1 



412 



NATURE 



[September 3, 1908 



The Dcinliardt-Schlomann Scries of Technical Dic- 

 tionaries in Six Languages, German, English, 

 French, Russian, Italian, Spanisli. By Kurt Dein- 

 hardt and Alfred Schlomann. Vol. iii., Steam 

 Boilers, Steam Engines, Steam Turbines. Edited 

 by Wilhelm Wagner. Pp. xi+1322; illustrated. 

 (London : A. Constable and Co., Ltd., 1908.) Price 

 16s. net. 



The third volume of Deinhardt and Schlomann 's 

 illustrated technical dictionary cannot fail to prove of 

 er^ormous value in international relations. The idea 

 is quite original. The dictionary is in six languages, 

 German, English, PVench, Russian, Italian and 

 Spanish, and the necessity for elaborate definitions is 

 obviated by the insertion of small sketches. The 

 vocabularies are classified under the following 

 heads : — A. Steam boilers : (i) fuels, (2) production of 

 heat, (3) furnace installations, (4) transmission of 

 heat, (5) evaporation, (6) testing of materials, (7) 

 boiler construction, (S) types of boilers, (9) boiler 

 fittings, (10) boiler erection, (11) feed apparatus, (12) 

 steam superheaters, (13) management of steam boilers, 

 (14) boiler explosions, (15) boiler inspection, (16) range 

 of steam pipes. B. Steam engines : (17) theory, (18) 

 machine parts, (19) condensers, (20) tvpes of engine, 

 (21) erection, (22) working of engines. C. Steam tur- 

 bines : (23) theory, (24) steam-turbine parts, (25) 

 turbine plant. 



The volume concludes with an alphabetical index 

 of the German, English, French, Italian and Spanish 

 terms, showing the page and paragraph at which each 

 word is to be found. .\ second index is devoted to the 

 Russian words. The type is small but clear, and the 

 volume is tastefully bound. Well-known experts all 

 over the world have helped to revise the text, and the 

 result is that Deinhardt and Schloniann's technical 

 dictionary is undoubtedly the most accurate that has 

 yet been published. 



Das Wetter und seine Bedciitiing fiir das prahiischc 

 Leben. By Prof. Carl Kassner. Pp. vi -I- 148 ; 

 illustrated. (Leipzig : Quelle and Meyer, 1908.) 



This book is No. 25 of the series "Wissenschaft und 

 Bildung," edited by Dr. Paul Herre. It is published 

 to meet the wish for information respecting the prin- 

 ciples of weather prediction consequent upon the 

 establishment of the public weather service in the 

 German Empire (June, 1906). The Deutsche See- 

 warte, Hamburg, has issued an excellent dailv weather 

 report since February, 1876, for the whole' of Ger- 

 many, but the new system divides the Empire into 

 fifteen forecast districts. The first section of Dr. 

 Kassner's work deals with the historical development 

 of weather prediction, and contains a concise sum- 

 mary of Dr. Hellmann's valuable researches into this 

 subject, from the Babylonian era, some 4000 years 

 B.C. This is a new departure from the usual course, 

 and will be very useful for reference. Part ii. deals 

 with the bases of modern weather prediction, with 

 the preparation of weather charts, with areas of high 

 and low barometric pressure, &-c. This portion of 

 the work, although not new, will be found most useful 

 to students._ The author has consulted all -available 

 sources of information, and presents a lucid cxposi- 

 tiori of the present state of the science ; various popular 

 notions, such as the supposed direct influence of the 

 Gulf Stream and of the existence of icebergs on our 

 weather (also lunar influence, part i.), are satisfactorily 

 dealt with. Part iii. deals with the importance of 

 weather in public and private life, and contains much 

 that is not usually found in similar works. 



NO. 2027, VOL. 78] 



LETTERS TO THE EDITOR. 



[The Editor does not hold himself responsible for opiniom 

 expressed by his correspondents. Neither can he undertake 

 to return, or to correspond with the writers of, rejected 

 manuscripts intended for this or any other part of Nature. 

 No notice is taken of anonymous commtttiicaiions.] 



Indiscriminaie Criticism, 



May I enter a mild protest ag.iinst a habit that seems 

 to me somewhat harmful? Many people, when they obtain 

 results w-hich disagree with the previous w-ork on the sub- 

 ject, seem to think it incumbent upon them to suggest that 

 the disagreement comes from some flaw of reasoning or 

 method in the work used for comparison. 



This habit, especially in experimental work, where the 

 chances against accuracy are always so great, is really a 

 hindrance to the advancement of knowledge ; " any stick 

 is good enough to beat a dog with," and so the slight ■ 

 approximation to truth that there may be in the depreciated 

 work will certainly not be enhanced by indiscriminate 

 criticism, and may, thereby, be overlooked. 



Surely it is understood that no one publishes results 

 which disagree with those previously obtained without 

 having a strong belief that his own work is, in one way 

 or another, an advancement — why not, therefore, leave it 

 at that? .'\ saving of time and words would resuh. 



I need scarcely say that I am not deprecating careful 

 criticism ; on the contrary, those who from facts and 

 figures will point out where a source of error may lie 

 confer a benefit on science. Unfortunately, people in the 

 position to do this seem to shun tjie task. 



What I have alluded to in the previous paragraphs may 

 well be illustrated by examples drawn from criticism passed 

 on researches lately published by Mr. Hartley and myself. 



(i) Prof. Callendar (Proc. Roy. Soc, vol. Ixxx., p. 466), 

 when comparing our experimental results with an 

 approximate calculation of his own, says, " the numbers 

 in the last column appear to indicate a systematic error 

 in the experimental numbers for strong solutions"; now 

 it is possible, nay, probable, that there are systematic errors 

 in the w'ork, but it seems to me more would be done to 

 elucidate the truth if he had either pointed out where the 

 errors are or stated to what degree of accuracy his 

 approximate calculation can be relied upon. 



{2) Mr. O. F. Tower (Jour. Am. Ch. Soc, August, 

 p. 1225) says, " Lord Berkeley and Mr. Hartley's results 

 for the vapour pressures of the solutions are undoubtedly 

 too low, showing probably that with their apparatus the 

 air was not completely saturated," &c. This is a flagrant 

 example ; for Mr. Tower on the same page gives a table 

 containing three sets of results, one obtained by Smits 

 by a direct method, ours by an indirect one, and his own 

 results, which are direct. Now Smits's numbers and ours, 

 for the same concentration, give values (by different 

 methods, be it noted) which are within i per cent, of each 

 other, while Mr. Tower's numbers differ from the other 

 two by nearly 10 per cent. ! 



.'Vs this letter is concerned with a more important matter 

 than the accuracy of our work, I will not discuss Mr. 

 Tower's assumption that the air was not saturated, &c. 



(3) Messrs. Lincoln and Klein (Jour. Phys. Chem., xi., 

 iqo7) say,' when deducing the relative value of Prof. 

 Kahlenburg's method of determining vapour pressures and 

 ours, that Mr. Hartley and I claim an accuracy of 5 per 

 cent. They omit, however, to state that our experimental 

 error is (ultimately) calculated on the difference between 

 the vapour pressures of solution and solvent, while theirs, 

 if reduced to the same basis, would show errors of 25 per 

 cent. 



(4) Prof. Morse (.Amer. Chem. Journ.. vol. xx.\iv., 1005, 

 pp. 24-25) s.ays, " we wish to speak of certain objections 

 to a method of measuring osmotic pressures which has 

 sometimes been practised. The attempt has been made, on 

 several occasions, to ascertain the pressures of solutions 

 bv bringing mechanical pressure to bear upon the contents 

 of the cell until," &c., and then .goes on to raise purely 

 imaginary objections to the method. It is interesting to 



^ Being away from h'me I .Tin unable to give the exact wording. 



