1837.] 



Meteorological Observations. 



205 



by the magnet, to loosen the index ever so little, then apply heat to the 

 bulb, and drive up the index with its superabundant mercury quite into 

 the air-vessel. This requires many trials and much patience. When 

 there, hold the instrument bulb downwards, and suspend the index by a 

 magnet at the top, allowing any globule of mercury to drop into the 

 origin of the tube below ; then heat the bulb cautiously over a very 

 small clear flame of an oil lamp, till the mercury rises to the very top 

 of the tube, and fairly unites with the globule there awaiting it. Let 

 the bulb cool, and the mercury will sink in one united column ; if not, 

 heat it again. When this is accomplished, the index may be set loose, 

 by withdrawing the magnet, and restored to its proper position in the 

 tube. 



A self-registering thermometer may be advantageously left (properly 

 secured) for a whole year, or parts of a year, on elevated summits or 

 rather remarkable points, to ascertain their maxima and minima of tem- 

 perature during absence. In such cases, take care to defend them from 

 discovery, or accident from wild animals, birds, snakes, &c. In taking 

 it up for reading off, observe not to derange the indexes, and do not leave 

 it without seeing that the indexes are in contact, and the temperature 

 that of the air at the moment. 



Of Thermometers buried in the Earth. — Thermometers buried at 

 different depths, for the purpose of examining the monthly changes of 

 temperature of the soil, should have their balls and lower part of the 

 scale well wrapped up in woollen cloth or pounded charcoal, and should 

 be placed in strong earthern vessels, which may be entirely withdrawn 

 from the ground so as to allow of inspecting and reading off the scale, 

 without exposing the balls to any possibility of changing their tempe- 

 ratures while under examination. The vessels should be fitted with 

 covers, to defend the scale from injury in burying and digging up. 



A pipe of earthenware (composed of separate pieces), or one of 

 wood, may be sunk ten or fifteen feet below the surface, into dry earth, 

 and a thermometer, defended as above, lowered by a chain. The pipe 

 being then obstructed at every two feet by some stuffing readily hooked 

 up, the thermometer may be easily examined, and a register of its indi- 

 cations kept with very little trouble. In like manner, the temperature 

 of wells may be registered. 



Of the Temperature of the 8ea.— The surface temperature of the 

 water at sea should be registered, as a matter of course, with 

 the same regularity and at the same hours as the barometer 

 and thermometer. It is more conveniently (and with quite ac- 

 curacy enough for the purpose) obtained by taking up a bucket- 

 full of the water and stirring round the thermometer in it. When- 

 ever a change to the extent of 2 8 Fahr. appears to have taken place 



