148 
THE COURSE OF STORMS. 
examined this question with care. He finds, that of the storm- 
centres which leave the American coast, one in nine passes over 
some part of England, while one in six comes sufficiently near 
to occasion a gale. 
A very instructive chart, as showing the erratic behaviour 
which some storms exhibit in their passage across the Atlantic, 
is published by Professor Loomis in a recent number of the 
American Journal of Science. In this case, the storm-centre, 
starting from a spot a little distance south of Greenland, and 
proceeding for some days in a regular easterly course, suddenly 
turns upon itself, and, after wandering in an aimless manner 
about the middle of the Atlantic, is found after the lapse of a 
fortnight at a place not many hundred miles from that, whence 
it started. Then, as if reminded of its duty, it takes a fresh 
start, and finally reaches the coast of Norway. A storm such 
as that must defy accurate prediction. 
For the solution of some of the . questions which I indicated 
just now, we may have to wait until observations have been 
collected and discussed from a wider area than at present. We 
need observations from southern, as well as northern latitudes, from 
the Pacific, as well as from the Atlantic. In the meteorological 
millennium, when we shall have the barometric pressure over the 
whole world charted for every day in the year, a flood of light 
will no doubt be thrown upon points which are now obscure, and 
if the forecasting of the weather should never arrive at the 
perfection which some anticipate, we shall at least have the 
satisfaction of knowing the reason why. 
