570 TRANSACTIONS OF SECTION G. 
for friction. The value of the pull is calculated from the readings of the 
frequency and the voltage induced in a special measuring coil, which is wound 
close to the armature so that the flux measured is independent of leakage. The 
‘factor’ of the instrument is determined experimentally by a method in which a 
standardised spring is substituted in place of the specimen, and a heavy mass 
is attached to the armature to reduce the amplitude of vibration. The maximum 
value of the magnetic pull is double that of the mean value which is thus deter- 
mined. It was shown that the influence on the wave of pull of higher har- 
monics in the wave of E.M.F. is small, as only the fifth and seventh harmonics 
are active. A search-coil is provided in the instrument for checking their values 
by means of an oscillograph. , 
Particulars were given of tests carried out with ductile and hard-drawn steel 
wires. From these it appears that specimens of a 0:47 per cent. carbon steel 
broke when tested under pulsating load, with about 85 per cent. of their 
breaking load under steady loads. A ductile low-carbon steel, on the other 
hand, broke with only 65 per cent. of the steady breaking load, 7.e.,-at a load 
close to the yield point. The extension of this material under pulsating load 
(20 per cent.) was very close to that obtained with steady load. 
3. Exposure Tests of Light Aluminium Alloys. 
By Professor Ernest WILson. 
During the past eleven years reports have been presented to the British Asso- 
ciation at fairly regular intervals. The tests show that alloying commercial 
sluminium with copper, unaccompanied by iron, nickel, or manganese, is not satis- 
factory. A 2°6 per cent. copper alloy has completely deteriorated in ten years 
and increased its electrical resistance 25 per cent. ‘Duralumin’ is a copper- 
manganese alloy of aluminium with the addition of about 05 per cent. mag- 
nesium. During the last year a specimen has increased its electrical resistance 
5°15 per cent. It would be interesting to know if this is due to the comparatively 
large percentage of copper (3°5 to 5°5) which this alloy is stated to contain, or if 
the percentage of manganese (0°5 to 0-8) is too low. This alloy has attracted 
attention in that a breaking-load as high as 90,000 lb. can be obtained, if 
desired, according to treatment. Its specific electrical resistance at 15° C. is 
about twice that of commercial aluminium. A specimen of high-conductivity 
copper wire has increased its electrical resistance 1-2 per cent. in one year. 
4. The Féry Bomb Calorimeter.1 By Ropert S. WHIPPLE. 
The bomb form of calorimeter is now frequently employed for the deter- 
mination of the calorific value of coal, because the combustion of the coal is 
more complete than in those calorimeters in which the oxygen is admitted at 
atmospheric pressure or at a pressure slightly above atmosphere. 
The instrument designed by Professor C. Féry,? of the Ecole de Physique 
et de Chimie, Paris, is of the bomb type, being in general design somewhat 
similar to the well-known Mahler instrument, but greatly simplified in its 
details. The bomb consists of a light iron vessel weighing about one kilo, and 
having a capacity of about 250 c.c. supported in the centre of a brass vessel by 
two discs of constantan (an alloy of copper and nickel). As the mass of metal 
in the bomb is small and no water-jacket is used, the rise in temperature of 
the bomb due to the combustion of the coal is large, averaging about 20° C. 
when 0°5 gramme of coal is burnt. It is evident that with such a large tem- 
perature rise the measurements need not be made with the same degree of 
accuracy as in the case of the usual type of bomb calorimeter, in which the 
temperature rise amounts to only 2° or 3° C. for the same. quantity of coal. 
The temperature rise is measured by the thermo-electric force generated by 
the constantan discs and the iron bomb, the surrounding envelope acting as the 
1 See The Electrician, September 13, 1912. 
? See also ‘Nouvelle Bombe Calorimétrique,’ by C. Féry, Journal de 
Physique, p. 550, July 1912, and Génie Civil, May 25, 1912. 
