247 



NATURE 



[January 19, 1905 



oi- reduced at will by blowing or suction, and it will 

 retain its size constant when placed in open communication 

 with the outer air by means of this tube. This is, of course, 

 thp only plane-faced polyhedron which can thus be formed, 

 faces, edges and vertices being entirely made out of soap 

 lilms. If, on the other hand, a figure has its dihedral 

 angles greater than 120°, then the internal bubble will 

 have concave faces, and will, if placed in communication 

 with the outer air, increase in size until it coincides with 

 the faces of the frame, and will then be kept in equilibrium 

 by their rigidity. This I verified in the case of the 

 eicosihcdron. 



There is one important law which must be mentioned. 

 I found a certain irregularity in the behaviour of the films 

 in the case of the octahedron and rhombic dodecahedron. 

 1 his was due to the fact that two films cannot cross one 

 another at right angles, a law which can be put to the 

 test by placing two plane loops covered with film at right 

 angles, when a small lanceolate film will be formed making 

 two curved lines of intersection with the film on the loops, 

 instead of allowing them to intersect in a single straight 

 line. In the case of the rhombic dodecahedron this slightly 

 modifies the form of the internal bubble, introducing a 

 small edge and a little curvature at each of the acute 

 vertices. This defect causes a serious convexity if the 

 bubble is small, but in general we have double curvatures 

 at the points in question, the remaining portion of each 

 face being plain while the figure retains the form of a 

 rhombic dodecahedron. W. F. \V.\rth. 



Reversal of Charge from Electrical Induction Machines. 



Tjie reversal of the poles of a \'oss machine by giving 

 some turns in the wrong direction, as observed in Nature 

 of January 5 (p. 221), is not an unknown phenomenon. It 

 is described in my paper " Essai sur la Th^orie des 

 Machines electriques k influence " (Gauthier-\'illars, Paris, 

 1898;, p. 38, together with a much more trustworthy and 

 simpler means — an improvement, in theory and in fact. 

 This consists in discharging by hand, at the same time, 

 both the inductors of the fixed disc. Then the reversal is 

 invariably observed without stopping the machine. 



V. .SCH.^FFERS. 



Louvain (Belgium), ii rue des R^collets. 



THE CONSTRUCTION OF SIMPLE ELECTRO- 

 SCOPES FOR EXPERIMENTS ON RADIO- 

 ACTIVITY. 



T^HE electrical method, where it is a|jplicable, is 

 ■*■ now by far the most sensitive method of detect- 

 ings small quantities of matter ; and the recent advances 

 in phy.sical science made by the method of measuring 

 small leakages of electricity, especially in connection 

 with the phenomena of radio-activitv, have excited a 

 very general interest in the experimental arrange- 

 ments employed. The writer hopes that the follow- 

 ing account of simple electroscopes for this kind of 

 work will be found to be of a practical nature and 

 of service to those who, though unfamiliar with many 

 of the devices in general use in a physical laboratory, 

 are nevertheless desirous of making quantitative ex- 

 periments on radio-activity or some other subject where 

 the electrical method is emploved. 



In general the final shape of the instrument will 

 depend very much on the purpose for which it is re- 

 quired ; in fact, it is one great advantage of the gold- 

 leaf electroscope that it can usually be fixed up in 

 any odd corner of the apparatus which happens to be 

 convenient. There is, however, one part of the 

 apparatus which is always the same in sensitive instru- 

 ments, and that is the gold-leaf system itself. Before 

 describing this it will perhaps make things clearer 

 if we consider for a moment one or two points about 

 the theory of the instrument. 



What we observe usually is the rate of decrease of 

 the deflection of a charged gold leaf from a vertical 



NO. 1838, VOL. 71] 



metal support to which it is attached. Now the de- 

 flection in question depends only on the shape and size 

 of the leaf and of the metal support, and on the electro- 

 static potential of the system, so that the rate of 

 collapse of the leaf measures the rate of decrease of 

 the electrostatic potential. But what we wish to 

 measure is the current or rate of alteration of electric 

 charge, and this is equal to the rate of decrease of 

 potential multiplied by the electrostatic capacity of the 

 system. Thus for a given current the rate of move- 

 ment of the gold leaves is greater the smaller the 

 capacity of the system. For a sensitive instrument it 

 is therefore absolutely necessary to have the parts 

 which are metallically connected with the gold leaf 

 as small as possible. 



Cutting gold leaves is a process which requires a 

 considerable amount of patience, especially from the 

 beginner. The process I always adopt is to take a 

 plate of glass and lay a sheet of smooth note paper 

 on it. On this the gold leaf is spread out flat by blow- 

 ing genti}- if necessary, and is cut by means of a 

 razor. To do this, all except a narrow strip at the 

 edge is covered with a second sheet of note paper, the 

 straight edge of which is pressed down with the 

 fingers so as to hold the gold leaf. A fine strip out- 

 side the edge of the paper is then cut off from the leaf 

 by dragging the razor gently backwards parallel to 

 itself and to the edge of the paper. It is not necessary 

 to exert any great pressure during this operation, but 

 a little practice will be necessary to get into the way 

 of doing the saw-cut stroke at the proper speed. Mr. 

 C. T. R. Wilson has succeeded in this way in cutting 

 uniform strips one-tenth of a millimetre across, but 

 for most purposes strips one millimetre wide are good 

 enough. In working with gold leaf much trouble 

 will be saved by working in a room which is free from 

 draughts and disturbances generally. 



For the metal support to which the gold leaf is 

 attached it is convenient to use a piece of wire of 

 about the same diameter as the thickness of the gold 

 leaf. To fix the leaf on to the wire it is suflicient just 

 to moisten the latter at the point of attachment with 

 the tip of the tongue; on allowing the end of the 

 gold leaf to come in contact with the very slightly 

 moist wire it will be found to attach itself sufficiently 

 firmly for all that is required of it. For obvious reasons 

 the cutting and mounting of the gold leaf should be 

 the very last operation in the construction of the 

 electroscope. 



In constructing an electroscope it is of the utmost 

 importance to have trustworthy insulation. When the 

 apparatus has not to be raised to a high temperature, 

 and great mechanical strength is not required, sulphur 

 is a long way better than anything else for this pur- 

 pose. Generally speaking, it is better to have as small 

 a quantity of insulating materia! as possible in order 

 to diminish irregularities caused by the superficial 

 charging up of the dielectric. Suppose we wish to 

 insulate the wire carrying the gold leaf from another 

 wire which supports it mechanically w'e should proceed 

 as follows : — Take a porcelain crucible and gently 

 heat a quantity of pure flowers of sulphur in it until 

 it just melts and forms a clear yellow limpid liquid. 

 It is important that it should not be heated so strongly 

 as to become dark coloured and viscous, as this appears 

 to diminish its subsequent insulating properties. The 

 end of one of the wires is then dipped into the liquid 

 sulphur, when a coating of sulphur forms on the wire. 

 This is allowed to cool until it has solidified, and the 

 operation is repeated a number of times until a bead 

 of sulphur like that shown in Fig. i .\ has formed on 

 the end. The end of the other wire is now heated 

 gently in the flame and applied with a slight pressure 

 to the point a, when it niclls its way inlo the su'.nhur ; 



