TRANSACTIONS OF SKCTION A. 541 



Thus in the case of six thermometers, on three of which the measurements 

 necessary for the calibrations were made at Kew, and on three at the Owens Oolle"-e, 

 by Messrs. G. Baker and M. Ilirakoa respectively, the difference between any o'lie 

 measurement of the length of a mercurial thread expressed in terms of the corrected 

 scale and the mean length of that thread equals or exceeds 0'01°C. (i.e. about 0"1 m.m.) 

 in eleven only out of a total of b80 measurements. 



The Kew thermometers had been calibrated, as is usual at that observatory, by 

 Welsh's methodCRep. Brit. Ass.,' 1853, p. 34). Mr. G. Baker afterwards made, 

 in accordance with the author's instructions, the measurements necessary for the 

 application of Bessel's method, the calculation being undertaken by the authors. 



It was a matter of some interest to determine in this case the magnitude of the 

 errors of the original calibration. 



Welsh's method is undoubtedly open to the objection that the errors are addi- 

 tive and Von QSttingen (foe. cj;!. p. 49) seems to consider that it must always be 

 necessary, for accurate work, to correct further a thermometer calibrated by it. 



The three thermometers were numbered 561-2-3, they read from 11 -.5° to 29° 0. 

 from 19-5° to 68°, and from 51° to 107° respectively. The average lengths of a 

 degree were 12-9 m.m., 11'2 m.m., and 9-5 m.m. respectively. The application of 

 Bessel's method to the first was made by 1 10 measurements of the mercurial threads 

 to the second by 163, and to the third by 185. 



The original calibration was so accurate that the second approximation of 

 Bessel's method was unnecessary in two cases, and was only partly carried out in 

 the third. 



The maximum positive and negative corrections were in the case of 



Th. 561 + 0-004° 0. and - 0-004° C. 

 „ 562 +0-012° „ -0-005° 

 „ 663 +0-008° „ -0-Oir 



As will be seen from the above description of the thermometers, the larger of 

 these quantities are about equal to the limit of certainty in readino-. 



In no case would the calibration error in the determination of a difference of 

 temperature have amounted to 0-02° C. It may therefore be concluded that Welsh's 

 method, as applied at Kew to selected tubes, and with a measuring instrument of 

 :great accuracy, is capable of giving first- rate results. The errors which remain 

 when it has been applied are so small that they may be neglected in all cases 

 but those where the thermometers are to be used under the most favourable con- 

 ditions, i.e. with the stem at the same temperature as the bulb, &c. 



This satisfactory conclusion is confirmed by the fact that Professor Rowland 

 has recently stated that the calibration of the Kew thermometer used by him in his 

 research on the Mechanical Equivalent of Heat was practically perfect. 



8. On the General Coincidence between Sun-spot Activity and Terrestrial 

 Magnetic Disturbance. By the Rev. F. Howlett, F.B.A.8. 



The object of this paper was to inquire how far solar activity, more especially 

 as regards sun-spots, is wont to be accompanied by terrestrial magnetic disturbances 

 ;a8 recorded by the automatic magnetic declination cui-ves at Kew and Greenwich, 



The data for such an investigation were furnished by comparisons instituted 

 between the more striking instances of sun-spots gathered out of a long series of 

 solar observations carried on by Mr. Howlett from 1850 to the present epoch, and 

 the synchronous conditions of the magnetic curves at the observatories above men- 

 tioned. The telescopic drawings of the spots were obtained with an achromatic of 

 three inches' aperture, by Dollond, and of forty-eight _' 2hes' focal distance, project- 

 ing the sun's image on a large white screen in a darkened chamber. 



By employing a Huygenian eyepiece magnifying 120 linear, and placing the 

 screen at the distance of five feet, two inches from the eyepiece, a beautiful image 

 of the sun was obtained, five feet four inches in diameter, and of which every inch 

 corresponded to just 30" of the celestial arc. 



Not only were the measurements of all the solar phenomena rendered thereby 



