ON BALANCES. 63 



For all weighings requiring special accuracy, the highest and 

 lowest points reached by the needle in successive oscillations 

 of the "balance are read on the index scale. In each case the 

 mean between the highest and lowest readings is noted as the 

 resting point of the balance. 



Balances of precision are always enclosed in glass cases, 

 with a view both to their preservation, and more especially to 

 keep their "action in weighings as far as possible from being 

 influenced by draughts of air and alternations of temperature, 

 which would affect the accuracy of the results. 



19. Amongst the balances of the Standards Department 

 there is, however, one balance of peculiar construction and of 

 extreme delicacy, to which I should now call your attention. 

 It was used by Professor Miller for all his weighings during 

 the construction of the new Imperial Standard pound, in- 

 cluding his weighings of a kilogram. It has been lent to 

 the International Metric Commission at Paris for the weigh- 

 ings of the new kilograms, or it would have been sent here 

 for exhibition as a scientific balance of considerable historical 

 interest. This balance was constructed by Barrow, and is 

 similar in construction to Robinson's balances. The distance 

 between the extreme knife-edges is 15*06 inches. The knife- 

 edges work upon quartz planes. The middle knife-edge is 

 1*93 inch long. Index scales marked upon thin and nearly 

 transparent slips of ivory, a little more than half-an-inch long, 

 are fixed to each side of the beam and oscillate with it. There 

 are 50 divisions of the scale, O'Ol inch apart. The scale is 

 viewed through a compound microscope, fixed in the glass 

 case of the balance, and having a single horizontal cobweb in 

 the focus of the eyepiece. A glass screen is interposed 

 between the observer and the front of the balance case. The 

 mean value of one division of the scale was found to be 

 about 0'002 grain with 1 Ib. in each pan, and 0*005 grain 

 with a kilogram in each pan. All the results of his com- 

 parisons were noted by Prof. Miller in hundredths of a division 

 of the scale. 



20. In describing the mode of observing the oscillations of 

 a balance of precision, I should not omit also to call attention 

 to the excellent method used by the late Dr. Steinheil of 

 Munich, through the adoption of a principle originated by 

 Gauss. The movements of the beam are observed by means 

 of a small mirror placed immediately over the central knife- 



