142 



NATURE 



[i\PRIL I, 1920 



seconds. This already high degree of precision 

 was, however, surpassed in the particular section 

 under notice. Thus over a total of sixty-eight 

 triangles the average closing error was o'63" and 

 the maximum error of any one triangle i'86". 

 This pitch of excellence was moreover attained 

 without any increase of time spent at the stations ; 

 indeed, it is claimed, we think with justice, that 

 the arc establishes a "record,*' both technically 

 and financially. The average cost per point occu- 

 pied was, in fact, lower than has been attained 

 with any previous work of the same class, and 

 as, owing to the nature of the country, higla and 

 expensive signal scaffolds were necessary, it seems 

 that the reduction made in the cost of the actual 

 observing was even more notable than appears on 

 first inspection. 



This conveys a lesson which may be taken to 

 heart by those responsible for future survey opera- 

 tions. It seems clear that the difference in method 

 of execution between what we are accustomed to 

 call first order or primary triangulation {i.e. tri- 

 angular error under i") and secondary work 

 (triangular error under 3") lies mainly in the nature 

 of the signals. If lamps only are used it is a 

 matter of indifference as regards rate of progress, 



and hence as regards cost, whether a large in- 

 strument capable of first order precision or a 

 smaller one capable only of second order is used. 

 In either case one observer can complete the ob- 

 servations at a station in one night, and no 

 reduction in size of instrument, in number of rounds 

 taken, or in order of accuracy will enable him to 

 do more. The difference in cost of transport 

 between the two instruments is in most cases 

 negligible. The only extra cost involved is that 

 caused by the necessity of providing five lamp- 

 men or lamp parties and moving them from point 

 to point. In rough country this might undoubt- 

 edly prove a formidable addition, but in the case 

 of future boundary commissions or land surveys 

 in Africa it is anyhow worth serious consideration 

 whether a backbone or net of primary triangulation, 

 planned so as to fit in with a comprehensive 

 geodetic scheme, cannot be undertaken without a 

 prohibitive increase in expenditure. 



This is the sort of question for which the co- 

 ordinated experience and authority of a geodetic 



I institute would prove invaluable, and it is ro be 

 hoped that it will not be long before such an 



i institution, long overdue, is established in England 

 for the British Empire. E. H. H. 



Science and Research in the Air Service. 



'X'HE Air Estimates for the year 1920-21, 

 -*■ recently presented to Parliament, show a 

 total estimated expenditure of 21 million pounds 

 compared with 54 million pounds in the previous 

 year. The apparent saving in cost is 33 millions, 

 but it is really greater, for in the year 1919-20 

 the cost of the experimental and research services 

 was borne jointly by the Admiralty and the 

 Ministry of Munitions, and is now wholly included 

 in the Air Ministry vote. 



As regards the Royal Air Force, the number 

 of officers, warrant officers, non-commissioned 

 officers, airmen, and boys provided on the estab- 

 lishment (exclusive of those serving in India) has 

 fallen from 150,000 in 1919-20 to rather less than 

 30,000 in 1920-21 — a striking reduction. The 

 21 million pounds for the new financial year 

 includes rather under a million for civil aviation 

 and two and a half millions for experiment and 

 research. This latter sum would have been more 

 than three millions (3,177,000^.) had not an 

 "appropriation in aid," due to the sale of certain 

 airships for 6oo,oooI., come to the relief of the 

 vote. The actual figures are as follows :■ — 



Liquidation of war liabilities ... ... 1,334,000 



Works, buildings, and lands 140,000 



Aeronautical inspection 80,300 



Airship constructional establishment... 315,000 



R.A.E., Farnborough 401,200 



Technical equipment and materials ... 844,390 



Salaries and wages 48,800 



Miscellaneous 13*850 



NO. 2631, VOL. 105] 



3,177,540 



An expenditure of more than three millions 

 for research alone in a single year would appear to 

 be a generous provision, but an examination of the 

 foregoing figures shows that much of the expendi- 

 ture will not be employed for this purpose. 



The air vote for meteorological services has 

 risen from i2,oooZ. in 1919-20 to 77,629!. in 

 1920-21, and part of this will doubtless be used 

 in research work of some kind, though these 

 services are not part of the research directorate, 

 but come under the civil aviation side of the Air 

 Ministry. The sum of 77, 629L includes the pro- 

 vision of only 358Z. for "experimental stations," 

 which is such a very modest amount that we as- 

 siume experimental research in meteorology is 

 provided for by other aid. In any event, the 

 amount cannot represent the degree to which at- 

 tention is given to research, since in meteorology 

 there is ample scope for original work based upon 

 the observations from what may be termed routine 

 stations. 



The printed Estimates convey the intention of 

 the Government to make liberal provision for 

 research in aeronautics, but it is impossible to 

 determine precisely what sum of money is thereby 

 devoted solely to "experiment and research," 

 since such work is sometimes carried on at the 

 ordinary air stations. Moreover, 40,000!. for the 

 National Physical Laboratory is not borne on the 

 Air Estimates at all, but on those for the Civil 

 Service. The Estimates do, however, include the 

 sum of 20,340!. for research "grants to scientific 

 bodies," and 6oo,oooZ. as an encouragement to 

 invention. 



