NATURAL PHILOSOPHY. 221 



compass should be gradually turned round until one of the lines upon the 

 glass remains coincident with one well-defined edge of the cloud as it 

 passes across the field of view. The angle indicated by the magnetic needle 

 being then read off, the azimuthal bearing of the cloud's track from the mag- 

 netic north is at once ascertained. The convenience of this instrument might 

 be increased by having an eye-piece attached to it, capable of being fixed ha 

 such a manner as to point to the intersection of the cross lines hi the centre of 

 the circle, so that the eye may be kept steadily in the same direction. By 

 means of an apparatus on the principle of a camera obscura, the direction of 

 the wind could be easily ascertained by observing the compass bearing of the 

 cloud's track. And, hi the absence of better instruments, the reflection by a 

 mirror ought certainly in ah 1 cases to be preferred to the indication of vanes, 

 whose action must always be vitiated more or less by friction, and perhaps by 

 other causes, besides being liable to be acted upon by currents which have 

 been distorted from their true direction by obstructions due to houses, trees, 

 and the configuration of the earth's surface. The changes of wind and weather 

 so characteristic of our climate, might, perhaps, be more certainly or more 

 speedily predicted by comparing the motions of the clouds in the higher 

 regions of the atmosphere, with those nearer the earth's surface, than from 

 information derived from other sources. Mr. Stevenson has observed a 

 change of wind apparent in the direction of the high clouds for two days 

 before the currents near the earth's surface were affected, although they ulti- 

 mately assumed the same direction. 



IMPERFECTIONS OF THE ANEROID BAROMETER. 



Prof. Guyot, at the last meeting of the American Association, hi a commu- 

 nication on the Aneroid, freely acknowledged its great conveniences, but 

 he entered his formal protest against dependence on it for nice measure- 

 ments of mountain altitudes. The instruments are all individuals, their 

 errors subject to no rule except that of great variability. He had made many 

 experiments and comparisons with good mercurial barometers, and found it 

 worthy of reliance as a scientific instrument, only under the condition that it 

 is kept stationary, and individually tested to learn the corrections for tem- 

 perature, &c. Had he trusted to his aneroid barometer hi a recent visit to the 

 Black Mountains, he would have been led to errors of 400 to 500 feet, as was 

 proved by the two good mercurial barometers that he carried. A traveller 

 who carries an aneroid alone with him, must not expect accuracy within two 

 or three hundred feet. Simply from motion or from having been subjected to 

 great changes of pressure, it will change its zero without giving any external 

 indication. 



ON THE SIMILARITY OF FORM IN SNOW AND CAMPHOR UNDER 

 CERTAIN CONDITIONS OF CRYSTALLIZATION. 



Mr. Glaisher, in a recent paper before the London Meteorological Society, 

 stated that Mr. Spencer had endeavored to observe snow crystals, but, from 

 their perishable nature, he has experienced a great difficulty, as they require 



