FOR MEASURING TENSILE STRENGTHS. 393 



secured at each end between gummed paper_, which was 

 fastened to fine wire triangles ; a triangle was hung upon 

 the fixed support, and the other brought on to the hook of 

 the floating cylinder by means of an adjustment on the 

 fixed support (not shown in the sketch) ; the fibre was drawn 

 until nearly tight ; water was then slowly drawn ofi^ until 

 the fibre was observed to be quite straight and taut ; 

 then the lever was set and noted, and water slowly run off 

 into a measuring-vessel until the fibre broke. The long 

 arm of the lever had moved over if division of the scale, 

 = 0*15 inch; but, as the lever multiplies six times, the 

 actual stretch or depression of the cylinder before breaking 

 was only 0*025 inch. The quantity of water drawn off was 

 found to be 3300 grains, from which must be deducted 62 

 grains for the stretch, leaving 3238 grains as the breaking- 

 quantity; this, being divided by 21*09, gives I53'5 grs. as 

 the actual breaking-weight for this fibre. 



It will be observed that this apparatus has two special 

 features — ^namely, the very gradual manner in which the 

 tension can be applied without jerks or shocks, and the 

 extreme sensibility of which it is capable. The smaller 

 floating cylinder can easily put on a measured strain of 

 0*002 grain ; and with larger water- vessels or smaller float- 

 ing cylinders there is hardly any assignable limit to its 

 delicacy. Of course great care and numerous precautions 

 are necessary when measuring strains so small. 



On the other hand, the apparatus may be enlarged to a 

 size that could put on a strain of a hundred tons, and could 

 perhaps be applied to many of the cases of testing wires, 

 chains, or castings. 



I believe that this apparatus may be advantageously 

 employed in many experimental researches, where it is de- 

 sired to exercise a measured and constant strain. Hitherto 

 I have only employed it in ascertaining the elasticity and 

 tensile strength of textile fabrics, for which it seems spe- 



2 d2 



