The next wheel may have an axle worn little or none. (See Fig. 28) Now, if these brasses are exchanged, we have the conditions as shown in Figs. 29 and 30, and we must expect they will heat. The remedy was simply to keep each brass upon its own journal. To do this the brasses were fastened to the axle by a piece of small wire, and went with it to the lathe and press. When its truck was reached, the brass was there with its journal. Worn-out brasses, of course, could not be put in, and new ones were substituted. The little trouble from that source that followed the change showed the efficacy of the remedy. [Illustration: FIG. 29 and FIG. 30] The manner in which the tires of engines were to be changed, when the final day came, was a serious question. The old-fashioned fire upon the ground could not be thought of. The M. & O. had used a fire of pine under the wheel, which was covered by a box of sheet iron, so arranged that the flame and heat would be conveyed around the tire, and out at an aperture at the top. (Fig. 31.) Many thought this perfect, while others were not satisfied, and began experiments for something better. A device for using gas had been patented, but it was somewhat complicated, as well as expensive, and did not meet with general favor. A very simple device was soon hit upon. A two inch pipe was bent around in a circle a little larger than the outer rim of the wheel. Holes 1/10 in. in diameter and 3 or 4 in. apart were drilled through the pipe on the inside of the circle. To this pipe was fastened another with a branch or fork upon it. To one branch or fork was connected a gas pipe from the meter, while to the other was connected a pipe from an air pump. With the ordinary pressure of city gas upon this pipe it was found that the air pump must keep an air pressure of 40 pounds, that the air and gas might mix properly at the branch or fork, so we could get the best combustion and most heat from our "blowpipe," for such it was. (Fig. 32.) [Illustration: FIG. 31 and FIG. 32] We were able to heat a tire so it could be moved in ten to twenty minutes, and the machine may be said to have been satisfactory. Gas, however, was not to be had at all places where it would be necessary to change tires, and the item of cost was considerable. To reach a result as good, if possible, experiments were begun with coal oil (headlight oil). They were crude and unsatisfactory at first, but soon success was reached. A pipe was bent to fit the lower half of a wheel pretty closely and then turned back under itself about the diameter of the pipe distant from it. This under part had holes 1/10 in. diameter and 3 or 4 in. apart drilled upon its upper side or under the upper pipe. Connected with the upper pipe at its center was a pipe which ran to one side and up to the can containing the kerosene. Between the can and the pipe under the wheel was a stop cock, by which the flow of oil could be controlled. [Illustration: FIG. 33]