CAN WEATHER BE PREDICTED IN THE BRITISH ISLES? 171 
when the results of his system are compared with those obtained 
at the present day, the fact of the utter dissimilarity of the two 
methods of warning is entirely ignored. At the present time 
the signals only cover the period of forty-eight hours, but they 
are held to be justified by the occurrence of a storm anywhere in 
the neighbourhood of the place where they are exhibited. It 
is therefore clear that the two systems are not, in the slightest 
degree, intercomparable. 
If we look at our present position as regards storm warnings, 
and compare it with Le Vender’s proposal of sixteen years ago, 
we see that that still describes very fairly the extent of our 
weather knowledge. 
When we see that a disturbance of the atmosphere actually 
exists, we can issue intelligence of the fact to the districts 
likely to be affected by it, and the value of such warnings 
depends on the correctness of the ideas formed by us as to the 
character and extent of the disturbance itself — by this is meant 
particularly the point or points whence the wind will blow, the 
force which it will exert, and the area it will cover — the direc- 
tion of its advance, and the rate of its motion. 
On all these points there prevails as yet a great amount of 
uncertainty ; but that, on the whole, a satisfactory result of the 
system is achieved appears from the fact that a total percentage 
of success of warnings of strong winds exceeding 7 5 has been 
maintained for the last four years, while of actual severe gales 
following warnings the percentage is as much as 44. 
It therefore appears that warnings are more astray on the 
ground of over-precaution than of negligence ; but yet it would 
be absurd to deny that in almost every year some of our very 
severest gales have come on us suddenly and swept much of 
the coasts before any warning has been issued. 
This is mainly owing to the fact that the funds available for 
the purpose will not allow of sufficiently frequent communica- 
tion between the outposts and head-quarters, and that much 
time is always lost in the necessary telegraphing to and fro. 
It has repeatedly been urged on us that we should institute 
a regular service of telegrams from America, but we had for 
some years such a service with Newfoundland, and derived 
little benefit from it. What would really be of value to us 
would be that a thoroughly competent meteorologist on that 
side should telegraph daily a general resume of the weather 
prevailing on the Atlantic seaboard of the States, with intelli- 
gence of any disturbances leaving them which showed signs of 
an intention to cross the herring-pond.. This has been at- 
tempted of late, with some measure of success, by the “ New 
York Herald,” which prophesied a storm on our coasts for 
