In 1878 this machine was shown, as I have mentioned, at the Franklin Institute. It attracted the attention of Mr. Garrett, a typical Philadelphia Quaker, who was then the agent of the Brush Company, which was just beginning to put out some few arc lights. A few ma-chims were in use, giving two to four arc lights to the machine. I must explain here that these machines worked in this way: whenever you wanted a light you ran a separate circuit out and back to the machine. They were what we may call single-arc multi-circuit machines. One set of armature coils fed to this line, another this, and so on, and for four lights you had to have four dynamos in one. The idea of putting lights one after the other in series was not developed at that time. Mr. Garrett asked if we could get up a four-light machine that would run these single-circuit lights, and he was told that we would try. "Well, " he said, "if you want to try I will bear the expense and see how you come out." So I set to work at once. Evenings and whole nights were spent in getting information together and calculating out, as best I could, what that machine should be and how to build it in a small machine shop. I would not trust anybody to do the winding but myself. I personally wound the armature and the field, and did everything necessary of an electrical nature to ensure that nothing would go wrong.

It must be borne in mind that no armature winders were to be had in those days. It would have taken a great deal of superintendence to get any man to wind an armature correctly, or to do anything with insulated wire where the voltage was considerable. This machine was wound and run, and it gave four lights in four separate circuits. As the machine was being completed Mr. Garrett came in one day and said, "Can you not run those lights on one wire ? I hear that the Brush Company is doing it. " " Oh, yes, " I said, " I can. " We soon had them running on one wire, 20 amperes to each light. They were very large and beautiful arc lights, such as are not often seen nowadays. The lights ran all night in a bakery, and soon after a number were installed in a brewery, and so the business began. The bakery was fearfully hot in summer, and the temperature was about 140° in the room where the machine was running. We had to stay in it and so got baked as well as the bread and the arc machine. But somehow or other the machine stood up, and we stood up also.

Not long after the machine had been in operation another inquiry came from Mr. Garrett:-"Can you make that machine give half as much light per lamp and twice as many lamps, or can you put on more lamps and split the current up ?" I said we could, for I had thought it all out before, believing it was coming; I did not wish to push him, but waited for him to push me in that respect, but the machine was all ready so that the circuit could be divided and made into what we called an eight-light 10-ampere series arc machine. For years after that the 10-ampere arc circuit was the standard, not only for ourselves, but for the other arc-light companies, with but one or two exceptions. The above conveys a general idea of how the Thomson-Houston arc system was started.

Mr. Garrett began to build machines after the first model and to sell them. One went into a brewery. An instance of something which happened there will show the way in which those new things were regarded at the time. We had established a seven-arc-light machine in this brewery, the proprietor, a good friend of ours, being willing to stake his lighting on our success. We were able to give him what he wanted, so put in the proper machinery and lighted this brewery. One night, for some reason or other, the hay in the loft above the stable which was part of the building took fire. It was thought somebody had been smoking there. The engineer at once shut down the lights, but the proprietor said, "No ! No ! Keep the lights on; I want to get the horses out." But when the firemen came, being unused to such lights, they played the hose on them and could not put them out. They were astonished to find that the globes could be full of water and that the light was still burning under water. Thus they learned that the electric light was different from other lights. There is no doubt in my mind that the fact that the electric light was kept going during the fire saved the building, because it enabled the firemen to work, and the only damage was in the upper story, in the malt room. The lights were run for a week or so afterwards with the wire lying across the burned portion of the building, so to speak. The circuit was passed through this burned portion and still ran the lights. I mention this as an incident of the early days, when the nature of electrie lighting was still little known.

It required, as you may well believe, a considerable amount of courage to start an enterprise of this kind; not knowing what market there might be for electrical apparatus. It required that we, as it were, should prejudge the future. But it seemed as if an era was opening in which electricity should have a great part - at least, so it seemed to me - and shortly afterwards a company was organized to begin operations in New Britain, Conn., and it was called the American Electric Company. It was formed for the exploitation of the system which we had been developing, and it carried on the work there for a year or two. The management was not very satisfactory at the start - was not pushing, not energetic - and the work dragged; but we kept at it and kept developing new things and perfecting our arrangements so that by the time the beginning of 1882 came there was a 25-light arc machine in existence and a number of appliances that made the system a very workable one - reg-ulators, etc. In other words, we had, during this time of what one might call indifferent business management, got everything in good shape so that we could make a creditable effect with our arc-lighting system, when a better business outlook presented. I may say here that our arc-lighting system was the thing we began with, and that alone - simply a dynamo, regulators and arc lights. In 1882 our lights were shown in Boston for the first time.