PRESIDENTIAL ADDRESS. 315 



in 1S53-60 it was time of halt at an extreme: the period remained stationary 

 and the variation repeated itself eleven times in closely similar fashion, so that 

 Pogson concluded it would continue in the same way. How many instances 

 suffice for an induction? Many inductions have been based on fewer than eleven. 

 Unfortunately the period was just beginning to change sensibly, and we lost 

 much valuable information, for no one else repaired Pogson's neglect adequately : 

 and the whole swing of period occupies about forty years, so that the opportunity 

 of studying the changes he missed has only quite recently returned. We are thus 

 reminded how disastrous may be a break in the record. It should be one of the 

 articles of faith with an observer that the record is sacred and must not be 

 broken. Most of them indeed act on that principle already, but there arc hsretics, 

 and it pained us to find even Professor Schuster himself tinged with heresy. 

 On the very occasion when he did so much for the observer by presenting his 

 beautiful method, he suggested that it might even be advisable to drop observing 

 for a time in order to apply the method to accumulated observations. He may 

 possibly be right, but the observer had better believe him wrong. There ought 

 to be an ' observer's promise ' like the promise of the boy scout ; and one part 

 of it should be not to interrupt the record, and another should be to publish 

 the observations regularly, and never to let them accumulate beyond five years. 



The method of Professor Schuster is not the only one that has been recently 

 proposed for dealing with large masses of observations. We have also the 

 methods of Professor Karl Pearson. These have been far more widely adopted 

 for use than the periodogram, and they have also been more adversely criticised. 

 As regards criticism, I think it is fair to say that it has chiefly been directed 

 towards the nature of the material on which Professor Pearson has used his pro- 

 cess rather than on the process itself, and at present we need not be concerned 

 with it. The processes themselves are sound enough ; one of them, for instance, 

 is much the same as the old method of least squares in a simple form. Put if 

 the same criticism is made as has been made on the method of the periodogram— 

 viz., that it is not new, we can also reply in almost the same words in the two 

 cases : the mathematical calculus may not be new, the novety is the insistence 

 on the application of it, and the application to all possible cases. Professor 

 Pearson ceases to look for one principal factor only, and examines all possible 

 factors, just as Professor Schuster examines all possible frequencies. Let us 

 recur for a moment to the words of Sir George Darwin previously quoted : — 



A mere catalogue of facts, however well arranged, has never led to any 

 important scientific generalisation. For in any subject the facts are so 

 numerous and many-sided that they only lead us to a conclusion when they 

 are marshalled by the light of some leading idea. 



Let us take, for instance, a catalogue of variable stars such as those of Mr. 

 Chandler. Particulars for each star are given in separate columns, exclusive 

 of the name and number. We might wait long for a leading idea to guide us 

 in marshalling the facts, and so far as I know wo have waited till now without 

 any such idea occurring to anyone. But Professor Pearson insists on the plain 

 duty of determining the correlation between each and every pair of these columns, 

 and any others we may be able to add. Anybody could have made the sugges- 

 tion, and there was plenty of elementary mathematical machinery in existence for 

 carrying it out; but so far as I know nobody did, any more than the critics of 

 Columbus suggested how to stand up an egg. But the suggestion having been 

 made by Professor Pearson, it was so clearly sound that I did what lay in my 

 power to follow it up : with the result that certain correlations were at once 

 indicated which at least pave the way for further inquiry. If we cannot say 

 more than this it is simply because the catalogue of facts was not large enough. 

 So far from the observers having wasted their energies by observing without any 

 theory to guide them, more work of the same kind would have been welcome, for 

 it would have reduced the probable error of the correlations indicated. As an 

 example I may quote the following. It has already been mentioned that a 

 variable-star-maximum, though it may recur after a more or less definite period 

 on the average, is subject to a swing to and fro like the time of sunrise. Let us 

 call the average interval the (lay of the star and the period of swing the year, 

 without implying anything more by these names than appears in the analogy. 



