CUBIC PARABOLA APPLIED AS A TRANSITION CURVE. LXI. 
Discussion. 
Mr. C, O. Burge said that the adaptation of the cubic parabola 
to transition curves had been first brought before the Royal Society 
by Mr. Walter Shellshear, M. Inst. C.E., about nine years ago. Since 
then, the author of the present paper had communicated a paper 
on the subject, with tables, which had been adopted by the Rail- 
way Construction Department, and he would like to explain how 
these tables were utilised in the Government Railway surveys, all 
curves of twenty chains radius and under being transitioned. 
There were usually two surveys for every line: first, the trial 
survey, on which the estimates were based, and secondly, the per- 
manent staking, on which construction was carried out. Formerly 
the trial survey was very sketchy in character, and the final selec- 
tion was left to the later one. Now, however, the trial survey 
was carefully aligned, and the second one was in many cases more 
or less of a mechanical operation. In the trial survey, however, 
the transitions were not set out, but four chains, as a minimum, 
of straight was left between the sharper reverse curves. This 
gave room for the insertion afterwards, in the permanent staking, 
of a four chain transition to each curve, half of each transition 
being absorbed by half of the straight, and half on two chains of 
the circular curve. The four chain transition was adopted almost 
without exception for all the curves dealt with. 
Mr. Burge was of opinion, however, that though it was a good 
thing to have the simplicity gained by a uniforin length of transi- 
tion, this length was not enough for main lines where high passenger 
speeds were to be expected, and that, if the Department had to 
deal with such lines, the transitions should be lengthened. It 
should be remembered that, at high speeds, four chains would be 
passed over in less than four seconds, and the sudden oscillation, 
above the springs, set up by the rise in the outer rail, in the four 
chains, would be very severe. In a short curve the fresh oscilla- 
tion set up by the descent of the rail, at the end of the curve, 
might synchronise with the original one, and thus intensified 
