July 28, 1923] 



NA TURE 



147 



Problems of Fundamental Astronomy.^ 



By Prof. W. de Sitter. University of Leyden 



THE science of astronomy has, in the past twenty 

 or thirty years, developed most remarkably. 



The marvellous applications of photography and 



spectroscopy on one hand, and the sudden growth 



of statistical stellar astronomy consequent upon the 



discovery of the two star-streams on the other, have 



led to so many unforeseen results and so many new 



points of view, that it almost appears as if the whole 



science were born anew and the astronomy of to-day 

 • had only very slight connexions with that of the last 



century : we are apt to think that the great problems 



of the past have lost all their interest to us. This, 



however, is not so. On the contrary, I think the 

 . central problems of fundamental astronomy have 



gained an enhanced importance even by the newest 



■ developments of the science. 



Astronomy is essentially the science of space and 

 time. It is not my intention, in thus assigning to 

 astronomy this wide field, to annex to it the whole 



, of physical science. On the contrary, I am quite 



f content to consider astronomy only as a special branch 

 of physics, but, having at its disposal the largest 

 spaces and the longest times, it has generally had the 

 last word in all important questions. To mention 

 only a few cases at random : the discovery of gravita- 

 tion, of the finite velocity of light, and of aberra- 

 tion, all these are astronomical discoveries, and the 



; three crucial tests of Einstein's theory are all three 



■ astronomical. 



In our exploration of space and time we are com- 

 pelled to make all our measures from this earth, to 

 which we are tied, as a starting-point. The problems 

 of fundamental astronomy are those which arise from 



I this fact, that all our observations are necessarily 

 referred to a moving origin. These problems are, 



;. from their nature, not very liable to change of aspect 



;, with time or fashion ; they are essentially the same 



\ to-day as they were in the time of Hipparchus, the 



; founder of astronomy, and they will remain the same 

 so long as science lasts, and will require ever more 

 accurate and more complete solutions, as we pene- 

 trate more deeply into the constitution of the universe. 

 Fundamental astronomy thus consists essentially of 

 a scrutiny of the last decimal place. This striving 

 after extreme accuracy, this fidgeting over small 

 ciuantities, may appear uninteresting, or even 

 pedantic. But we must not forget that great prob- 

 lems always turn about the measurement of small 

 quantities. 



The problems of fundamental astronomy are, of 

 course, all interconnected with each other, but, for 

 the sake of clearness, they may be classified under 

 three heads. There are, first, the problems connected 

 ith the system of constants. The motion of the 

 rth, and the system of measurement based on it, 

 are defined by several numbers, such as the solar 

 parallax, the constants of precession and nutation, 

 the ellipticity, the mean radius and the mass of the 

 earth, etc. Between these several constants there 

 exist relations, connecting two or more of them with 

 each other and with other universal constants such 

 as the velocity of light and the constant of gravita- 

 tion. The problem here is essentially one of adjust- 

 ment, so as to get a consistent set of constants satis- 

 fying all the connecting relations. The set of con- 

 stants in actual use in the national ephemerides is 

 not consistent. The discordances are, however, not 

 very large, and changes should not be introduced 

 unless by general international agreement. 



' Synopsis of a lecture delivered at the Imperial College of Science and 

 Technology, South Kensington, on May 7. 



NO. 2804, VOL. I 12] 



Another set of problems are those connected with 

 the rotation of the earth. The paramount practical 

 value of this rotation is that it is used as our standard 

 measure of time. 



Time is measured by observing the changes occur- 

 ring in some physical system — i.e. in the relative 

 positions of some material bodies, which positions at 

 any time are determined by our theories, so that 

 from the observed positions we can infer the time. 

 Such a mechanism — by preference periodic — that is 

 used to measure time may conveniently be called a 

 " clock." But there is no absolute measure of time, 

 nor an absolute test of the accuracy of any clock ; we 

 can only test one clock by another. If the two do 

 not give the same time, then one or both must be 

 wrong, i.e. our theories of the mechanism of one or 

 both must be incomplete. The standard clock to 

 which all others are generally referred is the rotating 

 earth. Is this standard absolutely trustworthy ? Do 

 all observatories give the same time, and if so, is this 

 a truly uniform time ? In other words : does the 

 earth rotate as a rigid body, and if so, is this rotation 

 strictly uniform ? 



It has long been suspected that the earth's rota- 

 tion is very gradually slowing down, owing to the 

 friction of the tidal wave.^ But lately other doubts 

 have arisen as to the trustworthiness of our universal 

 standard. As a matter of fact, it is not the rotation 

 of the earth, but the rotation of a definite point 

 on the earth — Greenwich Observatory or any other 

 observatory — that is used as our standard, and now 

 that the wireless distribution of time signals has made 

 comparisons so easy, occasional discrepancies between 

 the times of different observatories, amounting some- 

 times to several tenths of a second, have come to 

 light.' 



It appears probable that these are due to errors 

 in one or more of the parts of the mechanism used 

 to determine the time at some or all of the observa- 

 tories — the transit instruments, the clocks, the 

 astronomers — but it also may be that they are due 

 to real differences in the rotation of the different 

 observatories, which would mean that the earth does 

 not rotate as a rigid body, but some parts of its 

 surface are moving relatively to other parts.* Here 

 evidently is a most important problem, the solution 

 of which must be found sooner or later. 



Besides the rotating earth, we have other " clocks," 

 of which the moon must be mentioned in the first 

 place. It is well known that in the motion of the 

 moon there are irregularities of a much longer period, 

 called " fluctuations " by Newcomb, for which no 

 explanation has yet been found. Brown ^ and 

 Glauert « have pointed out similar irregularities in 

 the motions of the sun, Venus, and Mercury. If this 

 were confirmed, and if also other bodies — especially 

 Jupiter's satellites — should show the same thing, then 

 it would become very probable that the true origin 

 of these fluctuations is in the rotation of the earth, 

 or at least of the outer crust of the earth. 



Other problems connected with the rotation of the 

 earth, and the question whether it rotates as a rigid 

 body, are those involved in the variation of latitude. 



* Taylor, Mon. Not. R.A.S. Ixxx. 308 ; Jeflreys, ibid. 309. 

 ' .Sampson, Mon. Not. R.A.S. Ixxxii. 215 ; Dj^on and Bowyer, ibid. 



Ixxxii. 193. 



* Dr. Innes (Johannesburg Circular 55) has recently directed attention to 

 irregularities in the moon's motion of the same character as tlm discordances 

 between the times of different observatories referred to above. But these 

 are derived from several observatories, giving concurrent results, so that 

 it would appear that the error is in the moon, and not in the time. 



• Brit, .\ssoc. Report, Australia, 1914. 



• Mon. Not. R.A.S. Ixxv. 489. 



