8 APPENDIX. 



intensity of the draught, which they supposed to cause incrustation by burning the material. 

 A tolerable draught appears however to be preferable, and not injurious to a boiler, being 

 always under control ; moreover, I have invariably observed, that boilers with insufficient 

 draught expend more coals, with a considerable addition of labour, than others adapted to 

 engines of equal power where the draught is capable of extracting the whole of the combus- 

 tible matter of the coal, and vitrifying the residue ; for a great waste attends the former, from 

 the constant necessity of clearing the bars. I am also of opinion, that no serious deposit can 

 be formed, if change of water is duly attended to and the evaporation constantly supplied ; 

 and that the sound plates will not experience injury if water alone is in immediate contact 

 and the essentials of construction properly observed : for when the fires are in a state of 

 active combustion, the iron or copper rapidly absorbs the caloric from the fuel, and as 

 speedily transfers it to the water, which is accordingly in a greater or less state of ebullition, 

 and not of incrustation, the temperature of the plates not very much exceeding that of the 

 water. 



18. In iron boilers it is difficult to insure the duration of the plates, for various reasons. 

 Boiler plates are often on inspection apparently sound and of good quality ; but when 

 submitted to the heat of the furnaces, films of the metal opposed to the fire, where adhe- 

 sion was not perfect, become separated in the form of blisters or cracks, (according to 

 the nature and situation of the latent defects,) which are on examination frequently found to 

 exist through the whole plate, to an extent often exceeding- half its thickness. On testing 

 plates of various kinds for experiment, by heat, punching, breaking, &c., those of the most 

 perfect exterior were often found to be much laminated ; and those parts of boilers which 

 have been previously heated and bent to form the bridges, and on which (when con- 

 structed) the tire incessantly impinges, will, if defective, most readily afford opportunity of 

 detection. I have known boilers to run for- some months on occasional service in the 

 Channel during the existence of defects, which did not exhibit themselves till the plates 

 were overheated ; no doubt, through partial incrustation, experienced on a voyage to the 

 Mediterranean, immediately succeeding. 



19. Instances seldom occur of copper plates cracking, unless from causes already 

 described ; a material difference existing between the manufacture of copper and iron plate, 

 which renders it incompatible that they should display similar effects from the cause of heat 

 alone. The former, after certain refining processes, is cast in open iron vessels, forming con- 

 venient cakes, and of course solid and free from air and extraneous matters, prior to being 

 heated and rolled ; while the latter, in masses prepared from scraps, or slaps, is raised to a 

 welding heat, with its heterogeneous accompaniments, then submitted to pressure under the 

 rolls; on the judicious observance of which processes a perfect union of the parts and con- 

 sequent solidity mainly depends. 



