26 



SGIENGE. 



[N. S. Vol. III. No. 53. 



level in latitude 45° and under normal condi- 

 tions, 0.089873 ± 0.0000027. 



In 1891 Prof. Morley published * a series of 

 determinations of the volumetric composition 

 of water. The results of these determinations 

 were extremely concordant and there can be no 

 reasonable doubt that the same ratio would be 

 obtained again by the same method. When, 

 however, this ratio is combined with the ratio 

 of the densities given above, the resulting value 

 for the atomic weight of oxygen does not agree 

 with that which Prof. Morley has obtained by 

 the direct weighing of oxygen and hydrogen 

 and of the water formed by their union. Scott 

 has recently determined f the volumetric ratio 

 and finds the value 2.00285. This ratio, when 

 combined with the ratio of densities as found 

 either by Lord Eayleigh or by Prof. Morley, 

 gives the same value for the atomic weight as 

 that found by the gravimetric method. Prof. 

 Morley has, therefore, determined the volu- 

 metric ratio by another method. In a series 

 of ten experiments he determined the density 

 of electrolytic gas obtained from a solution of 

 caustic potash. He also determined the excess 

 of hydrogen present in the gas. From the re- 

 sults obtained, and, taking into account the 

 change in pressure occasioned when one volume 

 of oxygen is mixed with two volumes of hydro- 

 gen and the mixture is made to occupy three 

 volumes, the value 2.00269 for the volumetric 

 ratio was calculated. 



It seems to be established, therefore, that the 

 values obtained by Prof. Morley with the eudi- 

 ometer were not correct as representing the 

 volumetric ratio and that the density of a gas 

 in a tube is different from that in a globe, the 

 effect on the density being different for a light 

 gas from that for a heavy one. 



The gravimetric composition of water was 

 determined in a series of twelve experiments. 

 In these the oxygen was weighed in large 

 globes, the hydrogen (three and one-half 

 grams), in palladium, and the two gases were 

 burned in an apparatus so devised that the 

 water formed was also weighed. In this way 

 each experiment gave two independent deter- 

 minations of the atomic weight of oxygen. 



* Amer. Journ. of Science, 41, 220. 



tPhil. Trans. 184, A, 543 (1893). 



The results were : 



From the ratio of hydrogen and oxygen, 15.8792 

 From the ratio of hydrogen and water, 15.8785 



These values agree to the third decimal with 

 the value calculated from the volumetric com- 

 position and the ratio of densities as given 

 above. 



The final results of Prof. Morley's determina- 

 tions are : 



Grams. 

 Weight of one liter of oxygen, latitude 45°, 1.42900 

 Weight of one liter of hydrogen, latitude 45°, 0.089873 

 Atomic weight of oxygen, chemical method, 15.879 

 Atomic weight of oxygen, physical methods, 15.879 

 Molecular weight of water, chemical method, 17.879 



In conclusion a summary of previous deter- 

 minations of the constants in question is given. 

 Omitting the earlier determinations, which were 

 manifestly inaccurate, ani the results of one 

 more recent experimenter, whose work appears 

 to have been affected by some source of constant 

 error, the mean of all the other determinations 

 of six different observers gives the value 15.879 

 for the atomic weight of oxygen. 



It is impossible, in a brief sketch of this kind, 

 to convey any adequate idea of the pains which 

 was taken at every step to secure the greatest 

 possible accuracy in the work, nor of the genius 

 which has been displayed in devising compli- 

 cated apparatus adapted for the determinations 

 to be made. The work is classical and must, 

 hereafter, be consulted by every one who wishes 

 to do the best work in this field. 



W. A. NOYES. 



bimee's evolution of butterflies.* 

 Peof. Eimee, of Tiibingen, is an enthusiastic 

 opponent of Darwin's theory of Natural Selec- 

 tion, and has a theory of his own to replace it. 

 The theory of Eimer has been defended by 

 him on various occasions, his main exposition 

 being given in his work on the origin of species 

 published in 1888. His investigations on butter- 

 flies (thus far of the genus Papilio auct. only) 

 are intended to afford proof of his theory in a 



*Die Ai'tbildung und Verwandtschaft hei den 

 Schmetterlingen. II. Theil. von Dr. G. H. Theodor 

 Eimer unter Mitwirkung von K. Fickert. Text 8vo. 

 Pp. viii, 153. Atlas Folio Tafeln v.-viii. Jena, 

 Gustav Fischer. 1895. 



