26 



SCIENCE. 



[N. S. Vol. IV. No. 79. 



vertebrates. And on the evidence of cyprino- 

 donts it must be admitted that several of the 

 characters which have been almost universally- 

 regarded as stable landmarks in morphological 

 studies should be given comparatively little 

 definite importance. For in this group, rank- 

 ing only as a family, oviparous and ovovivi- 

 parous forms have been evolved, together with 

 a broad range in intromittent organs and in 

 embryonic nutriment. So that, for example, 

 we must admit that structures like the 'claspers ' 

 of sharks are of but little moment in separating 

 the phylum of the elasmobranchs from that of 

 the Dipnoan or of the Teleostome. In the 

 morphological portion of his work Mr. Garman 

 has directed especial attention to varietal 

 changes, notably in the case of Fundulus hetero- 

 clitus, and to structural variation as shown 

 principally in teeth, digestive tract, urinogenital 

 system and vertebral column. Under the latter 

 head he attributes the decrease in the number 

 of vertebrae of fishes in general, in and toward 

 the torrid zone, to the ' lessening of the com- 

 parative activity of the species,' due to an 

 enlarged food supply and to a decreased need 

 of nutriment, the decreased number is not, there- 

 fore, attributable to the direct action of tem- 

 perature, as several writers appear to have 

 inferred. Bashford Dean. 



A NEW DETERMINATION OP THE RELATIVE 



DENSITIES OF OXYGEN AND HYDROGEN 



AND OF THE RATIO OF THEIR 



ATOMIC WEIGHTS. 



To one familiar with the work of Prof. Morley 

 on this subject it would seem that any one who 

 wishes to add anything to our knowledge of 

 the matter must be an experimenter of unusual 

 ability, and must be willing to expend a very 

 large amount of labor on his determinations. 

 That Julius Thomsen, who has recently pub- 

 lished the results of his experiments in this 

 field,* is an experimenter of unusual ability 

 every one will admit. That the results ob- 

 tained can carry with them any considerable 

 weight in comparison with those of Prof. Morley 

 is very doubtful. It is, however, of very con- 

 siderable interest to find that, by the use of 



*Zeit. fiir Anorg. Chem. 11, 14; and 12, 1. 



comparatively simple apparatus and by methods 

 diflTering in almost every detail, he has ob- 

 tained results which are in fairly close agree- 

 ment with those of Prof. Morley' s elaborate re- 

 searches. 



The method employed for the determination 

 of the ratio of the atomic weights consisted in 

 the determination, first, of the ratio between 

 the weight of a certain amount of aluminium 

 and of the weight of hydrogen evolved by its 

 solution in a strong solution of caustic potash, 

 and, second, of the weight of oxygen required 

 to burn the hydrogen evolved by the solution 

 of a known weight of aluminium. 



The density of hydrogen was determined by 

 measuring over water saturated with hydrogen, 

 the gas evolved by the solution of a known weight 

 of aluminium. The density of oxygen was 

 determined in a similar manner, the gas being 

 evolved by the decomposition of potassium 

 chlorate. The volume of gas measured was 

 approximately one and one-half liters, while 

 Prof. Morley measured a volume of hydrogen 

 amounting to forty-tivo liters. The results ob- 

 tained were : 



Katio of Atomic Weights 1:15.8690±0.0022 

 Density of hydrogen at 0°, 



760 mm. and 45° Lat. 0:089947 ±0.000012 



Density of oxj^gen " 1.42906 

 Prof. Morley 's values were: 



Eatio of Atomic Weights 1:15.879 ±0.00032 

 Density of hydrogen at 0°, 



760 mm. and 45° Lat. 0.089873 ±0.0000027 



Density of oxygen " 1.42900 ±0.000034 



In discussing the result of his determination 

 of the ratio of the atomic weights. Prof. 

 Thomsen remarks that, as he has avoided the 

 weighing and measuring of large volumes of 

 gases, it is probable that his result is nearer the 

 truth than that obtained by others, and that 

 the uncertainty does not extend beyond the 

 fourth decimal. He seems to have overlooked 

 the fact that, in his syntheses of water. Prof. 

 Morley weighed his hydrogen absorbed in pala- 

 dium, and also weighed the water formed by 

 its combination, and that, while he weighed 

 the oxygen in gaseous form, the sum of the 

 weights of oxygen and hydrogen agreed almost 

 exactly with the weight of the water. 



