29 



2nd. The given ratio of the density of the air to the maximum 

 density of water. 



3rd. The ratio of the density of the metal of which the 

 standard is composed at 0° to its density at the given 

 temperature, and co-efficient of the lineal expansion of 

 the metal. 



4th. The weight of the standard in grains, and by deducting 

 from the sum the loss. 



5th. The ratio of the density of the standard at 0° to the 

 maximum density of water. 



In these calculations the latitude of the place of observation 

 together with its elevation above normal sea level have to be 

 taken into account. 



In order to observe the needle of the balance with an 

 accuracy commensurate with such corrections as I have des- 

 cribed, it has been found necessary to use microscopes. 



To the balance known in the department as Barrow's Balance 

 two microscopes with a single horizontal hair line have been 

 attached, so that two persons can simultaneously observe 

 each needle. Provision is made for illuminating the scale if 

 required. A double glass screen is placed between the obser- 

 vers and the balance, and an arrangement is provided for 

 shifting the pans from one arm to the other without opening 

 the glass case in which the balance is contained. A holder is 

 fixed to carry two thermometers, so as to adjust the bulbs to 

 the level of the centres of gravity of the weights in each pan. 



A vacuum balance to weigh up to one kilogramme (made by 

 Oertling) was shown to me by Mr. Chaney, of which the draw- 

 ing in Plate XL is a sketch from memory. This, I understand, is 

 only used occasionally as a test after the process above described 

 has been effected. It is not in common use. A powerful air-pump, 

 constructed by Troughton & Simms, was used in connection 

 with this balance, and almost a perfect vacuum could be ob- 

 tained. The glass is one and three-quarter inches thick. The 

 balance is provided with thermometers, gauge, and arms for 

 removing and changing the pans and removing the weights. 

 These are not made air-tight by stuffing boxes, but are sealed 

 by a 30-in. seal of mercury, which hangs down underneath the 

 table. Very minute weights can be observed with this instru- 

 ment. I was also shown a small scales weighing down to the 

 '0001 of a grain, which is to carry weights up to one ounce. 



The beam is formed of filaments as fine as hairs braced to- 

 gether for stiffness. 



Another balance was shown to me, enclosed, as is usual, in a 

 glass case ; the scale was read by a telescope, by which means 

 the observer could be removed twenty feet away from the instru- 

 ment to avoid disturbing the temperature by the heat of thebody. 



