October 17, 1907 



NATURE 



613 



LETTERS TO THE EDITOR. 

 [The Editor does not hold himself responsible for opinions 

 expressed by his correspondents. Neither can he undertake 

 to return, or to correspond with the writers of, rejected 

 manuscripts intended for this or any other part of Nature. 

 No notice is taken of anonymous communications.] 



The Advancement of Science. 



So.ME weeks ago Sir Oliver Lodge directed atlention to 

 the congestion uf work in Section A of tlie British 

 Association at Leicester. Last week Dr. Chree remarked 

 upon the comparative neglect on our part of such scientific 

 subjects as terrestrial magnetism and the local variations 

 of gravity, which cannot be pursued adequately within 

 the walls of a laboratory, but depend upon observations. 

 " in the field." In the meantime, Nature has noted Dr. 

 Mill's protest against the scanty opportunity for the dis- 

 cussion of meteorological subjects at Leicester. 



My own experience at Leicester supports Dr. Mill's 

 protest. I do not refer to what happened to my own 

 paper. I have no complaint to make against the officers 

 of the section, who, with the rest of us, were victims of 

 an unworkable system. I refer to the proceedings with 

 regard to the papers by Mr. Petavcl and his fellow-workers 

 at Manchester on the investigation of the upper air. 



Since the meeting, I have learned that the results of 

 the work of the international week at Manchester (the last 

 week in July) were of remarkable interest as showing in 

 an e.\ceptional manner all the characteristic features of 

 the variation of temperature in the atmosphere up to a 

 height of 20 kilometres. The papers were among a large 

 number on the programme for Tuesday, August 6. When 

 I left the section at three o'clock to attend a conference 

 as the delegate of Section A, they had not been reached. 

 When I returned at half-past four I was told that the pro- 

 ceedings of the section for the year had already been con- 

 cluded, with the usual votes of thanks. Whether the 

 papers had been read in the interval or withdrawn I do 

 not know, nor is it of much consequence. If the only 

 time to be found for a subject of such general interest 

 is after three o'clock on Tuesday afternoon, it is clear 

 that some change is required. 



The sectional proceedings on Tuesday opened with a 

 discussion upon new methods of treating observations, an 

 important practical matter for the observational sciences. 

 In the circumstances, it was evidently desirable that the 

 opening paper should be printed in extenso ; but the 

 recorder pointed out to me, and quite rightly, that such 

 a proposition could not be entertained by the British 

 Association, because the committee of Section A had 

 adjourned for the year on the previous day. 



These things do not make for the advancement of 

 science. 



I wish, however, to take up the point raised by Dr. 

 Chree, and to emphasise the fact, already too obvious to 

 those who have to do with such things, that subjects like 

 teri-estriat magnetism, seismology, atmospheric electricity, 

 and the physics of the globe generally, without any refer- 

 ence to meteorology in particular, suffer very seriously in 

 this country from the congestion of work in Section A. 



On the one side, work is done of which the scientific 

 public know little or nothing. .Atmospheric electricity is 

 a flourishing study on the Continent ; seismology is now 

 the subject of an international organisation with Govern- 

 ment support; terrestrial magnetism has called for expendi- 

 ture on a large scale for an establishment to replace Kew 

 as the normal observatory. It is desirable that the 

 association should know what is going on in such 

 matters. 



There are, moreover, a number of departments of 

 Government the work of which has at least its scientific 

 side. Papers of scientific interest could probably be 

 obtained for the asking, from a number of competent 

 workers, by an energetic president or secretarv. anim,ated 

 by the meritorious wish to use the meeting of the .nssoci- 

 ation to bring the scientific staff of the various denart- 

 ments into touch with the scientific public; but the officials 

 in charge of such work have not the advantage of 

 academic long vacations. The time spent at the British 

 Association must be taken either from short leave or from 

 duty. The matter must, therefore, be treated in a 



NO. tq8i, vol. 76] 



business-like way, which in present circumstances i? 

 impossible. 



it would be absurd, for example, for a secretary or &\\ 

 organising committee to ask, let us say, the hydrograplicr 

 of the Navy for a paper on submarine centres of magnetic 

 disturbance, or the Astronomer Royal for a paper on mag- 

 netic storms and sun-spots, or any other aspect of the 

 magnetic or meteorological work of the Royal Observ- 

 atory. There is the paralysing consciousness that the time 

 for reading the papers would have to be looked for in a 

 general scramble between three and half-past four on 

 Tuesday afternoon. What is true with regard to these 

 distinguished public servants is equally true with regard 

 to distinguished foreign workers in science. 



There is provision in the constitution of the association 

 for asking competent persons to prepare reports upon 

 recent progress in particular branches of science. The 

 provision is, unfortunately, a dead letter in the subjects 

 mentioned. One reason at least is obvious — there is no 

 time to listen to such reports, however valuable they 

 might be. 



It is time that we recognised that the attempt to include 

 in one section with mathematics subjects like laboratory 

 physics, in which workers are many and in constant inter- 

 communication, and subjects like terrestrial magnetism, 

 atmospheric electricity, and other branches of geophysics, 

 in which workers are few and widely scattered, is 

 disastrous for the one class of subjects ; and, judging by 

 the way in which a discussion upon so important a sub- 

 ject as the measurement of temperature by radiation was 

 received at Leicester, it is not too successful for the other 

 class. 



Some years ago there used to be a subsection for the 

 outdoor subjects, with the not very euphonious title of 

 " .Astronomy and Cosmical Physics " — perhaps astronomy 

 and geophysics might be better. It has disappeared — not 

 on account of any want of success while it lasted. It was 

 simply omitted from the South African arrangements. 

 The circumstances which called it into existence have now 

 become more pressing. Laboratory physics has become 

 more radio-active, and the other subjects have extended 

 their operations. The temporary expedient of a special 

 subsection is not now adequate. One special secretary at 

 least is required in the interest of those branches of geo- 

 physics which are not covered by astronomy. The 

 occasional treatment of such subjects in a presidential 

 address would be of real advantage to science in this 

 country. 



I ask, therefore, the hospitality of the columns of 

 Nature in order to appeal, in the name of the advance- 

 ment of science, for the establishment of an independent 

 section of the British Association which shall have 

 sufficient time at its disposal to promote the advancement, 

 not only of meteorology, but also of such subjects as terres- 

 trial magnetism, atmospheric electricity, seismology, and 

 geophysics generally. The briefest consideration of the 

 changes which have taken place since Section A was 

 initiated will show that such an appeal is not unreason- 

 able. \V. N. Shaw. 



October 12. 



On Correlation and the Methods of Modern Statistics. 



I REGRET that the pressure of work associated with the 

 opening of a new session did not permit of my replying to 

 Mr. Hinks last week. His letter (October 3) is so far 

 satisfactory that it gives evidence that one professional 

 astronomer realises the existence of stellar correlation ; but 

 Mr. Hinks will have to advance much beyond " scatter 

 diagrams " before he can hope to get much profit out of 

 modern methods. Further, may I suggest that he would 

 be more just to both Miss Gibson and myself if (i) he 

 read her paper carefully, and (2) he did not suppose that, 

 because we approach the subject from a different stand- 

 point from himself, we are of necessity both very ignorant 

 and very foolish ? 



.At the expense of reiteration, I must again refer to 

 one or two facts. There are, in my opinion, three points 

 of much interest in Miss Gibson's memoir; — 



(a) The correlation of magnitude and parallax is shown 

 to be low ; what correlation exists is shown to depend 



