METHODS OF INVESTIGATION 191 



recorded the strain on a drum. When the fibre gave 

 way the balance returned to rest, the drum moved on to 

 receive a fresh trace, the broken fibre with its cards was 

 thrown off , and, the clutch of the magazine being pulled 

 out at the same time, the cycle of operations started again. 



The set of cards in the magazine having been dealt 

 with, the paper was removed from the drum, and a fresh 

 magazine inserted, with a fresh batch of fibres. The 

 mean breaking strain was obtained quickly by adding up 

 the total deflections of the balance as marked on the 

 drum with a map-measurer, the whole operation with a 

 magazine of twenty fibres taking eight minutes, most of 

 which was simply spent in watching the machine do the 

 work (PL XVI.). 



Impact Testing. On leaving Egypt the author lost this 

 automatic tester, because some parts of it had been made 

 with Government material. The task of reconstructing it 

 was rather formidable, and a timely suggestion was 

 derived from a paper by Mr. J. H. Lester, in which he 

 points out the value of ballistic testing of yarn. The 

 chief feature, from the author's view-point, is that bunches 

 of yarns or fibres can be tested together. If several 

 fibres are slowly strained to determine the breaking 

 strain, the rupture of the first one throws its share of the 

 load on the others, which therefore yield in rapid succes- 

 sion, and the result is meaningless. 



In impact testing, on the other hand, the force measured 

 is kinetic, the breaking of each fibre subtracting a definite 

 amount of kinetic energy. 



The simple home-made form of the implement devised 



