136 M. Melloni on the Theory of Dew. 



agreed. The experiment was varied in another manner. At 

 the extremity of glass cylinders planted in the ground, they 

 fixed horizontal plates of zinc, copper, and glass ; each of 

 these plates had a cavity in the centre in which the bulb of 

 a thermometer was plunged, while the tube, supported by a 

 wire, rose vertically above the superior surface ; a thermo- 

 meter, suspended freely between these plates, was intended 

 to measure the temperature of the air. In this case, as in 

 the former, the apparatus employed in experiment indicated, 

 at the beginning of the night, calorific differences, which, at 

 a later hour, disappeared ; insomuch that, at the earliest 

 dawn of the following morning, all the thermometers were 

 found at heights sensibly equal. 



These facts appeared to the opponents of Wells' theory, 

 completely decisive ; and they thenceforward maintained that 

 *' the pretended nocturnal cold of bodies, indispensable to the 

 formation of dew, was a mere chimera !" For myself, I 

 should affirm that if anything in this appears imaginary and 

 fantastical, it will be found in the reasoning employed in 

 drawing such an inference ; for the experiments 'of these 

 gentlemen were made near the ground, in an atmosphere 

 charged with humidity. All the tubes of the thermometers 

 were uncovered, and, in the last mentioned experiment, the 

 bulbs of the thermometer communicated through the inter- 

 mediate plates, with the cylinders which supported the appa- 

 ratus. Now, the glass of which these tubes and cylinders 

 were composed radiates considerably, its temperature sinks, 

 and the cold thus acquired is communicated to the bodies that 

 touch it : the latter being placed in a very humid air, then 

 precipitate the aqueous vapour ; but we know that water ra- 

 diates and cools with as much energy as glass, varnish, and 

 lamp-hlack. It was, therefore, nothing surprising that the 

 thermometers in contact with the plates or plateaux of metal 

 indicated, after some time, the same temperature as the ther- 

 mometer surrounded with the most radiating substances. 

 From the circumstance of the metallic surfaces, which were 

 found covered with dew, becoming as cold as the vitrified or 

 blackened surfaces, it might well follow that the water, glass, 

 Sindiamp-dlack, are bodies possessed of emissive powers sen- 



