270 



NATURE 



[July 21, 1898 



argon. If that is the case there seems, as far as I know, no 

 a priori reason why sparking with oxygen should necessarily 

 remove the carbon. The ratio of specific heats must take care 

 of itself. It is a matter of the greatest interest to pursue the 

 subject ; for the origin of the spectrum, whatever it may turn out 

 to be, will probably throw much light on the source of the 

 spectra of comets and of carbon stars. 



Arthur Schuster. 



Liquid Hydrogen. 



Mr. Hampson seems insatiable of contradictions. He has 

 produced a vast quantity of irrelevancies with which I have no 

 concern. But I have denied the accusations he brings against 

 me, and every single statement of his that is relevant. Yet he 

 still complains that I do not deny enough. It is absolutely false 

 to say that I appropriated or profited by any plan, idea, or state- 

 ment of Mr. Hampson's, either directly or indirectly. I was 

 never informed of his visit, far less of any of the plans he brought 

 to the Royal Institution, nor would anything have induced me to 

 look at them. I have been long enough in this " temple of 

 science " not to know what that might involve. Mr. Hampson 

 got at my assistant behind my back, and persuaded him to look 

 at the plans. I infer from the public correspondence, that he 

 saw that they would not work, and he told Mr. Hampson why 

 they were unworkable. 



Even with this assistance it took Mr. Hampson another year 

 to perfect a provisional specification of his invention, which is 

 totally devoid of any plan or drawing of a workable apparatus. 



In the meantime Linde had completed his invention, and the 

 Royal Institution went on working on its own lines, just as it did 

 before Mr. Hampson was heard of, and as it would have done 

 had he never been heard of at all. 



Like the rest of us, Mr. Hampson was using ideas and 

 principles established by other men, and was trying to apply 

 them and combine them so as to reach a given result. He has 

 no property either in the principles or in the idea of combining 

 them, or in anything except the particular combination to which 

 he himself may give concrete form. 



Other men besides myself have successfully combined these 

 principles without any help from Mr. Hampson. 



Long before Mr. Hampson's patent was published, I said at 

 the Chemical Society in 1895 • " It is a mistake to attribute to 

 Linde the idea of using the cumulative withdra\Yal of heat for 

 the first time in his apparatus, but he has succeeded in making a 

 workable industi-ial machine, and that is a very important step." 



In the Society of Arts Journal I said : " Both Onnes and 

 myself used or economised the temperatures of the expanding 

 gas in order to cool the gas coming forward, but Linde was 

 entitled to every credit for elaborating a machine in which this 

 was done as perfectly as possible. " 



Further, in the Chemical Industry Journal the following 

 passage makes my position clear : "He (Prof. Dewar) was 

 willing to give all credit to Dr. Hampson, Dr. Linde, and any 

 one who effected improvements in these investigations. All he 

 asked was that they should not exaggerate their claims, and seek 

 to block the way to other people who were working in the same 

 direction. Dr. Hampson did not appear to realise that anybody 

 else could be working in the same path and utilising the same 

 ideas. It was quite clear, however, from the facts before them, 

 that that was precisely the state of affairs in the present case." 



Such extracts show that I have recognised to the full the 

 merits of the true inventor within the limits of his just claims. 



James Dewar. 



Summer and Winter in Relation to the Sunspot Cycle. 



The quality of a winter season may be fairly estimated from 

 the number of days on which the minimum temperature has 

 gone below a given limit ; and the quality of a summer season, 

 from the number of days on which the maximum temperature 

 has gone above a given limit Two tables issued from Green- 

 wich are here convenient for use ; one giving frost days (since 

 1841), the other days on which the temperature reached or 

 exceeded 70°. There are more of the latter than of the former ; 

 seventy-seven on an average, as against fifty-five frost days (in 

 September to May). 



We may roughly call a winter season severe or mild, accord- 

 ing as it has more or less than the average number of frost 

 days ; and a summer season hot or cool, according as it has more 

 or less than the average number of hot days (in the sense 

 specified). 



NO. 1499, VOL. 58] 



Confining attention to the groups of five consecutive years 

 having a sunspot maximum, or minimum, year third (or 

 central), I propose to inquire whether there is anything in the 

 winters and summers of these groups pointing to sunspot 

 influence. 



(The sunspot maximum years are 1848, i860, 1870, 1884, 

 1894; and the minima, 1843, 1856, 1867, 1879, 1890. Winters 

 may, for brevity, be designated by the year in which they end ; 

 thus, 1842 means 1841-42. In the tables exceptions are 

 marked e\ 



The following statements regarding winter may be verified : — 



( 1 ) In five-year groups having each a sunspot maximum year 

 third {or central), there are generally more mild winters than 

 severe. 



(2) In five-year groups having each a sunspot minimum year 

 third {or central), there are generally more severe ivinters thair 

 mild. 



The proof is in these tables : — 



>5 ' 



17 



The first table shows one exception to the rule (maximum of 

 1870). The five years' total, however, it may be stated, i& 

 under, not over, the average.^ In table B the winter 1840-1 

 has been included, though the record properly begins with 1841. 

 It was a severe winter. The table has one group with an 

 average winter, so that this group may be considered neutral. 



Coming now to the summer season, proof is offered of the 

 following : — 



(3) In five-year groups having each a sunspot maximum year 

 third {or central), there are generally more hot summers than 

 cool. 



The corresponding and opposite statement for minima seems 

 hardly warranted by the present data. The tables are these : — 



15 



14 



Here we have one exception to our third rule, the group for 

 min. 1884 showing two hot summers and three cool ones. The 

 table D has three cases pointing one way, and two the other.^ 



In view of these facts, I have sought light from a different 

 quarter, taking the mean temperature of the four months May to- 

 August, and dealing with the series (from 1841) in the same way. 



We thus obtain the following tables for summer : — 



E. 



Hot. 



Min. 

 groups. 

 1841-45 

 1854-58 

 1865-69 

 1877-81 

 1888-92 



Cool. 



4 

 3 

 3 

 3 

 4 



II 14 I I 17 8 



Comparing E and F with C and D, we find general agree- 

 ment of the two former (E and C), the exceptional group, 

 1882-86, remaining,^ while table F gets rid of the exceptions 

 of D. In fact, while the summer seasons 1854, 1855, 1856, 

 and 1867 had more than the average number of hot days, the 

 mean temperature of May to August was, in each year, under 

 the average. 



1 It is right to say that this criterion would make the neutral case in- 

 B an exception. 



2_ In the curve of hot days, there is evidently a long wave of variation,, 

 which may complicate matters. 



3 The sunspot maximum of 1884, I may point out, was abnormally low. 



