The assumptions involved in making these adjustments are inherently improbable, and it might be thought easier to assume that the time values of the post-Carboniferous strata were, as compared with the earlier strata, less. This emendation requires us to assume that the more recent materials were laid down about three times as fast as the earlier.

These are the alternative modes of adjustment of radioactive time to the stratigraphical column, leaving the latter on the whole intact. If we assume that the recent sediments have been overestimated in thickness, we can, by discarding about one-half the recorded thicknesses since Carboniferous time, produce an effect on the diagram equivalent to moving the origin to the right. With this particular numerical assumption the lead line will become steeper than it appears on the chart, and the early Algonkian point will remain at such a distance to the right of the Cambrian as will ascribe to the preCambrian sediments a thickness equal to that of the whole postAlgonkian accumulation.

The important question is, of course, as to how far such assumptions are permissible consistent with any degree of probability. There is much that is uncertain about data respecting rock thickness, not only as regards the actual field observations, but as to the real significance of what is observed. Again, the relative time equivalents of deposited rocks are not really known to us. Whether it is a detrital sediment forming in an estuary or a coral-reef building in clear water, the rate of growth must depend to some extent on the downward movement of the sea bottom, either induced by the load or taking place from other causes. Some sediments are, however, plainly of rapid and some of slow growth. Amidst such considerations we find no very definite grounds for numerical computation. So far as crustal yielding affects the question, the probable inference is, as I have stated above, that the earlier strata were in their greatest development more localized, and hence their time value should be less than the more recent. As regards the vertical distribution of definitely fast or slow collecting materials, a careful comparison of the materials throughout the geologic column is required in order to gather any evidence that may be forthcoming from these indications. At present, however, there seems nothing to support the different time values or amended thicknesses which must be assumed if we are to adjust the radioactive results in any way to the sedimentary record. What will prima facie appear most difficult to credit in the foregoing assumptions is the extremely slow rate which must be ascribed to the accumulation of the sediments even at their maximum. the recorded depths of sediment have taken 1,400,000,000 years to collect, the average rate has been no more than 1 foot in 4,000 years! This seems incredible; and if we double the depth of maximum sedi

If

mentation it still remains incredible. But, if possible, still more incredible is the conclusion respecting solvent denudation to which radioactive time drives us. If the sodium in the ocean has taken 1,400,000,000 years to accumulate, the rivers are now bearing to the sea about fourteen times the average percentage of the past—not less than nine times. It seems quite impossible to find any explanation of such an increase.

With these difficulties in view it is excusable to direct attention to the foundations of the radioactive method and ask how far they are secure. The fundamental assumption is that the parent radioactive substance, uranium, has always in the past disintegrated at the present rate. Is this assured? I am not now suggesting that the rate of change has been effected by external physical conditions, such as heat or pressure, but I assume that there may have been a different, and from the evidence as well as from probability, a greater rate of decay in the past, arising intrinsically, and ultimately due possibly to conditions of origin.

I venture to suggest-I do so with diffidence-that our assumption of a constant rate of change for the parent substances-uranium or thorium is really without any very strong basis. It rests upon analogy with the behavior of the substances which have been derived from them. But there may be a very profound distinction. The latter are of radioactive origin. That particular distribution of stability or of intrinsic energy among the atoms of these bodies obtaining at the moment of their formation, upon which the subsequent constant change rate depends,' may be conditioned by the events of radioactive transformation, or by their past history, or by both. In a word, a radioactive origin may be essential.

Now we know nothing as to the origin of the primary radioactive elements. No substances of greater atomic weight are known from which they may be derived. Nor is it unphilosophic to assume that they have had some other mode of origin, seeing that the radioactive ascent must terminate somewhere. Uranium can not be regarded, therefore, as in all senses one of a series any more than we should regard lead as such.

The matter seems to turn upon the legitimacy of the assumption that the mere existence of radioactive change progressing in the substance involves such a particular distribution of instability among its atoms as will insure that a constant fraction of these disintegrate each unit of time from their first origination-however this was brought about-till all are transformed. If such an hypothesis is not sufficiently secure to overbear the opposing evidence we must agree to judge the former by the latter. In this case the accumulation of

1 See Sir J. J. Thomson's Presidential Address to the British Association, 1909.

transformation products in minerals, in place of being a measure of geologic time, serves to shed light upon the rate of transformation of the primary radioactive bodies in the past. Apart from its interest in other respects, the importance of such a conclusion to geologic science would be great. If we supposed the curve, found by plotting the time results derived from lead ratios against the sedimentary thicknesses, represented an approximation to the facts, the rate of change of uranium 150,000,000 years ago may have been many times what it now is. The radiothermal effects of the whole series must have been proportionately increased, and the convergence of the radioactivity must have had an influence upon the secular cooling of the earth,

July 18, 1911.

INTERNATIONAL AIR MAP AND AERONAUTICAL MARKS.1

By CH. LALLEMAND,2

President of the French Association for the Advancement of Sciences.

1. PRELIMINARY ACCOUNT.

The dirigible air balloon, and more especially the aeroplane, which are scarcely out of the period of research and experiment, will soon enter into the area of practical politics. To-day, still mere instruments of sport or of military reconnaissances, they will become, to-morrow, valuable means of transport. It is time that aviators were given means for finding their way, similar to those which, for a long time, have existed for navigation and travel.

Whether on sea, land, or in the air, the pilot has always before him the same triple problem to be solved. He must from time to time

1. Recognize his position.

2. Determine the direction of the point to be reached.

3. Rapidly estimate the distance remaining to be covered. For terrestrial locomotion, the solving of these various problems was greatly simplified by producing special maps for the use of travelers and by erecting along the principal routes easily visible signs, such as milestones, plates indicating the names and distances of more or less remote towns, signposts at road junctions, etc. But the marking or buoying of routes presupposes a course fixed in direction and limited in extent. Admirably suited for travel by land, this system is still, to a certain extent, applicable to coasting trade, that is, for sea voyages along coasts or in estuaries. But, on the other hand, for travel on the high sea this method is unsuitable, and for aerial navigation quite useless in foggy weather or at night. In the latter case it is necessary to use a compass. In spite of fog, darkness, or the absence of marks, the use of the compass enables the pilot to follow the desired direction, which direction has been previously

1 British Association, Portsmouth, September, 1911. Reprinted by permission from The Geographical Journal, London, vol. 38, No. 5, November, 1911.

2 Inspecteur-Général des Mines, Directeur du Service du Nivellement général de la France, Membre de l'Institut et du Bureau des Longitudes.