Notes on Barometer Construction. 41 



of bore, and the exact diameter of the bore is also to 

 be ascertained, because when the tube is closed and 

 filled, and especially when bent into the syphon form, 

 the ascertaining of these points is no longer readily accom- 

 plished. The tubes chosen for making into barometers 

 will be often longer than is requisite for the instrument, 

 and the end cut off may be almost or quite the same dia- 

 meter as the upper end of the barometer ; when this is the 

 case it may be worth while to carefully label and set aside 

 this end piece, which would at any time answer any 

 question concerning the curve of the meniscus or any of a 

 kindred nature which might arise. Concerning capillarity, 

 a suggestion may be offered : — With any tube about to be 

 employed, or with the end piece of tube cut off as just 

 mentioned, a measurement of the effect of capillarity may 

 be made by a method given in Bunsen's Gasometry: — 

 Measure a column of mercury in the tube per se, and 

 measure the same column after covering it with a few drops of 

 corrosive sublimate solution : in the former case you have 

 the meniscus proper to the given diameter of tube in its 

 integrity; in the latter the mercury assumes a horizontal 

 upper surface, and the difference of height of the two 

 columns is that due to those physical causes which are 

 collectively spoken of as the influence of capillarity. 



Before proceeding to clean the inner surface of the tube 

 it will be well to become acquainted with what has been 

 ascertained concerning chemically clean glass, as especially 

 set forth in the papers of Tomlinson.* In the Chemical 



* When you have prepared with all precautions your supply of mercury 

 for the cistern and for filling the tube, I will suppose in a clean porcelain 

 yessel, with a nicely-polished glass bell jar for a cover, in a relatively dust- 

 free apartment, you may try a simple experiment which is suggestive of 

 the necessity of extreme cleanliness in barometer construction. Let the 

 experimenter elaborately wash his hands, and then press his finger against 

 the pure mirror surface of the mercury ; he will, if I am correct, produce a 

 minute and faithful oleograph of the skin structure— a picture of the skin 

 surface — drawn in sebaceous and epithelial particles, which the cuticle, 

 however well cleansed, is always ready to throw ofiE. Now if you take up 

 Deschanel's Manual of Physics, or other elementary work of the kind, in 

 which the barometer is figured and described, you will see a wood engraving 

 of the Torricellian experiment : — the hand inverting the tube filled with 

 mercury, and the finger about to be placed on the open end on the mercurial 

 column, before its insertion in the cistern — all very good for lecture table 

 demonstration, but certainly violating the rules according to which a good 

 barometer should be filled and erected. You cannot blow through a tube 

 or touch the end of it without making a fouled surface ; and although I am 

 not prepared with any suggestions for the best method of meeting this 



