324 



NATURE 



[January 5, 191 1 



93,000,000, except in the minds of those who are in some 

 measure acquainted with the laborious processes by which 

 the two extra figures are derived. In fact, I have not 

 infrequently heard the methods of observation used 

 described by some such epithet as " hair-splitting." For 

 this reason I thinli I cannot do better to-night than to 

 describe to you, without entering into technical details of 

 the methods employed, some of the aims and objects to 

 which modern astronomy of precision is devoted, and which 

 render essential none but the highest refinements that 

 human ingenuity can devise. 



Perhaps the primary reason why astronomy appeals to 

 the popular ima<jination in a higher degree than other 

 sciences is that astronomy is par excellence the science of 

 prediction. True, the days are now past when an astro- 

 nomer is regarded, except by the most ignorant, as gifted 

 with supernatural powers and capable of predicting events 

 that can have no conceivable relationship with the objects 

 of his researches, or when an unscrupulous astronomer 

 could utilise his powers of prediction for imposing on the 

 world at large in the face of the criticisms of fellow- 

 workers in collateral branches of science. Nevertheless, it 

 is only necessary to point to any of the leading almanacs 

 to establish the undoubted claim of astronomy to a con- 

 siderable predictive capacity in its own legitimate sphere. 

 These almanacs, prepared in advance, give from day to day 

 the positions of the sun and moon, the phenomena of 

 eclipses, and various other data with an accuracy which 

 can only be called in question by the most refined tests 

 available to astronomers. 



How, then, has astronomy acquired this faculty? The 

 answer to this question is — at least primarily — by con- 

 tinuous and patient observation, using always the most 

 refined methods of physical measurement available. 



A well-devised scheme of observation is sooner or later 

 bound to lead to the detection of laws governing physical 

 phenomena if such laws exist. Thus it was the planetary 

 observations of Tycho Brahe which led to the detection of 

 the laws of planetary motion associated with the name of 

 Kepler. 



Once such laws have been established and the necessary 

 initial data secured, the science of astronomical prediction 

 would for the future devolve on the mathematician rather 

 than the astronomer, were it not for two sources of un- 

 certainty with which the astronomer must continue to con- 

 cern himself. It is evident, on one hand, that we 

 cannot infuse into our predicted phenomena greater pre- 

 cision than that derived from the initial data, themselves 

 dependent on imperfect observations. However well the 

 laws governing planetary motions may be understood, the 

 predicted position of a planet to-day depends on its observed 

 positions at some earlier epoch or epochs ; and the falli- 

 bility of the observations made at these earlier epochs will 

 not only pervade all future predictions, but will inevitably 

 increase in amount as the epoch of prediction recedes from 

 the epoch of observation. For this reason, if the standard 

 of accuracy of prediction is mereh' to be maintained — and 

 the growing requirements of science will scarcely rest con- 

 tented with this — continuous observation must be main- 

 tained and the data on which predictions are based revised 

 from time to time. 



I have dealt so far only with the effects of the unavoid- 

 able inaccuracies of observations, even when pushed to 

 their utmost refinement, as influencing results of prediction. 

 A second consideration of even greater importance is the 

 validity of the laws associating the predicted with the 

 observed phenomena. I may illustrate my point again by 

 reference to the laws of Kepler. It is now well known 

 that these laws are only rough approximations to the 

 actual truth, and that though they might serve as a useful 

 basis for prediction over a short interval, a few years at 

 most would suffice, by showing a rapidly increasing de- 

 parture between the observed and predicted positions of 

 the planets, to Indicate that these laws require amend- 

 ment. 



That the direction this amendment should take followed 

 so soon on the original discovery of Kepler's laws was 

 due to the geniu? of Newton, who showed that the theory 

 of universal gravitation propounded by him not only 

 adequately accounted for the laws enunciated by Kepler, 

 and pointed to their imperfections, but served to coordinate 

 as due to a single cause even more recondite phenomena. 



NO. 2149, VOL. 85] 



such as the leading inequalities in the motion of the moon, 

 the precessional motion of the earth, and the phenomena 

 of the tides. This theory further reduced to order tho^' 

 astronomical vagaries the comets, . showing that, so lonj; 

 at least as they remained within the precincts of the solar 

 system, their motions tvere governed by it, while observa- 

 tions of double stars have established beyond question that . 

 even remote parts of the universe are still subject to thf 

 same laws. 



The dynamical laws propounded by Newton, which to- 

 day virtually form the basis of all astronomical prediction, 

 enable us to trace back as well as to trace forward th' 

 history of the solar system, and to confront modeii; 

 observations with historical records. Needless to say, in 

 but rare instances do these records possess the necessar\ 

 elements of precision to strengthen the existing data re- 

 quired by the astronomer ; but there are important excep- 

 tions. For instance, a very small uncertainty in th> 

 " elements," which in conjunction with Newton's law- 

 govern the motion of the moon, will produce by lapse ol 

 time a large change in the comparatively small area <<\ 

 the earth's surface over which a total eclipse of the sun 

 is visible as such. Thus a record that a particular eclip^ 

 was seen as total in a given locality becomes an observa- 

 tion of precision, provided only the chronological date n 

 which the eclipse occurred can be traced with sufiicien 

 certainty to ensure the identification of the eclips 

 concerned. 



The confrontation of modern with historical observ,'.- 

 tions of such a character has served to establish beyond 

 question the high degree of accuracy with which the law-s 

 of Newton represent the motions within the solar system, 

 and their trustworthiness as a basis of prediction for years, 

 perhaps for centuries, to come. It is, however, on 

 various grounds quite certain that these laws in themselves 

 are not absolute, far-reaching though they are, and that 

 they in turn, like those of Kepler, must be superseded b 

 laws still more exact. 



Until such laws are discovered there will always remai; 

 an element of uncertainty, apart from that due to tb 

 initial data affecting all predicted phenomena — an un- 

 certainty which can only be removed when the phenomena 

 cease to be prospective, and when they can be confronted 

 with later as well as with earlier observations. 



The fact, however, that the laws of gravitation yiekl 

 such a close representation of the observed motions within 

 the solar system throughout historic time renders the 

 detection of a departure from these laws a question of 

 extreme delicacy, but none the less essential, if predictio 

 is to be secured for long periods in advance. 



I have selected my illustrations largely from the sob^ 

 svstem chiefly on the grounds that, thanks to the New- 

 tonian laws,' it is here that, in spite of the immense 

 mathematical difficulties which have had to be faced, 

 astronomical prediction has attained its greatest triumphs. 

 I wish now to divert attention to the stars. In so far as 

 these form the fiducial points to which the motions of the 

 planets and other members of the solar system are re- 

 ferred, it is essential that the positions of a limited 

 number, at least, should be determined with the highest 

 possible accuracv. Any uncertainty in their positions will 

 undoubtedlv be 'reflected in the positions of the planets, 

 and will constitute one of those sources of error so liable 

 to increase with time, and render efforts at prediction, if 

 not entirelv nugatorv, at least partially ineffective. 



The universe of so-called " fixed stars " is not invariabl 

 in aspect, though its changes, for the most part, are of so 

 minute a character that they can only be surely detected 

 either bv the most delicate measurements or by their 

 cumulati've effect over long intervals of time. It is chieflv 

 through a study of these changes that our knowledge of 

 the stellar universe has been acquired in the past, and it 

 is largely to similar means that we look for an extension 

 of this knowledge. 



Among changes which lend themselves to observation 

 for this purpose we mav enumerate : — 



(i) Changes of the intensity of the light of the stai- 

 The origin of these changes, except in a few instanc* - 

 remains obscure. In certain cases, however, notably in 

 the case of variable stars of the Algol type, a satisfactory 

 explanation of the observed phenomena has been found in 

 the motions of a system, governed by laws similar to those 



