548 



SCIENCE. 



[N. S. Vol. XXI. No. 536. 



grown, and without further application of ni- 

 trogen. 



The final analyses of the crops and soils 

 showed that the organic nitrogen contained 

 in the cow peas, and accumulated largely from 

 the air, was utilized hy the succeeding non- 

 leguminous crop, and that there was a gain of 

 nitrogen in the soils where no cow peas were 

 grown, and that such gains were greatest 

 where the largest amount of manure had been 

 applied. 



The Detection of Methyl Alcohol: Heyward 



ScUDDER. 



The formation of methyl salicylate is not a 

 reliable test, because ethyl salicylate has al- 

 most the same odor. The Kiche and Bardy 

 test takes too long to be of practical value. 

 The Trillat test (delicacy 0.2 per cent.) is 

 uncertain. The deepening of the blue on 

 heating is characteristic. Wolf and Robine 

 state that Trillat's test is not reliable and give 

 modifications that may be of value in cases of 

 doubt. The Haigh (delicacy 5 per cent.) and 

 Spangle-Ferriere (delicacy 1 per cent.) tests 

 use phloroglucine to show the presence of 

 formaldehyde. The color obtained varies with 

 the concentration of the formaldehyde. A 

 blank test with ethyl alcohol must always be 

 made. The effect of heat is characteristic. 

 By using proper conditions a rapid test (deli- 

 cacy 2 per cent.) can be made. The Mulliken, 

 Scudder test (delicacy 3 per cent.) is very re- 

 liable because of the formation of flocks as 

 well as color. On account of the uncertainty 

 of color tests for small amounts, it is advised 

 first to try a rapid test and, if no definite 

 result is obtained to concentrate by fraction- 

 ation. 



The Origin of Radium: Bertram B. Bolt- 

 wood. 



It had been shown in previous papers that 

 the determination of the relative quantities of 

 uranium and radium in a number of minerals 

 strongly indicates a constant ratio between 

 the quantities of these elements occurring to- 

 gether. In the present paper improved meth- 

 ods for the quantitative determination of both 

 radium and uranium are described, and the re- 

 sults obtained from an examination of twenty- 



two separate samples, comprising twelve dis- 

 tinct mineral species, are given. These results 

 show a constant proportionality between the 

 quantities of radium and uranium in all of 

 the samples examined, and lead to the in- 

 evitable conclusion that uranium is the parent 

 of radium. F. H. Pough, 



Secretary. 



DiscLSSioy a:s'd correspondence. 



THE naturalist's UNIVERSAL DIRECTORY. 



A NEAV edition of the ' Naturalist's Direc- 

 tory ' has recently come to hand. In casually 

 turning the leaves I noticed that the names of 

 Joseph Le Conte and J. W. Powell were stiU 

 retained in the list of American naturalists, 

 although their deaths had occurred as long ago 

 as 1901 and 1902. These lapses suggested that 

 the directory might not be trustworthy in 

 other respects, and at my suggestion an asso- 

 ciate compared its lists with various other 

 lists of scientific men, for the purpose of test- 

 ing its accuracy and fullness. The practical 

 utility of Cassino's directory in the past seems 

 to be attested by the fact that it has reached 

 its nineteenth edition, and I therefore feel 

 justified in presenting, as a matter of general 

 information, some of the results of the exam- 

 ination. 



In order to judg« of its fullness a compari- 

 son was made with the contemporary lists of 

 a few scientific organizations whose member- 

 ship is carefully selected on the basis of scien- 

 tific ability or accomplishment. Of 90 mem- 

 bers of the National Academy of Sciences the 

 directory fails to include 28. It omits 129 of 

 the 329 members of the Washington Academy 

 of Sciences, 57 of the 233 members of the 

 American Society of Naturalists, 11 of the 46 

 fellows of the American Ornithologists Union, 

 and 54 of the 259 fellows of the Geological So- 

 ciety of America. Probably some members of 

 each organization are not within the scope of 

 the directory, but this remark can not apply to 

 such men as Outram Bangs, C. F. Batchelder, 

 Lyman Belding, Franz Boas, Lewis Boss, A. 

 P. Chadbourne, C. F. Chandler, S. C. Chand- 

 ler, W. W. Cooke, G. A. Dorsey, Wm. Butcher, 

 H. G. Dyar, H. W. Fairbanks, W. G. Farlow, 

 J. W. Fewkes, Henry Gannett, A. C. Gill, L. 



