134 
NATURE 
[OcTOBER 2, I913 
in a similar way to the lower connection in a 
moving-coil galvanometer. The binants of the 
box are connected to the potential difference to 
be measured. 
Advantages of the instrument are the wide pro- 
portionality between deflection to one side and 
potential difference, and the large range of poten- 
tial which may be given to the needle with satis- 
factory results; in addition we have the stability 
of the needle already mentioned. The deflections 
to one side are proportional to the applied 
potential difference over a range seven times as 
great as is the case when, with the quadrant 
instrument, readings to both sides are taken. 
This property has led to the construction of a 
portable binant instrument with a pointer, which 
can be used as a voltmeter, measuring directly 
potentials to a fraction of a volt without passage 
of current. If used idiostatically, the deflec- 
tions are, of course, proportional to the square 
of the potential, and, connected in this way, the 
instrument measures alternating potentials very 
effectively. 
0 200 +00 600 800 Volt 
Nadelladung 
Fic. 4.—The sensitiveness of the “‘ binant” 
and quadrant 
electrometer compared. 
The potential of the needle in the binant instru- 
ment can be taken as small or as large as may be 
desired. The variation of the sensitiveness with 
the potential of the needle is shown in the diagram 
(Fig. 4) for a quadrant and a binant instrument 
of similar dimensions throughout. The abscissz 
are the difference of potential of the two halves 
of the needle for the binant, the potential of the 
needle above earth for the quadrant, and the 
ordinates are millimetres deflection per millivolt 
applied potential. With the binant form the 
deflection is proportional to the potential of the 
needle up to about 400 volts, and still continues 
increasing up to 1500 volts (off the diagram); in 
the case of the quadrant instrument the sensitive- 
ness increases slowly with the potential of the 
needle, and reaches a maximum at about 300 volts, 
after which increasing the potential of the needle 
is disadvantageous. Further, for the quadrant 
electrometer the potential of the needle cannot be 
taken very small, as in this case the readings are 
too asymmetrical on reversal, as will be seen from 
the ordinary formula of the text-books. For the 
binant the poténtial of the needle may be taken as 
NO. 2292, VOL. 92| 
small as desired; in fact, by altering the potential 
of the needle alone measurements of potential can 
be made over a region of five powers of ten. 
The cause of the peculiar variations of the 
sensitiveness of the quadrant electrometer with 
the potential of the needle, increasing to a maxi- 
mum and then decreasing again, is to be found in 
the fact that the change of capacity per unit 
angular displacement is not constant, as assumed 
in Maxwell’s accepted treatment, but decreases 
with increasing needle potential and increasing dis- 
placement. This is due to the lines of force from 
the radial edges of the needle, which are to a 
large extent diffused not perpendicularly to the 
top and bottom of the box, but horizontally. The 
connection of such horizontal lines of force with 
one of the quadrants is unaltered by the displace- 
ment of the needle, and this influences the changes 
of capacity. The form of the needle and its 
position avoid these disturbances in the binant 
instrument; the narrow gap between the two 
halves of the needle, and their opposite potentials, 
cause the lines of force from the diametral edges 
to spring from one half to the other, instead of to 
the walls of the box, and the position of the gap 
perpendicularly to the gap in the box further dimin- 
ishes the effect. The wide proportionalities of the 
binant electrometer are largely attributable to, 
this result of its peculiar construction. 
E. N. pa C. ANDRADE. 
THE TECHNICAL PRODUCTION AND 
UTILISATION OF COLD.1 
HE appearance of an English translation of 
the work by Georges Claude (1), the success- 
ful French inventor in the field of the liquefaction 
and rectification of air, affords an occasion for re- 
viewing the progress made in this, which seems 
destined to become one of the leading departments 
of twentieth-century scientific industry. Eighteen 
years have elapsed since the inventions of Linde 
and Hampson solved the problem of the produc- 
tion of liquid air in quantity, and extended the 
range of low temperatures practically attainable to 
as great an extent as the electric furnace did in 
the opposite direction. It is sufficient to recall 
the names of Faraday, Andrews, Dewar, Hamp- 
son, and Ramsay to show that this country has 
not been behindhand in pioneers in this field, both ~ 
in regard to the attainment of low temperatures 
and to their utilisation for scientific investigation. 
But there, as in other cases, progress in this 
country seems to have come to a standstill, and 
the commercial application and utilisation of these 
results has been developed entirely abroad, in this 
case chiefly in Germany and France. 
It is on this side of the subject that the present 
book furnishes much information difficult to 
acquire easily elsewhere. Part i., dealing with 
elementary principles and the history of the 
subject, and part iii., with the properties of liquid 
1 © “ Liquid Air, Oxygen, Nitrogen.” By Georges Claude. Translated 
by H. E. P. Cottrell. “With a Preface by D’Arsonval. Pp, xxv. +418. 
(London : J. and A. Churchill, 19%) Price 18s. net. 
(2) ‘‘Le Froid industriel.” L. Marchis. Pp. xx+328-++104 figs. 
(Paris: Félix Alcan, 1913.) Price 2 .50 francs. 
