82 On the Production of Steam from heated Iron. 



of redness, numerous other facts had convinced me that the approach 

 to this period is greatly accelerated by the adhesion of any non-con- 

 ducting substance to the surface of the iron. Indeed, it often appear- 

 ed sufficient for the water to find and seize upon a mere point of such 

 material as a nucleus, to enable the fluid speedily to reduce the tem- 

 perature of the surrounding surface. By detaching a scale of oxide, 

 around which the effect just described had begun to take place, I have 

 sometimes succeeded in arresting the progress of vaporization, and 

 by giving the liquid once more a clean red surface, even with the 

 scale floating loosely in the w^ater, to establish once more the slow 

 evaporation which belongs to that state of the metal. 



To ascertain what effect the incrustation generally formed upon 

 the interior of a steam boiler might be expected to produce, in aug- 

 menting the rapidity of action in a case of overheating, I performed 

 the following course of nine series, employing for that purpose, the 

 basin used in the first course, commencing with its surface clean, and 

 having tried the effect of pure water at 212°, subsequently poured 

 in a portion of cold water, into a pint of which about two ounces' of 

 clayey garden earth had been put, producing a degree of turbidness 

 as great probably as any of our rivers possess in the time of freshets. 

 The iron was kept constantly over a brisk fire, and, in some of the 

 series, was permitted to come to bright redness before each experi- 

 ment ; while in others, the operation commenced with redness, but 

 was continued in so immediate a succession, as to reduce the metal 

 to a certain point of constant action 5 but never attaining the most 

 rapid period. 



It will be perceived that the first series w^as made in pairs, alter- 

 nately — two with clean water at the boiling point and two with the mud- 

 dy water above mentioned. The other series were made with similar 

 alternations of single experiments, with the exception that both hot 

 and cold water were free from impurities when laid upon the metal. 

 The ratios placed among the results of this course, will prove that 

 on an average, water at 212° laid upon hot metal under the circum- 

 stances described, requires 15|-^ per cent, longer for its evaporation 

 than a like quantity of water at 60°. This result, which appears at 

 first rather startling and paradoxical, is readily explained when we 

 consider the efficacy of cold water in bringing the coating and even 

 the surface of the metal down towards the temperature of most rapid 

 action, — a point, at which the mere difference of temperature be- 

 comes an insignificant element in the calculation, compared with the 

 vastly augmented speed with which the vapor is then generated. 



