262 On the Preservation of Iron Ships, fife. [Apr. 24, 



of the rivet-Tioles are rounded, so as to diminish their cutting action. The 

 mean of three experiments on joints with rounded holes, compared with 

 the mean of three experiments on joints with unrounded punched holes, 

 and with the mean of three experiments on joints with unrounded 

 drilled holes, gives the following results : — 



Shearing resistance 

 of rivet. 



Eounded holes 21*517 tons per sq. in. 



Unrounded holes I f^ff A f'ff » » 



I drilled 19-228 



This shows that the joint with rounded holes is, as regards the resistance 

 of the rivet, 12 per cent, stronger than the joint with drilled holes, but 

 only 2| per cent, stronger than the joint with punched holes. 



As regards the apparent superiority of hand-riveting as compared 

 with machine-riveting, the author observes that this is due to the fact 

 that in hand-riveted joints the rivet is slightly hardened by being 

 hammered after it is cold. In small experimental joints hand-riveting 

 has therefore a slight advantage in strength ; but the defects of hand- 

 riveting are not likely to be exhibited in experiments on a small scale. 



The experiments seem conclusive as to the inferiority of joints with 

 drilled holes ; and the author reiterates his conviction that the force 

 used in punching plates is a valuable practical test of their quality, and 

 tends to prevent the use of inferior iron. 



Appendix I. 



In this Appendix are given the results of Colonel Inglis's experiments 

 on joints with rivet-holes the edges of which are rounded. These expe- 

 riments agree with the author's in. most respects ; they show the supe- 

 riority of punched to drilled holes, but the difference of the strength of 

 the rivets in Colonel Inglis's experiments is only 2 j per cent. On the 

 other hand, in Colonel Inglis's experiments the superiority of the joints 

 with rounded holes is much greater than in the author's, amounting on 

 the average to 29 per cent. 



Appendix II. 



On the Theory of the Resistance of Riveted Joints. 



In this Appendix the ordinary theory of the resistance of riveted 

 joints is discussed, and some points in which it is defective are pointed 

 out. It is shown that a riveted joint may give way either : — (1) by the 

 tearing of the plates from the rivet-hole to the edge of the plates; 

 (2) by the tearing of the plates from rivet-hole to rivet-hole ; (3) by 

 the shearing of the rivet. When the plate gives way by tearing from 

 the rivet-hole to the edge of the plate, a bending stress is induced in the 



