Evidence gradually accumulated that the amounts of actinium in radio-active minerals were, roughly at any rate, proportional to the amounts of uranium. This result pointed to a lineal connection between them, and led Boltwood to undertake a direct attack on the problem. Separating a quantity of actinium from a kilogramme of ore, Boltwood observed a growth of 8.5 x (10 to the power -9) gramme of radium in 193 days, agreeing with that indicated by theory within the limits of experimental error. ("American Journal of Science", December, 1906.) We may therefore insert provisionally actinium and its series of derivatives between uranium and radium in the radio- active pedigree.

Turning to the other end of the radium series we are led to ask what becomes of radium-F when in turn it disintegrates? What is the final non- active product of the series of changes we have traced from uranium through actinium and radium?

One such product has been indicated above. The alpha-ray particles appear to possess the mass of helium atoms, and the growth of helium has been detected by its spectrum in a tube of radium emanation. Moreover, helium is found occluded in most if not all radio-active minerals in amount which approaches, but never exceeds, the quantity suggested by theory. We may safely regard such helium as formed by the accumulation of alpha-ray particles given out by successive radio-active changes.

In considering the nature of the residue left after the expulsion of the five alpha-particles, and the consequent passage of radium to radium-F we are faced by the fact that lead is a general constituent of uranium minerals. Five alpha-particles, each of atomic weight 4, taken from the atomic weight (about 225) of radium gives 205--a number agreeing fairly well with the 207 of lead. Since lead is more permanent than uranium, it must steadily accumulate, no radio-active equilibrium will be reached, and the amount of lead will depend on the age of the mineral as well as on the quantity of uranium present in it. In primary minerals from the same locality, Boltwood has shown that the contents of lead are proportional to the amounts of uranium, while, accepting this theory, the age of minerals with a given content of uranium may be calculated from the amount of lead they contain. The results vary from 400 to 2000 million years. ("American Journal of Science", October, 1905, and February, 1907.)

We can now exhibit in tabular form the amazing pedigree of radio-active change shown by this one family of elements. An immediate descent is indicated by >, while one which may either be immediate or involve an intermediate step is shown by .... No place is found in this pedigree for thorium and its derivatives. They seem to form a separate and independent radio-active family.

Atomic Weight Time of half Radio-Activity decay

Uranium 238.5 alpha > Uranium-X ? 22 days beta, gamma ... Actinium ? ? no rays > Actinium-X ? 10.2 days alpha (beta, gamma) > Actinium Emanation ? 3.9 seconds alpha > Actinium-A ? 35.7 minutes no rays > Actinium-B ? 2.15 minutes alpha, beta, gamma ... Radium 225 about 2600 years alpha > Radium Emanation ? 3.8 days alpha > Radium-A ? 3 minutes alpha > Radium-B ? 21 minutes no rays > Radium-C ? 28 minutes alpha, beta, gamma > Radium-D ? about 40 years no rays > Radium-E ? 6 days beta (gamma) > Radium-F ? 143 days alpha ... Lead 207 ? no rays

As soon as the transmutation theory of radio-activity was accepted, it became natural to speculate about the intimate structure of the radio- active atoms, and the mode in which they broke up with the liberation of some of their store of internal energy. How could we imagine an atomic structure which would persist unchanged for long periods of time, and yet eventually spontaneously explode, as here an atom and there an atom reached a condition of instability?