Remarks on Self-registering Thermometers. 161 



rather awkward contrivance I shall not now describe, as its ge- 

 neral principle is the same as in a thermometer which has oc- 

 curred to myself for exhibiting both maximum and minimum 

 results in the same instrument without the aid of indices, 

 which I believe has not before been attempted. 



In Figure 5 of Plate I. the dotted portions denote alcohol, 

 the parallel lines mercury. The upper part of the tube, for the 

 measurement of the greatest heat, is exactly upon Lord C. 

 Cavendish's plan, and the altitude of the mercurial part of 

 the column A denotes the actual temperature at any moment. 

 The lower part of the tube is bent upwards, and passed into 

 a cylindrical bulb B, close to one side of it, as shown in the 

 figure. It likewise terminates in a capillary orifice, and, as 

 by the contraction of the alcohol which fills the almost entire 

 bulb, the mercury is withdrawn from the tube and falls to the 

 bottom, the measure of minimum temperature in any period 

 will correspond to the existing temperature, (marked as before 

 by the height of the mercury A,) rninus the degrees of the. 

 tube next the bulb, which contains alcohol, measured by a 

 small scale of their own. 



The adjustment of the instrument for a new observation is 

 necessarily somewhat complex. It must first be reversed, and 

 the bulb heated with the hand till the column of alcohol joins 

 the quantity which has been expelled into the upper cistern ; 

 retaining the same position, the bulb must be cooled with 

 ether or some evaporating fluid till the alcohol has retired 

 from the lower extremity of the tube ; when, from th^ 

 position, the mercury will obviously join with the portion 

 which before lay in the bottom of the bulb ; and by again 

 heating it with the hand till it has nearly regained the temi 

 perature of the air, which was known at first by an observa- 

 tion of the summit of the mercury, all the uncertainty will be 

 done away as to when the instrument has regained its proper 

 temperature, which in my former paper I noticed as an error 

 of this principle. By a httle practice, too, the degree of heat 

 given artificially will be so nearly proportioned to the atmo- 

 spheric temperature, that little time will be required to wait in 

 making the adjustment. With regard to the necessity of ether, 

 it is to be observed, that it does little more than counterba- 



VOL. X. NO. I. JAN. 1829. L 



