636 



SCIENCE 



[N. S. Vol. XLIII. No. 1114 



the potash it contains. "We have oceans of 

 sea water carrying plenty of potash, if we 

 knew how to extract it. Don't say it can't 

 be done, for it is already done by miles of 

 seaweed. Why should we confine ourselves 

 to trying to take it away from the seaweed, 

 instead of learning what the seaweed knows 

 about getting it from the water ? You will 

 look supercilious, but until a large number 

 of chemists have studied semipermeable 

 membranes, there will always be this lack 

 of understanding of those simple reactions 

 of living matter going on around us. There 

 will always seem to me a possibility of doing 

 such physical and chemical processes more 

 nearly as we may wish to do them when we 

 know how these operate. 



When nothing new is being done by us it 

 wiU be a sure token of our decay. When 

 we stop increasing our experimental activ- 

 ities or fall for a considerable time behind 

 the activities of other countries, we may 

 expect to see our light become merely a 

 memory, like that of Greece or Rome. 

 Thus far we Americans have not reached a 

 fair average as investigators in natural sci- 

 ences, and yet we have incomparably supe- 

 rior conditions for the growth of research. 

 I can not look beyond the period when re- 

 search shall cease in a country and still 

 imagine that country a power in the world. 



There are no sharp lines to be drawn 

 through research to separate pure from 

 applied, scientific from practical, useful 

 from useless. If one attempts to divide 

 past research in such a manner, he finds 

 that time entirely rubs out his lines of 

 demarcation. At this particular time, how- 

 ever, one may imagine a more or less zigzag 

 zone which serves to divide research in a 

 commonly accepted way. 



In a manufactory the price of a new 

 product should include the cost of research. 

 No matter how complicated the system, this 

 is always true. Otherwise the industry 



would ultimately commit suicide. In prac- 

 tise it is common to apportion to particular 

 products the cost of their separate develop- 

 ment, and to fix the price so that within a 

 reasonable time, or by a reasonable volume 

 of sales, the so-called development cost may 

 be wiped out. Thereafter the product may 

 be sold on the basis of the continuing cost 

 of actual production. While this system 

 is extensive, it does not cover the cost of 

 many of those original researches which 

 may have been absolutely necessary. The 

 argon tungsten lamp, in its development 

 cost, did not carry the expenses of Rayleigh 

 and Ramsay's work, and so there will prob- 

 ably always be some such classification of 

 research work necessary. 



Under such a classification, the part of 

 research I am most interested in promoting 

 is what we may call the unpaid kind, not 

 because it is cheapest, but because it is the 

 most valuable. It is most neglected, most 

 poorly understood, most in need of appre- 

 ciative support in America. 



The separate industries do not need en- 

 couragement in research nearly so much as 

 the nation needs it. The industries can be 

 depended on to estimate its value to them, 

 for they take annual inventories. But a 

 country which keeps no books seems to have 

 to depend on instinct and environment for 

 its most valuable research work. 



It seems to me that our American col- 

 leges have been shortsighted in this respect. 

 This may foe explained by the rapidly in- 

 creasing demand in our growing industries 

 for analytical chemists and chemical engi- 

 neers, who could at once meet the existing 

 industrial requirements. This demand has 

 kept the chemical departments of our col- 

 leges and technical schools very busy with 

 the elementary and analytical side of chem- 

 istry and left little room for the synthetical 

 or experimental side. It has also naturally 

 tended toward the development of highly 



