AMERICAN INSTITUTE. 583 



MR. MEIGS ON TESTING STEAM BOILERS BEFORE USED 

 IN SEA OR LAND SERVICE. 



The explosive force so often developed in steam boilers resem- 

 bles generally that of gunpowder. Let the theory of its cause be 

 what it may — it is not the result, in all cases, of gradually in- 

 creasing power of steam, for it has been found that when that has 

 become superheated as in Frost's stame, and long before that in 

 Perkins' cylinder, under pressure it is flame quiescent. Perkins 

 made it red hot, &c.,&c. 



Now the expansive power of gunpowder appears to be more 

 uhder our control. But experience has demonstrated the indls- 

 pensible necessity of proof of even a gun before it is put into army 

 or navy, or sporting service. 



Hydraulic pressure is used in our steam boilers and is no proof 

 of capability of steam boilers or of cannon. Our Prof. Mapes, in 

 experimenting on -hydraulic pressure, found that in a properly 

 cast cylinder of iron, several inches thick, the water was forced 

 through the pores of the iron, and yet the cylinder did not ex- 

 plode. All the hydraulic power conceivable cannot act like some 

 of our steam boilers, be driven through walls across the Ohio, or 

 thrown up lUO feet in the air, as in a low pressure boiler some 

 years ago, &c., &c. I therefore propose that no boiler be allowed 

 for public service without having been fully tested by steam proof 

 to at least double the pressure ever to be actually used. And as 

 in gun proof, the time of service of every steam boiler shall be 

 limited as in the cannon. For it has been determined by experi- 

 ence that a cannon gradually becomes less strong by every dis- 

 charge. So that as a common rule, the cannon which has been 

 discharged a thousand times, is no longer safe. 



No other test can be found for boilers than steam itself, nor for 

 cannon than powder. Because every plate in any boiler has 

 unq.uestionably a different character, and while one pure plate 

 might bear perfectly 5U0 pounds per'square inch another would 

 burst at 100 pounds. And besides the wear of plates depends on 

 their position in the boiler and flues as to heat, some plates 

 nearest the hottest parts being very much more and sooner worn 

 than others. We ought to hav^e had, by this time, as good data 

 to determine the length of time boilers can be used as we have of 

 cannon. Experience has decided in the artillery service of Eng- 

 land, France, etc., that a cannon may be safely discharged about 

 one thousand times only. 



The successive strains upon the metal by discharges, like so 

 many percussions then have injured the cohesion of the atoms of 



