MELLONI OX THE NOCTURNAL COOLING OF BODIES. 469 



ness, the surface of contact with the air will certainly go on in- 

 creasing; but the underlying stratum, which freely radiates its 

 heat, will increase in the same proportion ; so that at first 

 sight it does not appear that after a certain interval of time 

 there would be any difference between the cooling of very thin 

 bodies and that of bodies having a certain thickness. But a 

 moment's reflection is sufficient to convince ourselves that the 

 spiders' webs are more apt to become cool than bodies of a larger 

 volume. In fact, the threads of which these webs are composed 

 being excessively fine, radiate from all points of their mass, and 

 receive but little or no heat from the stems, leaves or earth by 

 which they are supported*; for the conductibility, being in the 

 inverse ratio of the diameters, becomes sensibly nothing for cylin- 

 ders of extreme thinness. Therefore the nocturnal cooling of a 

 spider's web will be quicker and more intense than that of other 

 bodies; and since, according to the hypothesis adopted, dew 

 depends on the degree of cold produced, the abundance of it on 

 spiders' webs is favourable, and not the contrary, to the theory 

 of Wells; and those who pretend to find in it so formidable an 

 objection to the ideas received in science, only show their inca- 

 pacity to judge soundly of such scientific questions. 



We shall presently see analogous conclusions fi'om other facts 

 and other observations, which these incompetent judges regard 

 as contrary to the explanation of dew founded on the principle 

 of nocturnal radiation. But let us for the present remark, that, 

 according to this principle, the cooling of bodies must necessarily 

 precede the precipitation of dew on their surfaces. This funda- 

 mental proposition of Wells' principle may be easily demon- 

 strated by means of our apparatus. 



It is, in fact, sufficient to expose to the free air two of the 

 conical vessels of tin-plate which we have described, one of which 

 contains a thermometer armed with polished metal, and the 

 other a thermometer armed with varnished metal. The thermo- 

 meters in the closed vessel will both indicate the same tempera- 

 ture ; but if the covers be taken off, and after having verified the 

 fact already mentioned of the immobility of the metaUic thermo- 

 meter and the fall of the varnished one, we observe attentively the 

 surface of the latter instrument, it will be seen to be at first very 

 brilliant, then slightly clouded, afterwards more and more dull, 

 and finally covered with drops of dew. Every other radiating 



* 'I'licre must be a mistake here, cither in the oiigiiial Italian or the French 

 translation. — [Tr. N.] 



