EINSTEIN'S "MASCHINCHEN" 51 EINSTEIN'S "MASCHINCHEN" FOR THE MEASUREMENT OF SMALL QUANTITIES OF ELECTRICITY I In 1907 Einstein became interested in measuring very small quantities of electricity. These measurements could provide experimental support for the occurrence of volt- age fluctuations in condensers, an electrical phenomenon related to Brownian motion and discussed by Einstein in a paper written in December 1906.[1] On 15 July 1907, in a letter to his friends Conrad and Paul Habicht, he announced his discovery of a "new method of measuring very small quantities of energy."[2] Einstein's "new meth- od" consisted essentially of amplifying a very low initial voltage by means of a spe- cial electrostatic induction machine (or multiplier), whose output voltage could then be measured with a simple electrometer. In fact, Einstein's idea was not new, but goes back to the last decades of the eigh- teenth century, a time when the study of electricity was based purely on electrostatic experiments and observations, and high voltages were generated with friction ma- chines. Globes, cylinders, or disks made of glass (or other insulating substances) were rotated and rubbed by one or more leather cushions the friction charged the glass and the charges were accumulated on big brass conductors.[3] Electricity was measured by simple divergence electroscopes. In 1775 Alessandro Volta invented the electrophorus, a simple instrument capable of generating high potential differences by electrostatic induction,[4] and all subsequent electrostatic induction generators were based on Volta's electrophorus. The first small induction machines invented during the last decades of the eigh- teenth century were meant, like Einstein's, to multiply an electrical charge mechan- ically so that it could be detected and measured with an electroscope.[5] Unlike the friction machines, however, they were not powerful generators. The doubler pro- posed by Abraham Bennet in 1778, for instance, was a kind of double electrophorus. In the same year William Nicholson invented a doubler with two fixed disks and a rotating one, and in 1795 Tiberius Cavallo devised another instrument with three fixed plates and a fourth one oscillating between them. Many modifications and im- provements of these instruments were proposed in later years. In the first decades of the nineteenth century Giuseppe Belli, Professor of Physics at the University of Pavia, invented two types of multipliers: the "macchina ad attuazione" (actuating machine), an induction machine with a rotating glass disk carrying three metallic sec- tors, and a "doplicatore" (doubler).[6] These machines were bigger than earlier mul- tipliers and could be used instead of friction machines to produce high voltage elec- tricity for charging Leyden jars, for generating sparks, and for many other classical electrostatic experiments. The glass disk of the "macchina ad attuazione" was rotated between two massive insulated metallic boxes. At the beginning of the cycle one of the boxes, say A, carried a small electrical charge. When one of the metallic sectors