nd home again among them. As is now known, this disturbing effect had
actually dispersed the train.

The shower, which was well seen in 1866, was pretty confidently expected
in 1899, and to guard against the mischance of cloudy weather, it was
arranged that the writer should, on behalf of the Times newspaper, make
an ascent on the right night to secure observations. Moreover, it was
arranged that he should have, as chief assistant, his own daughter,
an enthusiastic lady aeronaut, who had also taken part in previous
astronomical work.

Unfortunately there were two nights, those of November 14th and 15th,
when the expected shower seemed equally probable, and, taking counsel
with the best authorities in the astronomical world, it seemed that the
only course to avoid disappointment would be to have a balloon filled
and moored in readiness for an immediate start, either on the first
night or on the second.

This settled the matter from the astronomical side, but there was the
aeronautical side also to be considered. A balloon of 56,000 cubic feet
capacity was the largest available for the occasion, and a night ascent
with three passengers and instruments would need plenty of lifting power
to meet chance emergencies. Thus it seemed that a possible delay of
forty-eight hours might entail a greater leakage of gas than could be
afforded.

The leakage might be expected chiefly to occur at the valve in the
head of the balloon, it being extremely difficult to render any form of
mechanical valve gas tight, however carefully its joints be stopped with
luting. On this account, therefore, it was determined that the balloon
should be fitted with what is known as a solid or rending valve,
consisting simply of balloon fabric tied hard and fast over the entire
upper outlet, after the fashion of a jam pot cover. The outlet itself
was a gaping hole of over 2 feet across; but by the time its covering
had been carefully varnished over all leakage was sufficiently
prevented, the one drawback to this method being the fact that the
liberation of gas now admitted of no regulation. Pulling the valve line
would simply mean opening the entire wide aperture, which could in no
way be closed again.

The management of such a valve consists in allowing the balloon to
sink spontaneously earthwards, and when it has settled near the ground,
having chosen a desirable landing place, to tear open the so-called
valve once and for all.

This expedient, di