140 PHILOSOPHICAL TRANSACTIONS. [aNNO 1757. 



would 4-equire the scale to be altered one 6oth part : and the error arising from 

 making use of the same scale will be about one 6th of a degree, if the thermo- 

 meter is observed when it has fallen 10 degrees. 



If the weight of the mercury be thought inconvenient, it may be avoided by 

 the construction described in fig. 2, where the bottom of the tube is bent so as 

 to point upwards, and is joined to a ball a, which communicates with a cylinder 

 placed above it. In all other respects it is the same as the instrument before 

 described. It is filled with spirit of wine and mercury ; the quantity of the latter 

 being sufficient to fill the whole tube and ball a. No part of the spirit in the 

 cylinder can get into the tube as long as the instrument is kept in an erect po- 

 sition, or even if it be carefully laid down fiat on a table. For though in this 

 last case some of the spirits may get into the ball a, it will rise to that part of the 

 ball which is then uppermost, and will not touch the orifice of the tube n ; which 

 was the reason for adding this ball, which would be unnecessary if the instrument 

 was kept constantly erect, or nearly so. If the spirit should come to touch the 

 orifice of the tube n, it would work up between the mercury and the glass ; 

 which would put the instrument out of order. 



The thermometer fig. 3 is designed for showing the greatest cold which 

 happens in any place during the time the instrument is left in it. The tube is 

 bent into the shape of a s)^hon of unequal legs standing parallel to one another, 

 the bend being at the bottom. The top of the shorter leg is bent to a right 

 angle, and immediately opens into a ball a, which, by means of a short bent 

 tube on the opposite side, communicates with a cylinder standing parallel to the 

 legs of the syphon, and pointing downwards. This cylinder contains the greatest 

 part of the fluid ; and is added only to make the thermometer more sensible 

 than it would be, if the ball a was made of a sufficient size to contain the proper 

 quantity of fluid. This instrument is filled with spirit of wine, with the addition 

 of as much mercury as is sufficient to fill both legs of the syphon, and about a 

 4th or 5th part of the ball a. The common degrees of heat are shown by the 

 top of the mercury in the longest leg, or by the top of the spirit, in case any of it 

 is left above the mercury. When the mercury in the longest leg sinks by cold, 

 that in the shorter leg will rise, and will run over into the ball a : from whence 

 it cannot return back when the thermometer rises again, as the surface of the 

 mercury in the ball is below the orifice of the tube n. Therefore the upper part 

 of the shorter leg will be filled with a column of spirits of a length proportional 

 to the increase of heat; the bottom of which, by means of a proper scale, will 

 show how much the thermometer has been lower than it then is ; which being 

 substracted from the present height, will give the lowest point that it has 

 been at. 



If no further contrivance was used, the mercury would fall into the ball a in 



