234 On the Mean Temperature of the 



the purpose, in the garden of' the University. For three niontlis 

 the highest and lowest temperatures were also daily observed. 

 For that purpose, in want of a better instrument, the following 

 arrangement was given to a common mercurial thermometer : — 

 After the above thermometer was finished, I warmed the bulb, 

 till the quicksilver ascended almost to the top of the tube ; it 

 was then quickly plunged with its open end in boiling alcohol, 

 a little of which entered the tube. A small cylinder of steel 

 (piece of a needle) was sunk in the alcohol, which, on holding 

 the thermometer in the perpendicular position, soon fell down 

 on the surface of the quicksilver. A portion of the alcohol was 

 then removed, till not more than a filament, ten degrees long, re- 

 mained in the tube ; a little mercury was then allowed to enter, 

 in order to confine the alcohol, and then air, which filled the up- 

 per end of the tube, which was closed with some cotton. It is 

 easily seen, that, if we first bring the small steel cylinder in con- 

 tact with the surface of the filament of mercury in connexion 

 with the bulb, and then place the thermometer in the horizon- 

 tal position, the cylinder will be continually pushed forwards, 

 while the mercury expands by heat ; but, when the tempera- 

 ture has reached it maximum, and begins to decline, the cylin- 

 der remains on the spot to which the mercury has pushed it. If 

 now we make an observation, the distance of the cyhnder of 

 steel, from the surface of the contracted filament of mercury, 

 will give the number of degrees which one must add to the tem- 

 jjerature at the time of observation, to find the previous maxi- 

 mum of temperature. To obtain now the corresponding mini- 

 mum, we bring the cylinder in contact with that filament of 

 mercury which confines the alcohol, by holding the thermometer 

 for some moments reversed, with the bulb uppermost, by which 

 the cylinder will slowly slide down. When now, by increasing 

 cold, the mercury contracts, the cylinder is pushed on by the 

 second mercurial filament, and finally rests (if the thermome- 

 ter lies horizontal), where the temperature has attained its mini- 

 mum. 



The following table contains the mean for every month, 

 from November 1827 to November 1828. The thermometer is 

 Fahrenheit's. 



