2oS 



NATURE 



[August i8, 1910 



of atomic weights. These require to be known with 

 all the precision of which quantitative chemistry is 

 capable. 



It is only by such knowledge that we may hope to 

 find solutions to some of the most interesting and 

 important problems with which contemporary 

 chemistry is confronted. There is, to begin with, 

 the fundamental question of the validity of the law 

 of the conservation of mass. Is there a dissipation of 

 matter, as of energy, in a cycle of chemical changes? 

 Is an atomic weight a fixed and unalterable quan- 

 tity? Or is it, as first suggested by Marignac, a 

 statistical quantitv varying within limits, doubtless 

 verv small, but still possibly appreciable? 



There is further the perennial question of Prout's 

 law, which, like the poor, seems to be always with 

 us. Modern views of the genetic relations of the 

 elements and of the dependence of their properties 

 upon their relative masses are intimately connected 

 with the exact values and numerical relations of 

 atomic weights. It is these and similar questions 

 lying at the very basis of chemical philosophy that 

 render it imperatively necessary that these constants 

 should be known with the greatest possible precision. 

 The greatest possible precision is, of course, relative; 

 it depends upon the degree of perfection of contem- 

 porary quantitative chemistry, and as this is progres- 

 sive, each decade seeing improvements, both in the 

 application of old methods and in the discovery of 

 new, it necessarily follows that there is no such thing 

 as finality in measurements of this kind. A large 

 number of atomic weights are now known with accu- 

 racy to the first decimal place ; even in the case of 

 those of high values a considerable proportion indeed 

 are known even to the second decimal, and a few, 

 especially those elements which are habitually 

 employed as a basis of comparison in atomic-weight 

 work, as, for examples, silver and the halogens, are 

 being ascertained with a still greater exactitude. It 

 is only when atomic weights in general are known to 

 a like degree of precision that we can hope for definite 

 answers to such questions as have been indicated 

 above. 



It is largely due to the attention which this subject 

 has received in America that our present position has 

 been reached, and it is especially to the Harvard 

 School of Chemistry that we are indebted for the high 

 standard of accuracy which is now incumbent on every 

 worker in this field of determinative chemistry. No 

 laboratory in the world can point to such a remark- 

 able sequence of memoirs as those which are embodied 

 in the short synoptical statement in which Prof. 

 Richards has dealt with the Harvard determinations 

 of atomic weights between 1870 and the present year. 

 Initiated by the late Prof. Josiah Parsons Cooke, 

 whose determination of the atomic weight of anti- 

 mony is still regarded as the best ascertained value 

 for that element, the work has been continued by 

 his assistant and successor, Prof. Theodore Richards, 

 partly alone, but mainly in collaboration with pupils 

 whom he has trained and imbued with his own high 

 sense of exactitude. What the outcome of this work, 

 extending over many years, has been is abundantly 

 illustrated by the significant table on p. 90 of Prof. 

 Richards's memoir. Of the eighty-three elements at 

 present known, and of which the atomic weights are 

 given in the annual tables prepared by the Inter- 

 national Committee on Atomic Weights, no fewer 

 than twenty-eight of those estimations which are re- 

 garded by the committee as among the best ascer- 

 tained values are to be credited to the Harvard 

 laboratory. 



It is remarkable that this work should have been 

 NO. 2129, VOL. 84] 



done in America. It is commonly held that no nation 

 is more keenly appreciative of the utilitarian value of 

 science than .America ; but there is no money to be 

 made out of the results of an atomic-weight deter- 

 mination. It is quite impossible to evolve a new 

 colouring matter out of it, or to turn it into a syn- 

 thetic drug. Not even the "smartest" and most 

 enterprising of German chemists could bring it within 

 the protective infiuence of the mystic letters D.R.P. 

 On the contrary, atomic-weight work requires money, 

 and that frequently in no small amount ; platinum 

 vessels, and apparatus of transparent quartz, electric 

 ovens, high-class balances, and pure materials, render 

 such work extremely costly. No doubt Harvard is 

 well endowed, and Prof. Richards presumably has 

 been liberally supported by his university. But the 

 beneficence of the university has been largely supple- 

 mented by the action of the Carnegie Institution of 

 Washington ; without the pecuniary help afforded by 

 the Trust, the work could not, says Prof. Richards, 

 have been carried out on so large a scale, nor could 

 it have reached the degree of precision which it has 

 attained. T. E. T. 



TESTS FOR COLOUR VISION. 



THE agitation concerning the official colour-vision 

 tests for seamen has entered upon a new stage. 

 The Board of Trade has announced its decision to 

 hold an inquiry into the matter, and the personnel 

 of the committee has been published (Nature, 

 June 30, p. 529). 



After the reiteration of the confidence of the Board 

 in the certitude of the official tests, this change of 

 front comes somewhat as a surprise, but that can 

 be forgiven in the welcome possibilities of a revision 

 of tests that have forfeited the confidence of those 

 most concerned. 



It is not to be expected that the constitution of an 

 official committee will please everyone, and already 

 protest has been made by letter to the Board of 

 Trade from the secretary of the Imperial Merchant 

 Service Guild. The Guild protests that it was given 

 to understand that the projected new committee 

 would be of small size and of strictly impartial char- 

 acter, but that it proves to be large, and in the view 

 of the Guild heavily weighted. The Guild states that 

 of the members of the committee, at least two were 

 prominent supporters of the official tests in recently 

 disputed cases where the official position was 

 admittedly wrong. 



The choice of tests for colour vision is not so 

 simple as it iT\av seem at first sight. The difficulties 

 presented in appreciating the mental picture of a 

 colour-blind person are very great ; and complexity 

 is introduced bv the several .conflicting theories of 

 colour vision which, consciously or unconsciously, bias 

 the opinion of those who essay to determine these 

 tests. 



It is perhaps unfortunate that theories of colour 

 vision should enter into the question of colour tests, 

 at any rate, in the present state of our knowledge. 

 It is, of course, conceivable, nay, even probable, that 

 the true theory of colour vision when that is formu- 

 lated and proven, supposing it for the moment to be 

 none of those extant, will show an infallible means of 

 testing the sense of colour in any and every person. 

 Until then it would appear better that the test should 

 be frankly empirical, and, .so far as possible, im- 

 biased bv any theory. .And for the reason that, 

 unless rival theories be eliminated from the field, we 

 can scarcely expect a reasonable uniformity in our 

 tests. 



