green's standard barometer. 



57 



dent. This was removed by the substitution of iron in the place 

 of wood ; but it was soon found impracticable, in this form of cis- 

 tern, to prevent damage from rust. These objections led to the 

 ]jresent plan of construction, which effectually secures the joints 

 without the use of any cement. The surfaces concerned are all 

 made of a true figure, and simply clamped together by the 

 screws, a very thin leather washer being interposed at the joints. 

 This would not be permanent, however, but for the especial care 

 taken in preparing the boxwood. The boxwood rings are all 

 made from the centres of the wood, and concentric with its 

 growth. They are worked thin and then toughened, as well as 

 made impervious to moisture, by complete saturation with shel- 

 lac. This is effected by immersing them in a suitable solution 

 in vacuo. The air being withdrawn from the pores of the wood, 

 is replaced by the lac. This, however, with the after-drying or 

 baking, requires care ; but when properly done, the wood is ren- 

 dered all but unchangeable. 



Another peculiarity consists in making the scale adjustable to 

 correct for capillarity, so that the barometer may read exactly 

 with the adopted standard, without the application 

 of any correction ; and this, too, without destroy- 

 ing the character of the barometer as an original 

 and standard instrument. Near the 30 inches 

 line. Figure 6, is a line v, on the main tube ; this 

 last line is distant exactly thirty inches from the 

 tip of the ivory point; therefore, when these lines 

 coincide, or make one line, the scale is in true 

 measurement position ; or the 30 mark is exactly 

 thirty inches from the tip of the ivory point in 

 the cistern. In this position, the amount of cor- 

 rection due to capillarity being ascertained, the 

 scale is then moved that quantity and clamped 

 firm. The barometer will now give the readings 

 corrected for capillarity, and thus avoid at once the labor of ap- 

 plying a correction, and the risk of error from an accidental 

 neglect of it. 



It must be borne in mind that this correction applies only to 

 the particular tube, and while preserved in good condition. 



If this tube is injured and again used, or another tube put in 

 its place, the scale should then be moved until the lines coincide, 



I'if.S 



