NATURAL PHILOSOPHY. 189 



and Pettit of England. This instrument consists of a glass ring of a pris- 

 matic section placed so as to surround the flame at such a height that all 

 the lateral rays of light proceeding from it are intercepted by the ring, and, 

 filling on its inclined exterior surface, are projected downwards and con- 

 centrated within the range of the refractor. This range will depend on 

 the angles to the interior and exterior surfaces of the prismatic ring, and 

 can evidently be increased or diminished by making the angles between 

 the sides and the base of the prism greater or less. The effect produced by 

 the arrangement is such as could not fail to excite the attention of even 

 the most cursory observer. It is not indeed under all circumstances that 

 it is desirable so to concentrate the light ; but for all show purposes, and to 

 enable delicate mechanical or other operations to be performed with arti- 

 ficial light, this is absolutely necessary, and there is no contrivance by 

 which this can be done so effectually as by the present. London Mechan- 

 ics' Magazine. 



f 



STANDARD BAROMETERS. 



The "Kew Committee" of the British Association, having been 

 requested by Lieut. Maury to give their advice respecting the best form of 

 a marine barometer, reported at the last meeting that they had taken the 

 subject into consideration, and, after examining several instruments, had 

 selected one, in which it is believed all the requisites for making correct 

 observations at sea will be found to have been obtained at a very moderate 

 cost, combining convenience and accuracy in observing with simplicity 

 and durability in its general construction. 



The great importance as to certain conditions reauisite in a good 

 barometer induced the Committee to have the action of this instrument 

 tested by such means as were at their command, and this was effected by 

 Mr. Welsh (accompanied by Mr. Adie, the maker) in a voyage to Leith 

 and back to London; subsequently the action of the instrument was 

 further tested by Mr. Welsh in a voyage to and from the Channel Islands. 



[The results of these experiments are detailed in a letter from Mr. Welsh, 

 included in the report.] Mr. Welsh says : 



" 1st. Any one of the three barometers is capable of showing at sea the 

 changes of pressure, with a probable error of about 0.005 inch, or at most 

 0.007 inch. 



2d. The tremor of a steamship is rather beneficial than otherwise to 

 the performance of the barometer, and (leaving the pumping out of 

 consideration) the barometer performs rather better at sea than on land. 



3d. For such a motion of the ship as must be very common, the amount 

 of contraction of the tube should be greater than in any of the three 

 barometers employed, say a contraction to 18 or 20 minutes. The mean 

 amount of pumping from ten observations in the return voyage was, for 

 the tube contracted to 5 minutes, 0.084 inch, and for the one contracted 

 to 10 minutes, 0.031 inch; the greatest observed being for the former, 



