lviii GENERAL SUMMARY OF SCIENTIFIC AND 



the author's intention to make a thorough investigation of 

 the vast field thus opened. 



Lundquist has given the results of his calculations to de- 

 termine the distribution of heat in the normal sun spectrum, 

 founded on certain measurements of Lamansky's. lie repre- 

 sents the intensity of this heat graphically, and gives curves 

 in which the ordinates represent intensities, and the abscissas 

 wave lengths. It appears from these curves " that in the 

 normal spectrum of the sun the maximum of heat is situated 

 about in the middle of the luminous spectrum, and diminish- 

 es on both sides of this point," thus confirming entirely the 

 experimental results obtained by Dr. John W. Draper in 1872. 

 In the electric spectrum, however, assuming Tyndall's results 

 as data, calculation gives a curve in which the maximum of 

 heat is near the line A. In this case the distribution of heat 

 is not equal in both halves of the visible spectrum. 



Iloorweg has repeated with great care the experiments of 

 Tyndall and Magnus upon the diathermancy of moist air, 

 with a view to reconcile the discrepancies in their results. 

 The general arrangement of the apparatus was similar to that 

 used by Wild, a Leslie's cube being placed on either side of 

 a Melloni's pile furnished with its conical reflectors. But for 

 the introduction of the moist air between the cube and pile 

 on the one side and the dry air on the other, two cylinders 

 were used, the one filled with moistened pumice, the other 

 with calcium chloride. These were placed beneath the line 

 joining the pile with the source of heat, so that, whenever a 

 current of air was driven through them, moist air rose at one 

 end of the pile and dry at the other. With a very delicate 

 galvanometer no deviation could be detected. A pair of 

 tubes, each 25 centimeters long, open at the ends, and bored 

 laterally with fine holes, was then substituted, but with 

 scarcely an appreciable result. Both tubes were now placed 

 on the same side of the pile, and a slight but distinct devia- 

 tion was observed, amounting to 1.7 per cent. The tube was 

 now increased to a meter in length, a heated copper plate 

 being used as the source of heat. The absorption of moist 

 air was 2 per cent. The experiments were repeated with 

 various sources of heat ; the absorption by moist air varied 

 from 3 to 0.4 per cent. Alcohol vapor absorbed, under like 

 conditions, from 6 to 27 per cent, of the heat. Iloorweg con- 



