170 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1937 



certainty the real meanings of the phrases which they used. He 

 became interested in the astronomical side of the records and took 

 the trouble to learn how to deal witli them according to the best 

 canons of the astronomer. He became thoroughly expert in this 

 part of the work, and the combination of knowledge and judgment 

 which he gave to it has produced excellent results. 



Some years ago Prof. G. I. Taylor added to our stock of information 

 by showing that the greater part of the friction came from the tides 

 in shallow seas. He calculated that the Irish Sea alone produced 

 about one-fiftieth of the necessary amount. A little later Dr. Jeffreys 

 made a similar survey of the whole earth. He found, though his 

 calculations were necessarily rather rough because only rough data 

 were available, that the total amount was about the same as that 

 given by observation. One interesting fact emerged, namely, that 

 the Bering Sea contributed the greater part of it. Had better data 

 for that stretch been available he would have been able to make his 

 calculations much more exactly. 



Tliis result of Jeffreys raises a point which is sometimes forgotten, 

 namely, that the present figure for the amount of the friction, and 

 consequently for the slowing down of the earth, may have no great 

 degree of permanence. The southern boundary of the Bering Sea is 

 a higlUy volcanic region in wliich changes are continually taking place. 

 It requires no great stretch of the imagination to suppose that in the 

 course of a few thousand years the depth of this sea and its boundary 

 may materially change and that in consequence, changes will occur in 

 the amount of the friction. Thus caution is needed when we use the 

 present figure for arguments concerning the past and future history 

 of the earth-moon system — a question I shall deal with briefly below. 



The amount of this gradual change is very small — the day is longer 

 now by less than one thousandth of a second than it was a century 

 ago. The error of the clock increases much faster — in fact, as the 

 square of the elapsed time. If the clock loses 1 second in a century 

 it will lose 100 seconds in 10 centuries, and in 100 seconds the earth, 

 in the middle north or south latitudes, turns a good many miles. 

 This is why the place of an ancient total eclipse is so sensitive to small 

 changes in the length of the day. 



Of much more scientific importance is the change in the moon's 

 motion. As the earth is losing energy, the moon, which is responsible 

 for the greater part of the change because of the fact that it is the 

 chief tide raiser, must be gaining energy. This makes it move around 

 the earth more slowly and sends it farther away from us. If tliis 

 went on indefinitely the moon would leave us. But it is calculated 

 that when the month is about twice as long as at the present time it 

 will stop receding. "\Miat will happen afterward is largely a matter 

 of speculation. 



