When, however, the term high electromotive force is met with in descriptions of the apparatus of that time, it must be taken to mean a great difference of potential between the terminals of the dynamo, not between the primary terminals of the transformers. Take, for instance, 100 transformers connected in series, run with a difference of potential at the dynamo of 1000 volts, although it was not known at that time how to produce so high an electromotive force, still this would give across the primary terminals of each transformer the modest difference of potential of 10 volts. In this way the difference of potential at the generator was determined by the number of transformers in series. This system had plainly the great disadvantage, that no matter how tortuous a path the lead must follow, it had to pass through the primary coils of all the transformers, and the principles of a proper system of distribution were not present. With the invention of the incandescent lamp the activity of inventors was given quite another direction. The systems of electric lighting up to this time were not sufficiently advanced to permit even of a _division_[3] of the electric light, that is, the ability to feed even a small number of lamps from one generator. We shall only mention this invention so far as it helps to further the history of the transformer. Gramme made the earliest arc lamp that could be employed alone; then followed Jablochkoff, as the first who carried out practically, and with good results, the use of arc lamps in series or in parallel arc with condensers. Siemens and Halske then replaced the Jablochkoff candles with their differential lamp, which, although not offering an opportunity for a good division of light, was unexcelled in construction and manufacture, pointing out the way for further progress in arc lighting. This class of lighting was brought nearly as far forward as it is to-day by the introduction of continuous currents for this use by Brush.