318 CRITICAL NOTICES OF NEW PUBLICATIONS. 



before the explosion. The sheet of copper was torn from the heads, un. 



rolled, and irregularly bent, adhering to the heads for only a short distance 

 near the top of each ; and the heads were bent outwards. The thickness of 

 the copper along the line of rupture varies from 0.25 to 0.35 of an inch, and 

 the metal appears to have been highly heated at one end of the torn por- 

 tion." 



To a gradual increase of pressure beyond that which the material 

 of the boiler would bear, the Committee refer certain well-known ac- 

 cidents, which are upon record ; as that in Wellclose Square, London, 

 that of the steamer, Rhone, &c. 



The safety-valve being the means commonly relied on to prevent a 

 dangerous accumulation of steam within a boiler, the Committee thus 

 enumerate the causes which may render it inoperative : — Design on 

 the part of the engineer or attendant, and adhesion of the valve to its 

 seat by rust or other cause. This adhesion, the Committee consider, 

 cannot be traced to the peculiar pTienomenon observed when a disk is 

 placed before an aperture, whence a fluid is rapidly issuing. This 

 phenomenon has been the subject of much examination. The first 

 observation of it is claimed by M. Clement, of Paris, and by Mr. 

 Roberts, of Manchester. Theory and experiment both shew it to 

 depend upon the relative dimensions of the disk and aperture ; and 

 as these are not very unequal in the case of the safety-valve, the ten- 

 dency of the disk towards the opening is inconsiderable. This de- 

 duction is further confirmed by the experiments of the Committee, 

 who used a graduated safety-valve, and compared its indications of 

 pressure with those of a mercurial gauge, or with those deduced from 

 the observed temperature of the water in the boiler. Two difierent 

 valves were used in different parts of the series of experiments, both 



