384 REPORT— 1903. 



bearing surface is not reduced below the length corresponding to D, c in 

 fig. 2. 



A machine has been made, under the superintendence of the Committee, 

 by the Cambridge Scientific Instrument Company, for the accurate 

 measurement of screw gauges. A description of this is given in the 

 Appendix. The machine has been placed in charge of the Committee of 

 the National Physical Laboratory, and the Director of that Institution 

 is prepared to undertake the measurement of gauges and screws submitted 

 for examination. 



The Committee have further to report that the Engineering Standards 

 Committee have appointed an influential sub-committee to deal with the 

 standardisation of gauges of all kinds, including screw gauges, and that 

 in their opinion the work which they have been doing may with advantage 

 be left to this committee and to the National Physical Laboratory to 

 carry on. The Committee have not considered in detail the question of 

 the limits of error in sci'ews purporting to represent the B.A. thread. 

 This matter they think it desirable to leave to the Engineering Standards 

 Committee, who will be able to discuss it in connection with larger screws 

 and gauges. On the other matters submitted to them they do not wish to 

 repoi't further. In consequence they present this as their final report, 

 and do not ask for reappointment. 



We, the undersigned members of the Small Screw Gauge Committee, 

 do not accept in its entirety the above report, as we consider that some of 

 the recommendations contained in it are not those held by the whole of 

 the Committee. 



1. The report appears to explicitly restore without modification the 

 form of the original B.A. thread as defined in the Montreal Report. 



2. It provides gauges for testing threads of a form differing from those 

 laid down in the earlier part of the same I'eport. 



The work of the first three years of this Committee showed that the 

 difficulty in constructing tools and gauges of the B.A. thread was at the 

 root of the inaccuracy complained of in commercial screws. On that 

 ground the Committee asked for an extended reference, and recommended 

 a new form of thread. The construction of taps for the old thread re- 

 quires exceptional manipulative skill, and does not admit of great exact- 

 ness. An average tool maker can produce taps of the new thread without 

 difficulty, and the process admits of extreme refinement. Moreover, since 

 the manufacture both of gauges and of screwing tools alike depends upon 

 the construction of accurate taps, the adoption of a form of thread which 

 is easily produced met at once the difficulties of the screwing tools, and 

 of gauges or trial pieces for testing screws when made. 



The present report from which we dissent confirms the recommenda- 

 tion of this form of thread for the gauges, but withdraws it from screws, 

 notwithstanding that on account of the number used the simplification in 

 the construction of taps is more valuable in the case of those for screwing 

 tools than in those for gauges. 



Although we admit the fact that screws made in dies or in screwing 



