1840.] On the Ijonstruction and use of Portable Baro7neters. 269 



a single tube barometer is useless for nice observations, unless it has 

 been compared with another, even if it is in good order, for the capil" 

 lary action in the tube can never he valued unless by comparison with a 

 barometer which can be depended on. The double syphon tubes I 

 have above remarked are free from this defect, but whatever the con- 

 struction may be, the mercury in a good boiled tube must be free from 

 specks or cloudiness throughout its whole length ; and it must jar 

 against the top of the tube (not hump), as a bar of steel might be sup- 

 posed to do. 



15. Since the above was written, I have had the opportunity of 

 examining another of Newman's barometers, like that described in para. 

 3, and I found the lid or cover of the cistern so shrunk and loose, that 

 nearly all the mercury had escaped, and also that the tube had been 

 attached to the cover by glue and tow, which, in the damp weather in 

 this climate, had been softened so much, that the mercury had been forced 

 completely thro-ugh the joint. As there are many of these instruments 

 in India, the following directions regarding them may be useful. 



16. Whenever any mercury has been found to have escaped into the 

 case of the instrument, it may be generally considered as certain, that 

 it has become useless until put again into order, and it should in con- 

 sequence be carefully kept with the cistern upwards, because if set up- 

 right, some air might get into the tube, and render necessary the trouble 

 of reboiling the tube. The brass ring at the bottom of these instruments 

 has 6 brass screws in it ; the three nearest the wood attach the ring to 

 the wood, the lower three screws are screwed into the wooden lid of the 

 cistern. In the lower piece of brass by which the bottom of the cistern 

 is turned, there are two iron screws, which open into the cavity of the 

 cistern ; care must be therefore taken that they are never taken out ex- 

 cept when necessary to take the instrument to pieces. The three brass 

 screws which hold the cover of the cistern being removed, the part of 

 the cistern which is movable being pulled, the cistern and tube wiU 

 come out easily, but great care must be taken not to turn it round in 

 the least, which would snap off the little thermometer. To put a baro- 

 meter of this kind in order, if any air has got into the tube, the first thing 

 to do, will be to ascertain if the lid of the cistern is tight. The two iron 

 screws in the movable bottom piece being removed, the brass cap 

 may be pulled off easily, and the cistern and tube, can be filled with 

 mercury through the holes. The cistern being quite full, the tube is to 

 be held downwards, and the bottom of the cistern turned round a quarter 

 turn. The tube is then set up in an erect position, and on jerking it 

 slightly up and down, if the lid of the cistern is loose, the mercury will 



