40 Scientfiic Proceedings, Royal Dublin Society. 
gas instead of air; no oxidation can be possible in the absence of 
oxygen. ‘The nitrogen in the jacket can be occasionally renewed 
without affecting the inside chamber, and as the inner chamber is 
always surrounded, as well as filled, with nitrogen, no evil effects - 
can ensue from diffusion. 
(B).—This portion of the apparatus is novel, and I therefore 
submit a design and detailed description. (See Plate IV.). 
It consists essentially of a uniform motion clock somewhat 
similar to what we use for our astronomical telescopes. 
This clock serves to drive two spindles, the first at a rate of 
once in 15 seconds, and the other once in 60 seconds. On the 
15-seconds spindle, there is a controlling apparatus very similar 
to what we use on our equatorials, by which the rate is checked 
every turn, 7e., every 15 seconds; and if there be any difference 
between this clock and the standard clock amounting to even 
one-tenth of a second, one or other of the correcting differential 
gears (0 or U’) is brought into action, and this error is wiped out, 
This spindle (A) may therefore be assumed to revolve exactly in 
accord with the standard clock A. 
On the 60-seconds spindle (B) is a similar correcting arrange- 
ment, which, however, only comes into play once in 24 hours, viz., 
on the arrival of the Greenwich signal, and its duty is to wipe 
away any difference + or—that may be present between the time 
as given by Greenwich and that given by the standard clock, 
and to register this on a paper slip for after-reference. 
This corrected 60-seconds spindle (B) is used, first, to show 
correct time on a suitable dial or set of dials in the hall, and, 
secondly, to drive or control all the 200 clocks in the various 
rooms of the building. 
(C).—For the distribution of time throughout the building, 
there is the choice of various systems. 
There is the pneumatic system and the electrical system, and 
the latter can be again divided into a system which regulates or 
controls clocks possessing a certain motive power in themselves, 
and that which actually drives the whole series of clocks by elec- 
trical impulses. 
Any of these systems can be worked from the distributing 
clock described under head B. 
