The electric field required to produce practical saturation varies with the intensity of the ionization and consequently with the activity of the preparations to be examined. For preparations which have an activity not more than 500 times that of uranium, under ordinary conditions, a field of 100 volts per cm. is sufficient to produce a practical saturation current. For very active samples of radium, it is often impossible to obtain conveniently a high enough electromotive force to give even approximate saturation. Under such conditions comparative measurement can be made by measuring the current under diminished pressure of the gas, when saturation is more readily obtained. The method to be employed in the measurement of this ionization current depends largely on the intensity of the current to be measured. If some very active radium is spread on the lower of two insulated plates as in Fig. 1, and a saturating electric field applied, the current may readily be measured by a sensitive galvanometer of high resistance. For example, a weight of ·45 gr. of radium chloride of activity 1000 times that of uranium oxide, spread over a plate of area 33 sq. cms., gave a maximum current of 1·1 × 10⁻⁸ amperes when the plates were 4·5 cms. apart. In this case the difference of potential to be applied to produce practical saturation was about 600 volts. Since most of the ionization is due to rays which are absorbed in passing through a few centimetres of air, the current is not much increased by widening the distance between the two plates. In cases where the current is not quite large enough for direct deflection, the current may be determined by connecting the upper insulated plate with a well insulated condenser. After charging for a definite time, say one or more minutes, the condenser is discharged through the galvanometer, and the current can readily be deduced. =55.= In most cases, however, when dealing with less active substances like uranium or thorium, or with small amounts of active material, it is necessary to employ methods for measuring much smaller currents than can be detected conveniently by an ordinary galvanometer. The most convenient apparatus to employ for this purpose is one of the numerous types of quadrant electrometer or an electroscope of special design. For many observations, especially where the activity of the two substances is to be compared under constant conditions, an electroscope offers a very certain and easy method of measurement. As an example of a simple apparatus of this kind, a brief description will be given of the electroscope used by M. and Mme Curie in many of their earlier observations. [Illustration: Fig. 11.]