SEPTEMBER 20, 1918] 
the fact that large gas-filled lamps cause con- 
siderable heating of the lamp sockets used with 
them, owing to conduction of heat from the 
bulbs, measurements were made of the tem- 
perature rise of ventilated and unventilated 
sockets, and of the leading-in cables when 
used in lanterns of both the open and the en- 
closed types. It was found that the cables were 
liable to a total temperature of about 140° C., 
and that little improvement resulted from the 
use of existing arrangements for socket venti- 
lation. 
Good progress was made with the research 
on buried cables. The electrical constants of 
all the cables under test were determined, and 
it was found that in general the rise of tem- 
perature of a cable drawn into a stoneware 
duct was twice that of a similar cable 
armored and laid direct in the ground, cables 
laid solid in bitumen occupying an intermedi- 
ate position. Other points, such as the length 
of time required by the cables to attain their 
maximum temperature, the effect of the ends, 
and the temperature gradient in the duct and 
in the cable, were also dealt with. The design 
of the special motor-generator set which will 
be used to supply continuous current for the 
cables, and which will also be used for heavy 
test work, has been settled and the machine is 
under construction. It consists of two continu- 
ous-current generators, each fitted with two 
commutators. Each commutator is arranged 
for a current of 1,250 ampéres at 6 volts, the 
ends being brought to a series-paralleling board. 
By this means currents up to 1,250 ampéres at 
24 volts will be available for the cable tests 
and up to 5,000 ampéres at 6 volts for the test 
work, suitable regulation to any required cur- 
rent being obtained by resistances in the ex- 
citing cireuits. The generators, which will be 
driven by a three-phase 3,000-volt motor run- 
ning off the town supply, will be separately 
excited, the excitation current being main- 
tained constant by a centrifugal spring regu- 
lator putting resistance into or out of the field 
of the motor. 
The heat division collaborated with the 
Heating and Cooking Panel of the Engineer- 
ing Standards Association in developing efti- 
SCIENCE 
285 
ciency tests for electrical plates and cooking 
stoves. A number of pyrometer sheaths of 
British manufacture were tested for their 
softening point and porosity at high tempera- 
ture. Some of the samples of porcelain, 
though unglazed, stood the porosity test re- 
markably well; in fact, when the interior was 
exhausted to a pressure of a few millimeters of 
mercury the walls collapsed under the external 
pressure before any leakage occurred. On the 
other hand, tubes of fireclay and shrunk 
alumina often showed a return to atmospheric 
pressure in about 10 seconds. A large series 
of aluminium alloys was investigated for 
thermal conductivity on behalf of the Light 
Alloys Sub-committee of the Advisory Com- 
mittee for Aeronautics, the advantages of 
such alloys, on account of their high conduc- 
tivity, for the construction of the cylinders 
and pistons of aero engines having been ap- 
preciated for some time. It was evident at the 
outset that measurements at room tempera- 
ture would not be sufficient, since the tempera- 
tures encountered in actual practise range up 
to 800° C., and an absolute and expeditious 
method of determining thermal conductivity 
and its variation with temperature was there- 
fore devised. 
In Metrology the year was distinguished 
by the development of the gauge-testing work. 
The new gauge-testing building was completed 
in May, and occupied in the early part of 
June. The numbers dealt with, however, con- 
tinued to increase and at the end of 1917 were 
more than double those of 1916. There was a 
steady increase in the quantity of screw 
gauges tested, but in the first three months of 
this year there was a considerable reduction in 
gauges of the more simple types. The types 
that continue to be received are, however, in 
general more complicated and involve more 
work in testing than those which have dropped 
out. The gauge-rectifying shop was found 
much too small to meet the demands made 
upon it, and the erection of a new workshop of 
10,000 square feet area was sanctioned. The 
building of this has been started. 
Arrangements have been made for the cali- 
bration of glass volumetric apparatus. Accom- 
