1863.] 455 



Cast steel stood twenty heatings, but was very much cracked all over 

 its surface. As respects the change of form of cast iron and steel, the 

 result was similar to that in wrought iron, but not nearly so large in 

 amount. The cast iron did not return to its original dimensions, but 

 the smallest diameter was about one inch above the water-line. 



Tin showed no change of form, there being apparently no inter- 

 mediate state between the melting-point and absolute solidity. Brass, 

 gun-metal, and zinc showed the eifect slightly ; but instead of a con- 

 traction just above the water-line, there was an expansion or bulging. 



The effect on wrought iron is best seen in the solid cylinder 

 (figs. 9 and 10), where the displacement of particles just above the 

 water-line appears to be compensated by the bulgings at the two 

 extremities. 



The specimens of wrought iron were submitted by Mr. Abel (Che- 

 mist to the War Department) to chemical analysis, and he informs 

 me that he found nothing noteworthy in the composition of the 

 metal; nor was there any appreciable difference in the specific 

 gravity of the metal taken from different parts of the specimen. It 

 appears therefore to be simply a movement of the particles whilst 

 the metal is in a soft or semifluid state. 



The following is an account of the experiments, which were 

 carried out under the superintendence of Mr. Butter, Draughtsman 

 of the Royal Carriage Department, to whom also I am indebted for 

 the accompanying diagrams. The exact dimensions of each specimen 

 before and after heating are given in a tabulated form at the end of 

 the paper, to facilitate comparison. 



In figs. 22 and 23 the changes in form of the 9" cylinders (one 

 immersed one-half, the other two-thirds its depth) are shown in 

 section after every five heatings (half the full size). 



Experiment. 1 A 4 ft. 2 in. hoop-tire of 3 inches breadth and f ths 

 inch in thickness (fig. 1) was heated and cooled by being immersed 

 to half its depth in cold water five times, by which the effect shown 

 in fig. 2 was produced. 



