[ 411 ] 
XVI. On the Relation of the Direction of the Wind to the Age of the Moon, as inferred 
from Observations made at the Royal Observatory, Greenwich, from 1840 November 
to 1847 December. By G, B. Airy, Esq., Astronomer Royal. 
Received December 27, 1850, — Read March 6, 1851. 
In the year 1849, in a voyage to Shetland, I heard allusions to the belief entertained 
generally by Norwegian seamen, that a northerly wind may always be expected 
about the time of new moon. The expression of this belief was so positive, and the 
implication of the interests of the persons entertaining it was so distinct, that it ap- 
peared to me extremely probable that there was some physical foundation for it. At 
the first convenient opportunity, therefore, I took measures for discussing, with re- 
ference to this question, the directions of the wind at the Royal Observatory, during 
a period of rather more than seven years, as ascertained from the records of Osler’s 
Self- registering Anemometer. I extended the research so far as to enable every reader 
to judge whether there is any probable relation between any Direction of Wind and 
any Age of the Moon. 
The collection and summation of the numbers was effected under the immediate 
superintendence of Mr. Glaisher. Great pains were taken to establish such checks 
on the operation that error is almost impossible. 
The general result is contained in the Table subjoined to this paper. And, while 
it shows that there is great uncertainty in the verification of an empirical law, even 
from nearly ninety lunations, it seems very distinctly to negative the asserted law 
which gave rise to the inquiry. 
In explanation of the Table, it is only necessary to remark that the civil day on 
which the new moon occurred is taken as the day of new moon (at whatever hour 
the conjunction occurred), and that the other days are counted in succession from it. 
The moon’s synodic period being nearly 29^ days, the month sometimes consists of 
twenty-nine days, sometimes of thirty, in almost equal proportions in the long run ; 
and thus the numbers in the last horizontal row of figures must, to make them com- 
parable with the others, be very nearly doubled. Although the sums of numbers in 
the other horizontal lines are not absolutely equal, they are so nearly equal that no 
remarkable error will be produced by assuming them as equal. 
