January 3, 1896.] 



SCIENCE. 



25 



The refinement attainable is not equal to that 

 reached in the case of inorganic compounds. One 

 cannot think of working with millionths of mil- 

 ligrams, and will at times have to be content 

 if a satisfactory result is reached with tenths of 

 milligrams. The classes of compounds dealt 

 with, in this first number of the book, are : 1. 

 The anthracene group ; 2. Phenols ; 3. Nitro- 

 compounds ; 4. Quinones, Ketones, Aldehydes. 

 It is to be hoped that the appearance of the 

 book will lead chemists to try the new methods, 

 as it appears that their work will be much facili- 

 tated by them. It must, of course, be borne 

 in mind that the problem of detecting minute 

 quantities of organic compounds does not often 

 present itself, though there are cases in which 

 it becomes of importance. Iea Kemsen. 



On the Densities of Oxygen and Hydrogen and on 

 the Ratio of their Atomic Weights. By Ed- 

 ward W. MoRLEY, Ph. D. Published by the 

 Smithsonian Institution, AVashington, D. C. 

 1895., 4°. xi. 117 pp. 



For more than ten years Prof. Morley has 

 been almost constantly engaged on the work 

 which is described in this paper. With a pains- 

 taking fidelity to the highest ideals of accurate 

 work which has rarely been equalled and has 

 never been surpassed, he has determined four 

 constants which are partly interdependent, and 

 which are of very great importance in physical 

 science. These constants are : the density of 

 hydrogen, the density of oxygen, the ratio of 

 the combining volumes and the ratio of the 

 combining weights of the two elements. 



The density of oxygen was determined by 

 three different methods. 



■ In the first series nine determinations were 

 made. From nine to twenty-one and one-half 

 liters of oxygen were weighed in large globes 

 which were filled at the temperature of the 

 laboratory, 



In the second series sixteen determinations 

 were made. Instead of measuring the tempera- 

 ture and pressure directly in this series the 

 oxygen was brought to the same temperature 

 and pressure as that of hydrogen contained in 

 another large globe. The pressure of the 

 hydrogen was previously measured at the tem- 

 perature of melting ice, thus making the globe 



containing it, in effect, a very sensitive air 

 thermometer. The difference between the co- 

 efficients of expansion of hydrogen and of oxygen 

 was of course considered. 



In the third series seventeen determinations 

 were made. The globes were filled at the tem- 

 perature of melting ice and, after weighing them 

 filled with oxygen, they were exhausted and 

 weighed again. The oxygen in this series was 

 prepared partly from potassium chlorate and 

 partly by the electrolysis of dilute sulphuric acid. 



The results of three series were : 



By use of thermometer and manometer D = 1.42879 

 By compensation D= 1.42887 



By use of ice and barometer D -- 1.42917 



Giving double weight to the last series, the 

 weight of a liter of oxygen under normal con- 

 ditions at sea level and in latitude 45° is 

 1.42900 grm., with a probable error of 0.000034 

 grm. 



Five series of determinations of the density 

 of hydrogen were made. 



In the first and second series the same 

 methods were used as in the first and third 

 series for oxygen. 



In the third, fourth and fifth series hydrogen 

 was absorbed in palladium, contained in a glass 

 tube, and, after weighing, was expelled into three 

 globes which were surrounded with melting ice, 

 and which had a combined capacity of forty- 

 two liters. By this means three and seven- 

 tenths grams of hydrogen were weighed in a 

 comparatively small apparatus, and the volume 

 occupied by the gas was accurately determined. 

 The method has the additional advantage that 

 any mercurial vapor contained in the globes was 

 without effect on the determination. In all, 

 sixty-four determinations were made. The re- 

 sults were as follows : 



Series I. D .— 0.089938 



" 11. D= 0.089970 



" III. D — 0.089886 ± 0.0000049 



" IV. D == 0.089880 ± 0.0000088 



" V. D = 0.089866 ± 0.0000034 



It is believed that mercurial vapor entered 

 the globes in the first two series and that the 

 results of those series are too high. They are 

 accordingly rejected. The remaining series 

 give as the weight of a liter of hydrogen at sea 



