102 On a Proposed New Method of Weighing, 

 Then, for this method the assay weights consist of: — 



10 grains and its subdivisions in grains. 

 1 ,, „ „ in tenths of a grain. 



•1 „ „ ,, in hundredths of a grain. 



and a rider of platinum or gilt silver wire carried by a lever 

 which slides parallel to the upper edge of the beam, per- 

 mitting the rider to be placed on any point of the divided 

 beam, or to be lifted off the beam altogether. This rider, 

 according to its position on the beam, represents thousandths 

 of a grain from '001 to 100, or its full weight when placed 

 in the pan, as will be obvious when we consider that the 

 beam is divided into 100 parts on each arm, and that the 

 rider weighs one-tenth of a grain. But because the ten- 

 grain weight, the assayer's pound, is regarded as unity, 

 and all the smaller weights are regarded as decimal fractions 

 of uuity, they are marked, conformably to this view, as 

 follows : — 



The 10 grain weight or pound is marked 1* 



The grain subdivisions are marked respectively *6 S *2 and -1 



The subdivisions of the grain ,, ,, -06 -03 -02 and -01 



The rider, weighing one-tenth of a grain, and used on the 

 principle of the steelyard, by placing it on one or other of 

 the divisions of the beam, furnishes all subdivisions of unity 

 from -01 to 0001. 



In assay balances, as they are now made, a few sub- 

 divisions of the beam, namely, those nearest the outer knife- 

 edges, are wanting, a deficiency resulting from the particular 

 pattern of the beam ends. This incompleteness of the 

 divided beam is certainly not of an insurmountable character, 

 should complete subdivision of the beam be required ; and 

 that this complete subdivision is wanted for the routine 

 about to be proposed will presently become apparent. 



But besides the weights just described, a second series is 

 commonly used. Taking the unit (or pound) weight, repre- 

 senting each bulk or bar of gold, and submitting this 

 quantity to the operations of the assay, and thus obtaining, 

 for each sample, the pure gold obtained in it ; the weight of 

 this pure gold product as ascertained with the above- 

 described weights, expresses decimally the fineness of the 

 particular sample, and enables us to value the bar or parcel 

 of gold which it represents, as far as gold contents are 

 concerned. 



For expedition and accuracy it is found best to have, in 

 addition to the above-described series of weights, a second 



