ON BALANCES. 73 



avoirdupois iron weights, of the same apparent bulk. But 

 one only is a just weight. The second and third are 

 fraudulent weights seized from itinerant dealers ; they 

 have had their insides bored out, and filled with matter of 

 less density, one with cork blackened at the bottom, 

 the other containing only air. The result is that the first 

 weighs exactly 1 Ib. or 16 oz., the second 7f oz., and the 

 third 7 J oz. 



39. The work of computing the weight of air displaced by 

 standard weights compared is to be avoided by weighing them 

 in a balance placed in a vacuum. The vacuum balance of 

 the Standards is a large and costly instrument of high scientific 

 character, and could not well be moved to this Loan Exhibi- 

 tion. I have here a drawing of it and a full description 

 as shown in Appendix VI II. to my Seventh Annual Report. 

 But the small Mendeleef balance is, as I have already stated, 

 intended to be made available as a vacuum balance ; and 

 although it is a new and untried instrument, we have en- 

 deavoured to get it sufficiently ready to show you the action 

 of a vacuum balance. The bell- shaped cover of the balance, 

 which you now see, is merely a temporary one. It is far 

 too large, and it is proposed to have one of a different form, 

 and made as close as possible to the balance. 



40. A further practical illustration of the effect of the 

 density of a weight when compared with another weight of 

 different density will now be exhibited to you, by comparing 

 first in air and then in a vacuum, or something near it,~the 

 platinum-iridium Ib. P-i, with the quartz standard Ib. of the 

 Standards Department, designated as Q. 



P-i was intended to be of the true weight of 1 Ib. in a 

 vacuum. Its actual weight in a vacuum, as deduced from 

 the mean result of all the comparisons with PS, has been 

 determined to be '01691 gr. in deficiency, and, as already 

 shown, it displaces nearly 0*4 gr. of standard air. Its 

 apparent weight in air is therefore about 0'4 gr. less than 

 its weight in a vacuum. 



Q was constructed to be as nearly as possible of the 

 weight of a brass Ib. in air. The commercial standard Ib. of 

 brass, which is a theoretical Ib. of the average density of 

 brass, displaces a little more than 1 gr. of standard air, 

 whilst Q displaces a little more than 3'2gr. Q was there- 

 fore constructed to be in a vacuum about 2*2 gr. heavier 



