sensitive screen. Further knowledge was needed before the phenomena could be interpreted with confidence as the action of a single individual atom.

In 1908 Prof. Rutherford and Dr H. Geiger devised a new method of bringing the movement of a helium atom before our eyes. All three types of radiation, a, ß and y rays, when passed through a gas, make that gas a conductor of electricity. Indeed, it is by this property that most measurements of radio-active intensity are made. This conductivity has been traced to the production by collision between the a particles and the molecules of the gas of ions, or electrically charged particles, which can move through the gas under an electric force and carry their charges with them. Each atomic projectile causes a train of ions to flash into existence along its path, probably producing them by knocking one of its constituent corpuscles out of each molecule of the gas with which it collides. A single a particle from radium produces some 43,000 ions before its course is run.

The amount of electricity carried by 43,000 ions is just about the limit of sensitiveness of our electrical instruments. Whether we use an electrometer, in which a light, spindle-shaped needle is moved by the attractions and repulsions of electrified plates surrounding it, or an electroscope, where a strip of gold-leaf is repelled so as to stand out at an angle from the brass plate to which one end of it is attached, this amount of electricity is the smallest that our present instruments will measure. Now all who have worked in a laboratory know how unsatisfactory results become when instruments are used near their limit of sensitiveness, and how impossible it is to have any confidence in measurements made in such cir