5G ANNUAL OF SCIENTIFIC DISCOVERY. 



dition, lie is disposed to suggest that the rate of evaporation of water per 

 pound of coal increases with, and bears some ratio to, increase of pressure. 



With regard to the deposition of sulphate and carbonate of lime and mud 

 in boilers, Mr. Graham stated that he had experimented, with more of less 

 success, Avith caustic soda, quick-lime, muriatic acid, soap liquor, sawdust, 

 spent madder, and logwood chips. Two facts in particular were noticed as 

 regards the tendency of hard water to "scale": 1. That the sulphate of 

 lime separates from the water when in contact with the bottom of the boiler, 

 or with other substances, such as sawdust or other materials floating in the 

 water; but that no precipitation takes place until the water has been concen- 

 trated, by continued evaporation, down to the state of a saturated solution, 

 or to that point which maybe termed the "salting point." 2. That carbon- 

 ate of lime and mud are principally liberated in the body of the water, and 

 have but little disposition to adhere to the boiler, unless cemented by the 

 sulphate of lime. 



Practically, therefore, it has been found that no scale of any consequence 

 will be produced on engine boilers, even Avith such hard water and hard tiring 

 as Mr. Graham has been accustomed to, if 100 gallons of the concentrated 

 liquor of the boiler, equal to 4 per cent, of the amount of feed-water used 

 daily, and 300 gallons, or 12 per cent., be run away on Saturday through the 

 usual mud machine, and if the boiler be run empty every sixth Satur- 

 day and brushed out. The water used w r as so hard as to require from 35 to 

 40 measures of Clark's test liquid to soften it. There is little loss incurred by 

 this mode of working, since the chief discharge may take place at the close 

 of each day's work ; and there is an incredible advantage gained by the sav- 

 ing of coal, the reduced wear and tear of the boiler, and the greater safety 

 of all persons concerned with it. 



OX THE RESISTANCE OF TUBES TO COLLAPSE. 



It has long been a desideratum in the strength of boilers to determine 

 some definite law by which the engineer could calculate the proportionate 

 strength of internal flues. Ever since boilers became a necessary appendage 

 to the steam-engine, we have acted upon the principle that the internal 

 cylindrical flues subjected to compression were absolutely stronger than the 

 outer shell opposed to tension. These opinions have, in reality, had no 

 foundation in practice, excepting from deductions drawn from occasional 

 explosions, and the failure of vessels under severe pressure. Hitherto, there 

 has been nothing definite, or any known principle by which we could calcu- 

 late the diameter, thickness of plates, or length of flues corresponding with 

 the strength of the boiler; and even in cases where explosions have taken 

 place in collapse, we have, too frequently, mistaken the original cause from 

 the debris surrounding the rupture, and the force which has torn to pieces 

 the scattered remnants of the outer shell. Numerous accidents of this kind 

 have occurred, accompanied by serious loss of life; these haA r e too frequently 

 been caused by the collapse and the rupture of the internal flues, which, 

 acting upon the interior of the boiler with an irresistible force, carries havoc 

 and destruction before it. The relative position and comparative value of 

 these resisting forces have never as yet been clearly ascertained, in so far as 

 respects the cause of rupture, and the anomalous condition in which many 

 of these constructions are affected, have greatly retarded the application of 



