296 



NA TURE 



[January 6, 1910 



provided for agricultural departments in the graded scliools 

 of the State. 



Referring to American colleges and universities. Dr. 

 Brown gives it as his opinion that among the leaders of 

 American university education there is a growing and sur- 

 prisingly unanimous conviction regarding the directions 

 in which improvement should be made in higher education 

 in the States. It is to be rendered more coherent, vital 

 and democratic. As President Butler has remarked :— 



The American college is under fire, no doubt. Well- 

 directed intelligent tiring will do it good. It is far from 

 perfect, but it knows its job, and is working at it with 

 the skill born of long and successful experience." The 

 democratic movement in higher education has been 

 emphas.sed during the year bv the effort to -organise in 

 -Massachusetts an institution which shall bring courses of 

 college mstruction home to all communities in the State 

 in which it may be desired— a project which has been re- 

 ferred by the Legislature to the new State Board of Educa- 

 tion for an opinion as to its advisabilitv ■ bv the step 

 taken by Cornell Universitv in the direction of the State 

 university form of ort?anisation, and bv the beginnings 

 at the University of Wisconsin of a more comprehensive 

 and widely diffused svstem of universitv extension 



Within the year the Universitv of "Wisconsin has been 

 a centre of public interest in a varietv of wavs, not the 

 least important of which is the Vilas' bequest, which is 

 expected to amount to 40o,oooZ., and to be administered so 

 that It shall eventually reach a total of 4,ooo,oooZ. The 

 act that the available income of this fund is to be devoted 

 largely to research renders it an epoch-making endow- 

 ment. ^ 



In addition to its treatment of the problems of American 

 education, the volume provides an admirable series of 

 summaries of educational progress in European and other 



Zblem!' •''h"''"-'~^-''u''"'''U ''"■'= ^"'"'^^^ t° educational 

 problems in Hawau, the Philippines. Porto Riro, the 

 -Argentine Republic and Chile. Great Britain and Ireland 

 France and Central Europe. Educational reform in CWnn 

 andcurren educational topics in foreign countries each 

 receive special treatment. 



These educational reports from Washin.cton have often 



?haT th?'r. r '''''' r,',""^"'- ^"'^ '^ ^•'" ^^^-^^ ^ ^^--^ 



mat the latest report fully maintains the excellence of its 

 predecessors. 



OLE ROMER AA^D THE THER.VfOAfETER.^ 

 T"P ,'''■'' thermometers of which the indications were 

 independent of atmospheric pressure appeared in the 

 latter half of the seventeenth centurv, but Fahrenheit was 

 the first one to succeed, in 1710, in solving the problem 

 of furnishing these thermometers with .such scales that 

 their indications agreed; these thermometers were much 

 admired and represented great progress. It mav therefore 

 be of interest to_ show th.at Ole Rbmer solved this problem 

 before Fahrenheit, and that it was from him that Fahren- 

 heit obtained his method. 



^J'T- """•" f.'T^'>',''<="i-'"ks which I happened to come 

 across in scientific literature of the eighteenth centurv I 

 c^st"'';- «'%I*r- P^bably occupied himself with" ih 

 con.struction of thermometers, and that some connection 

 hTL % T", ^T ■^'"^ Fahrenheit. These statements 

 work fn^h °.f '"4"'^'"S me to look for traces of Romer's 

 J^ork in the librar.es and archives here in Copenhagen 

 In the universitv library I found what I was looking for- 

 IZ ■'■ ¥'^/':. ^f'l'^d "Adversaria," a volume of 



uritten papers m foho bound in a brown cover ' 



The book contains a whole section about the thermo- 

 meter, besides some scattered statements about temp»ra- 

 ture measurements, which I shall return to later The 

 arrangement of Romer's thermometer seems to me to be 

 of considerable interest. Romer apoears to have been the 

 first to construct thermometers with the two fixed points, 

 the temperature of melting snow—" Nix sine gelu et 



„'' >'\^lj^'lP',.^''''''^^:- '^"■"Vers.w'hc^Khn-i Udvikline eenn^m Tiderne 

 og dels Forhold til vekslende Anskuelser om Varmen. Natur Giel erum 

 Eoshandd. Inaueural nlssertation. (Copenhagen, ,qo,.) ^J''I'='-"P' 



li.hed In .T^^'' T ^^""^"'^ " '' P^'l<^y '■••"in : the book will he puh- 

 SeUkab auspice? of Ihe Kgl, danske Videnskabernes 



NO. 2097, VOL. 82] 



calore " — and the boiling point of water, and with the 

 cubic contents of the tube divided into equal parts. Both 

 Rumer and Horrebow's remarks seem to indicate that this 

 took place about the year 1702. The first part of this 

 section is mathematical, and deals chiefly with the problem 

 of dividing the cubic contents of a conical glass tube into 

 equal parts. Rbmer finds a general method of making 

 such a division, and calculates approximate formulas by 

 the aid of which he may carry out more easily his calcu- 

 lations ; he employs these formulas in dividing the cubic 

 contents of a conical tube 8 inches long, intended for his 

 "original thermometer," into four equal parts, and he 

 gives the length of these parts when he determines that 

 the scale of the thermometer is to have sixtv divisions, 

 and these are to be arranged in such a way as to read 

 " boiling 60, snow without cold or warmth 7J." After 

 these preliminary investigations Riimer gives complete in- 

 structions in four paragraphs for " the construction of an 

 originaP thermometer." 



"(i) By means of a drop of n]ercur\' investigate whether 

 the cavity of the tube, be it cylindrical or conical, is 

 regular before the ball is blown out. Irregular forms are 

 to be rejected ; the cylindrical form may be employed with- 

 out further investigation. With regard to the conical 

 forms, proceed as follows : — 



" (2) From the middle of the tube towards the outer 

 points take the lengths of the drop of mercury. 



" (3) When by means of this experiment the divisions 

 have been divided into two equal parts, each of these parts 

 is in turn divided into two equal parts proportionally by 

 increase or diminution, and the whole tube will thus be 

 divided into four equal parts. 



" (4) When the thermometer is completed, filled and 

 closed, fis. by means of snow or crushed ice the point of 

 division y\, by means of boiling the point 60." 



.■\fter these instructions there are remarks written in 

 Horrebow's hand and with his signature which are sup- 

 plementary, and show also that Romer's thermometer 

 existed after his death (1710) : — " ... In 1730. Romer's 

 widow sent me five glas.ses for thermometers w'hich Romer 

 himself had filled and divided with two points in accord- 

 ance with his own rules given above. The alcohol in them 

 is rather pale, although Romer coloured it with saffron in 

 the usual manner. . . . .After this was written, 1 asked 

 Romer's widow if she knew whether Romer, after Ihacl 

 left his observatories, had made any change in his thermo- 

 meter. She said that she did not know, but she gave me 

 Romer's vade nieciim, in which I found a loose sheet, 

 which is pasted in here after the next sheet. On that T 

 read that Rcimer fi.\ed upon 8 as the dividing-point for 

 snow, and thus, so far as w^e know, the alcohol never 

 sinks below o in Copenhagen, and it is to be remarked 

 that January 7, 1709, the alcohol only sank to y,",,." 



The loose sheet which Horrebow mentions contains a 

 table of temperatures which gives the temperature for every 

 day from December 26, 1708, to .'\pril i, 1709. 



The two following pages contain a sort of table of 

 corrections for the four divisions. 



.After this short account of the contents of the eleven 

 folio pages which Romer devoted to the construction of 

 his " new " thermometer, it will be appropriate here to 

 give a short explanation of his method and to point out 

 what is new in it. 



The chief feature of the method is this : to base the 

 division of the thermometer on two fixed points, the melt- 

 ing point of thawing snow and the temperature of boiling 

 water, and to find the length of the degree by dividing 

 the cubic contents of the thermometer tube between these 

 two points into equal parts, taking into consideration 

 whether the tube is cylindrical or not. The size of the 

 degree is obtained on the basis of the fact that there must 

 be between the freezing point and the boiling point 52.5 

 degrees of equal cubic content. If the tube is cvlindrica'l 

 the whole length between the two fixed points is divided 

 into 52i equal parts, and 7-5 similar parts are added 



1 .Some weights which are still in existence from Romer's time, and probably 

 are those that he constructed as standards for the new system of weights and 

 tneasures intro'inced by the Act of May i. 1683, bear the inscription 

 "original weight." From this it may be inferred that "original ihermo- 

 metel- " means " standard thermometer," and that it was Romer's purpose to 

 introduce a standard for thermometers as for other unit: 



