.March 4, 1920] 



NATURE 



ouf*ln also to possess expert advisers in many branches 

 of science. Some kind of scientific service will there- 

 fore be needed, but this should not be allowed to 

 engross the whole field ; and the best results are sure 

 to be obtained in the future, as they have been in the 

 past, by untrammelled men of capacity working as 

 they please. Ronald Ross. 



In discussing the best ways of fostering research 

 work it is important to remember that, the word 

 "research" is used in two widely distinct senses: it 

 may stand either for the careful collection of observa- 

 tions, or for the deduction of the principles expressing 

 the relationship between one set of phenomena and 

 another. The difference between them is like that 

 between the discovery of a new country and the careful 

 mapping of one known in a general way but not in 

 full detail. 



It would be unfair to set either of these kinds of 

 research above the other; each is indispensable to the 

 other. Experience shows, however, that the power 

 to collect careful observations can be imparted to a 

 large number of men and women, while the power to 

 utilise the material and deduce from it anything more 

 than the comparatively obvious is rare and cannot 

 be imparted. Further, this ability is not equally 

 divided as between different classes of men or as 

 between men and women. 



The recognition of the necessity for each kind of 

 ability is essential to the proper conduct of a research 

 institution, and one of the g-reat difficulties is to find 

 deducers of new ideas and to ensure that thev shall 

 work harmoniously with the equally necessary, but 

 less rare, collectors of observations. The difference 

 between the two groups of workers is fundamental 

 and far-reaching, manifesting itself even in trivial 

 daily actions. One difference is particularly important 

 for the present discussion : the first group £?reatly 

 resent immediate direction ; the second do not, pro- 

 vided they see advantages therein. In all research 

 institutions of any size the chief problem is to keep 

 both groups of workers as nearly abreast as possible. 

 Deductions made in advance of facts are often wrong 

 and sometimes harmful. Facts and observations 

 accumulated without any illuminating hypothesis or 

 general principle are rather dreary and soon foreotten. 

 It is one of the tragedies of a life devoted to science 

 that so often the fruit falls stillborn and is entombed 

 in some journal, never again to see the light. We 

 have all known such workers : 



And, as year aft^r rear 

 Fresh nro-'iicts of their barren labour fall 

 From their tired hands, and rest 



Never vet com- s more near, 

 Gloom settles slowly down over their breast. 



The only way of avoiding the tragedy and its 

 accompanying waste is to ensure that both' groups of 

 workers keep together. 



It is not only between these two groups, however, 

 that co-operation is necessary; under modern condi- 

 tions there must also be close relationship between 

 the workers in different subjects. Science is becoming 

 increasingly specialised ; no one man now knows 

 much of any subject except his own. For the inves- 

 ti£?ation of phenomena such as those of agriculture, 

 which lie outside the present arbitrary divisions of 

 science, recourse must be had to team-work; a bodv 

 of voung workers whose minds are still elastic must 

 be interested in the problem and induced to work 

 together for its solution. 



Experience shows that successful co-operation is 

 achieved only when a deliberate attempt is made to 

 secure optimum conditions for each individual worker. 



NO. 2627, VOL. \oK\ 



How can a State system be adapted to fit these various 

 necessities? For financial reasons complete elasticity 

 is impossible; Treasuries must know their liabilities. 

 In any Civil Service system promotion is almost in- 

 evitably by seniority. Individual action and thought 

 would be intolerable; everything must go throueh a 

 chief, while anything repugnant to him must be sup- 

 pressed. In all these directions the State system is 

 absolutely incompatible with living research, although 

 it might be consistent with much careful accumulation 

 of facts, with survey work, and with the establish- 

 ment of some central collecting institute. For these 

 reasons I cannot believe that the intensely centralised 

 system proposed for India could succeed!! One man 

 may organise work in one institution where he is 

 accessible to the staff morning, noon, and night; but 

 he would indeed need to be a superman of most exalted 

 degree if he aspired to direct the research v<?ork of a 

 country. 



The system devised by the English Ministry of Agri- 

 culture is, in my view, much better. It possesses 

 some degree of financial elasticity. While it contains 

 the inevitable regulation about promotion by seniority, 

 this is qualified by clauses under which the best man 

 available can, nevertheless, be appointed to fill a 

 vacant post. There is no attempt to govern from 

 Whitehall ; no general director, deputy director, or 

 other official to run the research workers, but only 

 occasional friendly gatherings of the chief officers to 

 discuss common problems. Could not some such 

 system be tried in India? E. J. RussF.i.i.. 



Rothamsted Experimental Station. Harpenden. 



The question of reorganising and developing 

 scientific work in India discussed in the leading 

 article in Nature of February 19 is of the utmost 

 importance to all concerned with the welfare and 

 scientific reputation of the Empire. Now that there 

 is a prospect of recognition by the Government of 

 India and the Secretary of State of the necessity for 

 increased expenditure on scientific investigation, it is 

 essential that the new era should be inaugurated 

 under the most favourable conditions. Two policies 

 are apparently under consideration, which may be 

 referred to respectively as centralisation and decen- 

 tralisation ; these are clearly defined in the article of 

 February 19. The advantage of organising research 

 within certain limits is g-enerally admitted ; facilities 

 should be afforded for supplying information, for sug- 

 gesting problems, and for the co-ordination of the 

 activities of individuals or institutions, but it would 

 seem that the policy of centralisation advocated by 

 the Indian Industrial Commission, presided over by 

 Sir Thomas Holland and " favoured by a number of 

 administrators," is much more than this. It is. in 

 short, a proposal to bring scientific investigation into 

 line with routine official work— a procedure which, 

 one learns with surprise, has the support of several 

 scientific witnesses examined by the Commission. 

 If there is one thing vital for the successful prosecu- 

 tion of scientific research of the best type and for the 

 encouragement of the full development of a re- 

 searcher's capacity, it is freedom of action. 



It is safe to predict that verv few men possessing 

 what may be called the research temperament would 

 consent to submit to a bondage that would be not 

 onlv irksome and irritating, but also fatal to indivi- 

 dual initiative and enthusiasm. If adequate remunera- 

 tion is offered and reasonable laboratory facilities are 

 provided, good men will be easily secured. Given 

 the richt sort of men, T venture to think that the 

 only rational course is to trust them to work out in 



