W. MARRIOTT — METEOROLOGICAL OBSERVATIONS. 213 



Thermometers. — The Maximum Thermometer may be on either 

 l^egretti and Zambra's or Phillips' principle. Negretti's registers 

 in the following manner : — The tube is bent and contracted near 

 the bulb in such a way that it allows the mercury to pass in ex- 

 panding, but on contraction the column of mercury in the tube is 

 broken off, and the extremity shows the highest temperature that 

 has been attained. It is set by simply holding the instrument 

 bulb downwards and shaking it. It is to be mounted horizontally. 

 In Phillips' thermometer the index is formed by a small portion 

 .of the mercurial column, separated from the main thread by a 

 minute air-bubble, and is pushed on before the column when it 

 expands, but does not return with it when it contracts. It there- 

 fore rests at the extreme position to which it has advanced, and 

 the end furthest from the bulb registers the highest temperatm-e 

 which has been attained. As the air-bubble is liable to be dis- 

 placed, and the instrument to be put out of order, jS'egretti's ther- 

 mometer is the more trustworthy of the two. 



The Minimum Thermometer generally used is that known as 

 Eutherford's. The fluid used is spirit, and in it there is immersed 

 a steel index. When the temperature falls, the spirit draws the 

 index along with it, but on rising again, the spirit passes by the 

 index, leaving it at the lowest point to which it has been cbawn, 

 thus registering the minimum temperature. The instrument is set 

 by raising the bulb and allowing the index to flow to the end of the 

 column of spirit. The thermometer should be mounted very nearly 

 horizontally, the bulb being a trifle lower than the other end. 



The Dry- and "Wet-bulb Thermometers should be of precisely 

 the same make, have small bulbs, and be about four inches apart. 

 The wet-bulb should be covered with a single piece of very fine 

 muslin, in the form of a jacket to fit the bulb if it is cylindrical, 

 and should project a little below it. A conducting thread of six or 

 eight strands of darning cotton should be tied round the neck of 

 the bulb over the musKn, the other end passing into a water 

 receptacle through a small orifice at the top, placed about three 

 inches from the bulb. Clean rain-water alone should be employed. 

 The muslin and conducting thread must be washed in boiling 

 water prior to use, and must be changed at least once a month, 

 and oftener if there is any appearance of drrt or deposit. In damp 

 and foggy weather the thermometers should be carefully wiped 

 about five minutes before the observation. When the temperature is 

 below the freezing-point, the wet-bulb requires careful management. 

 Instead of a film of water round the bulb, there must be a coating 

 of ice, which can be obtained by applying, with a camel's hair 

 brush, water taken from under ice. This must be done half an 

 hour or so before the time of observation. It is best to remove the 

 muslin entirely, as a lower and more correct reading will be ob- 

 tained if the coating of ice is on the naked bulb ; for this purpose 

 thei-mometers with roughened or ground bulbs should be used, as 

 the ice will form more readily on them than on smooth bulbs. As 

 Boon as the temperatui'e rises above the freezing-point, the bulb and 



