Marcu 7, 1912] 
NARGRE 
on 
Dover the wind attained the velocity of 71 miles an 
hour. For a long time past cyclonic disturbances have 
arrived in proximity to our coasts from the Atlantic 
with considerable frequency, but, due to the persistent 
high barometer over western Europe, the incoming 
storm systems have followed a track to the north- 
ward, skirting our western and northern coasts. The 
storm area which arrived on March 4 completely 
traversed the British Isles, and probably subsequent 
disturbances arriving will for a time now follow a 
similar path. In connection with the disturbance, a 
severe squall, accompanied by thunder and lightning, 
passed over Kew at 4.30 p.m. on March 4, when the 
wind attained the velocity of 60 miles an hour, and 
a similar squall passed over South Kensington at 
4.40 p.m., the wind velocity recording 42 miles an 
hour. At Dover a squall, with the wind blowing 
68 miles an hour, was experienced at 4 p.m., and a 
corresponding disturbance passed over Valencia, in 
Ireland, at 7.40 a.m., which gives a rate of travel of 
rather less than 50 miles an hour. 
THE past winter, comprised in the three months 
December, January, and February, proves to be one 
of the warmest experienced of recent years, notwith- 
standing the severe frost which occurred at the close 
of January and at the beginning of February. A 
summary of the weather for the thirteen weeks ended 
that the mean temperature for the winter was in 
excess of the average over the entire kingdom; the 
greatest excess occurred in the east and south-east of 
England and in the Midland counties. The aggre- 
gate rainfall for the winter was largely in excess of 
the average over the whole of the British Isles, except 
in the north of Scotland, where the deficiency amounted 
to 2°95 in. The greatest excess was 5°3 in., in the 
south-east and south-west of England. In the Mid- 
land counties the excess was 4°3 in., and in the south 
of Ireland 4°5 in. The rainy days were also in excess 
of the average everywhere, except in the north of 
Ireland. The duration of bright sunshine was de- 
ficient, except in the north of Scotland, but the differ- 
ence from the normal was nowhere very large. 
Greenwich the mean temperature was above the 
average in each of the three winter months, the 
excess being respectively 5°, 2°, and 4°; the mean for 
the whole period was 42'5°, which is 3° above the 
normal. There has only been one winter as warm in 
the last thirty-five years, the mean for the three 
months being 435° in the winter of 1898-9. 
rainfall was also in excess of the average in each of 
the three months, the aggregate excess being 3°3 in. 
The duration of bright sunshine was in good agree- 
ment with the normal. 
In the December (1911) number of the Annals of 
Tropical Medicine and Parasitology, issued by the 
Liverpool School of Tropical Medicine, some novel 
“Tables of Statistical Error” are given by Sir Ronald 
Ross and Mr. Walter Stott. The tables show, for a 
given true percentage, how many observations must 
be made in order that the odds may be m: 1 that the 
observed percentage lies within given limits. The 
limits taken are +1, +2, +3, £4, “£5, or +6 per 
NO. 2210, VOL. 89] 
At | 
The | 
| to be projecting from 
March 2, issued by the Meteorological Office, shows | 
cent., the odds 99999: 1, 9999: 1, 999: 1, 99:1, 9: 1, 
and 1:1, and the percentages are tabulated by steps 
of a unit. As the calculation appears to have been 
made, however, on the usual basis of a normal dis- 
tribution, it is not clear what meaning’ can be attri- 
buted to the figures given for very low percentages, 
where the number of observations is not nearly 
sufficient to justify such an assumption. The tables, 
which are obtainable as a separate publication, should 
do much to lessen the publication of results based on 
quite inadequate statistical data, and thus serve a very’ 
useful purpose, but the point to which we have 
directed attention should have received more attention 
in the explanatory introduction. 
Mr. G. R. M. Tempie 
sends us from York a copy 
of a photograph, here re- 
produced, which illustrates 
very clearly the result of the 
expansion of water by freez- 
ing during the recent severe 
frost. The bottle was filled 
with clean water and tightly 
corked; when the water had 
frozen a stem of ice about 
44 in. in length was found 
the 
neck of the bottle, as shown 
in the illustration. This stem 
represented, of course, the 
increase of volume undergone 
by the water in passing from 
the liquid to the solid state. 
The bottle must have been 
cracked while — solidification 
was going on, otherwise the 
water would have escaped. 
Protrnding stem of ice formed 
by the freezing of water in a 
bo.tle. 
Tue illustrated article by Prof. E. F. Northrup on 
a photographic study of vortex rings in liquids, pub- 
lished in our issue of February 1 last (vol. Ixxxviii., 
p. 463), has prompted Mr. A. W. Ackermann to send 
a description of some experiments performed by him 
twenty years ago in the production of vortex rings in 
liquids. He took a cubical vaseline tin of 85 cm. 
edge, cut a hole in the lid 1 cm. in diameter, filled 
the tin with a solution of permanganate of potash, 
and placed the tin in a bath 6 ft. long. By means 
of a long stick impulses were given to the top or 
end of the tin, and the vortex rings were ejected at 
pleasure and studied. At Mr. Ackermann’s sugges- 
tion, Prof. C. V. Boys, F.R.S., was asked if he had 
investigated the matter. He reminds us that the late 
_ Prof. Guthrie had a large glass trough made in the 
early eighties of the last century for experiments on 
liquid vortex rings. In the centre of one end there 
was a ‘‘ gun” with a thick sheet india-rubber back. 
The gun was filled with a solution of rosaniline. Dr. 
Guthrie’s trough was used later by Sir Arthur 
Rticker, F.R.S., while professor of physics at the 
Royal College of Science, London. Prof. Boys goes 
on to inform us that he would have expected that a 
trough wider than 12 cm., as described in Prof. 
Northrup’s article, certainly not less than 30 cm. or a 
