32 SEVENTH REPORT—1837. 
kind upon record. If, however, this phenomenon were to take place, 
it might not be referred by the observer to the proper cause. A ter- 
restrial object seen by reflection in this manner would be reversed ; 
but it is, perhaps, not impossible, that although the internal surface of 
the liquid air might not reflect mountains in this manner, it might re- 
flect so bright an object as the sun. Halos might, perhaps, always be 
produced by transmitted light. It might also be remarked, that the 
liquid heterogeneous thickness of air opposes a difficulty to the caleu- 
lation of astronomical horizontal refractions by the method of mecha- 
nical quadratures, devised by the late Mr. Atkinson, and employed by 
him in the Transactions of the Astronomical Society, and by M. Biot 
in the Conn. des Temps, unless somewhat modified. 
On the Principle of Mr. Wurwk.u's Anemometer. 
The author rapidly sketched the principle on which his instrument 
registered the quantity of aérial current passing any place. He had 
exhibited the instrument in an unfinished state at the Dublin meet- 
ing, and in a more matured state of its existence at Bristol; it had 
since received some valuable improvements, which were suggested 
by the practical working of the machine. That he might not oc- 
cupy the time of the Section too long, it would suffice at present to 
say, that in it a small set of windmill vanes, something like the ven- 
tilators placed in our windows, were presented to the wind by a 
common vane, let the direction of the wind blow how it might: the 
aérial current as it passed set these vanes into rapid motion, and a train 
of wheels and pinions reduced the motion, which was thence com- 
municated to a pencil traversing vertically, and pressing against an 
upright cylinder, which formed the support of the instrument, and that 
10,000 revolutions of the fly only caused the pencil to descend the one- 
twentieth of an inch. The surface of the cylinder was japanned white, 
and the pencil as the vane wavered kept tracing a thick irregular line, 
like the shadings on the coast of a map: the middle of a line was 
readily ascertained, and it gave the mean direction of the wind actually 
exhibited before the eye by a diagram, while the length of the line was 
proportional to the velocity of the wind, and the length of time during 
which it blew in each direction; which therefore gave what he called 
the integral effects of the wind, or the total amount of the aérial 
current which had passed the place of observation in the direction 
of each point of the compass, during the interval which had elapsed 
since the time of last recording the instrument. This, it was well 
known, was a subject of much importance in meteorological specula- 
tions, but had not been hitherto accomplished. It was indeed deemed 
of much consequence to obtain even the mean direction of the wind 
at a given place, and the celebrated Kimtz, in his Meteorologie, has 
made a collection of several results of this kind; but, in the ordinary 
way of registering even the direction of the wind, which is by stating 
