May 28, 1914] 
NATURE 
on 
22 
JI 
until recent years. The estimated height was 
37000 ft., but Glaisher lost consciousness for thirteen 
minutes, and his estimate is therefore uncertain; the 
highest point may not have been much more than 
30,000 ft. 
In 1875 the French Academy of Sciences arranged 
ascent, 5 
for an and M. Gaston Tissandier and two 
companions 
ascended to a 
great height; 
im Spite of 
oxygen  in- 
ipanteat io mS. 
however, his 
1wo co m- 
panions lost 
their lives, 
and Tissandier 
himself barely 
eos ¢ a p-e.d 
as phy xiation 
when the 
balloon 
mesa ci ie d 
a height of 
28,000 fetes 
This disaster 
prevented any 
very high 
ascents for 
some _ years, 
but in 1894 
Po Bier son 
made the first 
of his series of 
ascents that 
have — eclipsed 
all previous 
records. In’ July, 1g01, Berson actually took a 
reading of the barometer corresponding to a height of 
34,500 ft. or 103 km., and in spite of oxygen inhala- 
tion, he, too, became unconscious This may be taken 
as the highest ascent yet made by man. ~ 
The danger to life at great elevations led to another 
method of research. In 1891 M. Bonvallet sent up a 
number of paper balloons carrying post-cards asking 
Fic. 4.—Dines winding-gear for kites. 
Fic. 5.—Balloon meteorograph with Bourdon tube barometer and 
bimetallic thermometer, 
the finder to post them, with a note of the time and 
place of finding. The experiment was repeated by 
MM. Hermite and Besancon with larger balloons and 
simple recording instruments. One of these balloons 
having reached a height of 30,000 ft., they made a 
still larger one, provided with a better recording in- 
strument. This was the earliest registering balloon, 
NO. 2326, VOL. 93] 
| cannot count on Govern- 
_ graph he has made one 
and it made its first ascent on March 21, 1893, reach- 
ing a height of 15 km., or nearly g3 miles. In Ger- 
many these experiments were soon repeated, Under 
the auspices of the German Society for the Promotion 
of Aerial Navigation, Dr. Assmann sent up a balloon 
which rose to a height of about 22 km., or 135 miles. 
These were sensational experiments, but they seem 
to have attracted little attention in this country. 
In the very early days of kite-flying ordinary mini- 
mum thermometers used to be sent up; but when the 
study began seriously, special instruments had to be 
designed. In the ordinary pattern, pens, actuated by 
some form of barometer, thermometer, and hygro- 
meter, trace a record on a revolving cylinder. — For 
the barometer an ordinary aneroid box was used at 
first, but this has given place in most of the instru- 
ments used on the Continent to a tube which acts in 
the same way as the Bourdon tube pressure gauge. 
The same system was used for the thermometer, the 
tube being filled with spirit, the expansion or con- 
traction of which changed the shape of the tube. But 
a bimetallic thermometer has also been much used; 
this consists of a coil of metal made of two pieces 
which expand or contract at different rates with rise 
or fall of temperature. In both thermometers the 
resulting motion is 
communicated to the 
pen by levers. 
In this country we 
are badly situated for 
balloon work; many of 
our free balloons are 
lost in the sea, and we 
ment support in the re- 
search as can some of 
our more fortunate 
neighbours. I think 
I am not overstating 
the case if I say that 
but for the ingenuity of 
Mr. Dines we should 
have had practically no 
upper air research in 
this country. In_ his 
light balloon meteoro- 
of the most 
meteorological 
ments. It 
shillings what 
pounds; and, 
striking 
instru- 
costs in 
the other 
weighing as_ it 
Fic. 6.—Dines light meteorograph. 
in 
two 
instruments cost 
does under 
| ounces with its case, it can be sent up with quite 
small balloons. An aneroid box, as it expands with 
decrease of pressure, carries two pens across a copper 
plate; the thermometer pen is moved by the relative 
contraction of a strip of German silver compared with 
an invar steel bar. Two lines are thus scratched on 
the copper plate; the length of the lines from the 
origin represents the pressure, and their distance apart 
the temperature. The trace is very minute, the whole 
plate being about the size of a postage stamp, and it 
has to be read under a microscope; but the expansion 
and contraction of the thermometers used in the Con- 
tinental meteorographs have to be magnified mechanic- 
ally, and, as Mr. Dines has pointed out, the optical 
magnification is perhaps less liable to error. The 
instrument is so small that some who had used th: 
other instruments looked on it as a toy rather than 
as a serious instrument. But it was soon found to 
give as good results as the other forms; and when at 
Manchester University twenty-four balloons carrying 
these instruments were sent up in the space of twenty- 
four hours, one each hour, it made a considerable 
