292 



NATURE 



[May 6, 1920 



Letters to the Editor. 



[77je Editor does not hold himself responsible for 

 opinions expressed by his correspondents. Neither 

 can he undertake to return, or to correspond wtlh 

 the writers of. rejected manuscripts intended for 

 this or any other part of Nature. No notice is 

 taken of anonymous communications.] 



Organisation of Scientific Work. 



I HAVE only recently seen the article in Nature of 

 February 19 and the correspondence so unanimously 

 supporting the view that the present decentralised 

 system of team work by experts in different branches 

 of science in agricultural, forestry, and medical re- 

 search institutes is greatly superior to the proposed 

 centralisation in distant Simla of each separate science 

 — chemistry, botany, etc. — under directors of research 

 with autocratic powers to decide what each original 

 worker in his branch throughout India shall investi- 

 gate and publish ; for it is clearly impossible in these 

 days for one man to be suiiliciently conversant with 

 each special division of his science adequately to fulfil 

 such a stupendous task. I desire to associate myself 

 with that view, which may be illustrated by my 

 experience in organising the Calcutta School of 

 Tropical Medicine, shortly to be opened, for which I 

 have just obtained endowments from two successful 

 European and Parsi business men for a whole-time 

 biochemist, in addition to two other chemists for the 

 analysis of indigenous drugs and of food and water 

 respectively, all three of whom will aid nine medical 

 research investigators in team work at important 

 medical problems under a medical director. Could 

 anyone contend that these very specialised chemists 

 would be better controlled by a purely chemical 

 director a thousand miles away in Simla, who could 

 know nothing of the medical problems they will 

 investigate ? 



On the other hand, if the Government of India is 

 to provide the large sums urgently required for the 

 further development of scientific research in India, 

 it will require some organisation to co-ordinate and 

 report on the work it \yill be financing. May it 

 not learn a lesson from the Medical Research Com- 

 mittee of eminent medical men of science, which is 

 wisely utilising the large sums supplied by the British 

 Government in assisting the investigations of univer- 

 sity and medical-school workers with established 

 reputations and with a minimum amount of inter- 

 ference? A very similar and successful organisation 

 was set up in India when the late Sir Pardey Lukis 

 persuaded the Government of India to hand over five 

 lakhs (some 50,000?.) a year to the Indian Research 

 Association, administered by a governing body on 

 which the medical members, through their special 

 knowledge of the subject, exercise a preponderating 

 influence; while I have recently obtained a purely 

 medical governing body to administer the endowments 

 of the Calcutta School of Tropical Medicine with an 

 income of some ii,cooZ. a year, which I have been 

 fortunate enough to raise to provide both men and 

 apparatus without the long delays, usually of several 

 years' duration, involved in obtaining the sanction of 

 the Government of India and of the Secretary of 

 State for new posts. 



This plan has thus already proved its value and is 

 capable of extension, while "boards composed of a 

 number of men of science of high standing will corn- 

 mand much greater confidence than an autocratic 

 director of research. The nucleus of such a body 

 already exists in the Board of Scientific Control, which 

 meets twice a year in Delhi and Simla, and might 

 with greater advantage holJ its principal meeting 

 NO. 2636, VOL. 105] 



I coincidently with the Indian Science Congress in one 

 j of the large centres of research. At a meeting in 

 Delhi in December, 1918, 1 advocated, in place of an 

 autocratic director of medical research, that an in- 

 spector of research might be appointed, who would 

 not attempt the invidious task of laying down what 

 each research worker should investigate and publish, 

 but would visit different research laboratories and 

 consult with their respective directors or councils 

 regarding the financial and other needs of the insti- 

 tutes, and help to co-ordinate the work in different 

 parts of India to prevent waste through overlapping. 

 The recent correspondence in Nature confirms me in 

 that solution of the difficulty, and I venture to think 

 that in some such ways as I have suggested the 

 established advantages of the present decentralised 

 system may be retained and strengthened by greater 

 and more ' elastic financial aid, and be better co- 

 ordinated, without introducing the highly objection- 

 able autocratic and distant centralised control of the 

 proposals now before the Indian Government to which 

 you have directed such timely attention. 



Leonard Rogers. 

 South Devon, April ?.g 



The Small islands of Almost-Atolls. 



The familiar inductive series of fringing _ reefs, 

 barrier reefs, and atolls may be further subdivided so 

 as to contain six members : Normal fringing reefs, 

 offshore fringing reefs, narrow-lagoon barrier reefs, 

 broad-lagoon barrier reefs, almost-atoUs, and atolls. 

 Almost-atoUs, or atoll-like reefs encircling lagoons in 

 which one or several small islands rise, are of interest 

 as affording a critical test of certain competing coral- 

 reef theories, as follows: Murray's theory of out- 

 growing reefs around still standing islands explains a 

 completed atoll by supposing that the original volcanic 

 island is slowly worn down as the encircling reef 

 grows outward and the lagoon is excavated behind 

 the growing reef by solution, the degraded central 

 island eventually disappearing in a way not clearly 

 explained, perhaps by outwash of its detritus from the 

 lagoon by currents which are fed by the influx of surf 

 over the windward reef and discharged by outflow 

 through passes in the leeward reef. Under this 

 theory the small island of an almost-atoll would be a 

 nearly worn-down central island, which would exhibit 

 rolling hills of low relief surrounded by delta flats; 

 or in a later stage, after the delta deposits had_ been 

 swept away, the low hills of the vanishing island 

 would be encroached upon by the lagoon waves and 

 cut back in low bluffs fronted by low-tide rock plat- 

 forms that gradually deepen into the lagoon. 



According to Daly's Glacial-control theory, atoll 

 reefs are built up from the margin of platforms 

 abraded bv the waves of the lowered Glacial ocean 

 across still standing islands that had been previously 

 worn down to low relief by long-continued normal 

 erosion, the reefs being built up as the ocean rises 

 in post-Glacial time. Under this theory an almost 

 destroyed central island would have a surface of 

 rolling hills, cut back by cliffs which would now — 

 except for fringing reefs that may border them— 

 plunge into the lagoon waters to a depth of twenty 

 or more fathoms. This inference is well supported by 

 the occurrence of strongly clift islands surmounting 

 submarine banks of moderate depth in the extra- 

 tropical seas. According to Darwin's theory of up- 

 growing reefs on intermittently subsiding- foundations 

 Submergence bv subsidence being faster than 

 erosional degradation — atolls are produced when the 

 central island of an up-growing barrier reef has sunk 

 out of sight. Under this theory the lagoon of an 



