September 8, 1923] 



NA TURE 



357 



from " experience " or " infinite objectivity " from 

 " finite subjectivity." 



These principles are called " qualitative/' but there 

 are also five " quantitative " ones, and three further 

 " anoegenetic " principles of reproduction, disparition, 

 and variation of clearness. I have space only for a 

 few remarks on these heads. 



The first three of the five quantitative principles run 

 as follows : (i) Every mind tends to keep its total 

 simultaneous cognitive output constant in quantity 

 however varying in quality ; (2) the occurrence of any 

 cognitive event produces a tendency for it to occur 

 afterwards ; (3) the occurrence of any cognitive event 

 produces a tendency opposed to its occurring afterwards. 



Of these (i) looks as if it meant that every sleepy 

 mind tended to have the same cognitive output as it 

 has when alert. This, however, is not what is meant. 

 Our author seems to mean instead (p. 131) that the 

 occurrence of any one noegenetic process tends to 

 diminish the others. The second noegenetic process, 

 however, presupposes and includes the first. 



(2) and (3) are flatly contradictory, so that it is 

 difficult to know what to do with them. They are 

 called respectively Retentivity and Fatigue. It may 

 be worth remarking, then, that Fatigue, in ordinary 

 language, does not contradict (2). When you are tired 

 you are likely to stop, but afterwards you may begin 

 again. 



I do not mean these criticisms to be verbal, but I 

 should be glad if they were. For the author's courage 

 and resource I have nothing but admiration, and his 

 vigour is always refreshing. John Laird. 



Our Bookshelf. 



Memoirs of the Geological Survey. Special Reports on 

 the Mineral Resources of Great Britain. Vol. i : 

 Tungsten and Manganese Ores. Third edition. By 

 Henry Dewey and H. G. Dines ; with Contributions 

 by C. N. Bromehead, T. Eastwood, G. V. Wilson, and 

 R. W. Pocock. Pp. iv + 83 + 3 plates. (South- 

 ampton : Ordnance Survey Office ; London : E. 

 Stanford, Ltd., 1923.) 2s. net. 

 The latest edition of the report on the British ores 

 of tungsten and manganese merely brings up-to-date 

 the information contained in the previous editions, but 

 it cannot be said that it has brought out any new facts 

 of importance. Perhaps it only serves to emphasise 

 the industrial unimportance of the British sources of 

 supply. Temporarily the War directed attention to 

 the domestic sources and caused these to be actively 

 worked, but under post-War conditions, the home 

 deposits have again been found to be unable to compete 

 with the richer deposits that exist abroad. This is 

 well exemplified by the ores of tungsten, the British 

 output of which touched nearly 400 tons per annum 

 during the War, while the output to-day is probably 

 less than a quarter of this quantity ; in the same way 



NO. 2810, VOL. I 12] 



the price, which during the War reached 555. per unit, 

 is to-day only about 12s. Furthermore, the total 

 output from Great Britain is only about 2 J per cent, 

 of the world's production. 



In the case of manganese ores the figures show the 

 same tendency, though not to so marked an extent ; 

 this is due in part to the fact that the great bulk of the 

 British ores of manganese are of low grade compared 

 to the imported ores. The chief centre of our home 

 supplies is in North- W^est Wales, in Carnar^' onshire and 

 Merionethshire, both of which districts are well described 

 in the present report. These ores appear to average 

 less than 30 per cent, of metallic manganese, while 

 imported ores contain at least 50 per cent. Even so, 

 however, the tonnage of domestic ores is barely i per 

 cent, of the world's production, and only about 2 per 

 cent, of our imports. Economically, therefore, the 

 British production of both these ores is negligible, 

 and a careful study of the report before us affords no 

 ground for hope that it will ever become a factor to be 

 reckoned with in the world's markets for either mineral. 



Cements, Limes and Plasters : their Materials, Manu- 

 facture and Properties. By E. C. Eckel. Second 

 edition, revised and partly rewritten. Pp. xxxi4- 

 655. (New York: J. Wiley and Sons, Inc. ; London: 

 Chapman and Hall, Ltd., 1922.) 32^. 6d. net. 



Eckel's treatise on cements, although dealing almost 

 exclusively with American practice, is one of the most 

 thorough works on this subject, and the new and 

 revised edition will be accepted as a standard authority. 

 The scanty references to English practice are not 

 always accurate, and the casual reader might suppose 

 that the industry in Great Britain was insignificant, 

 but the information in regard to the United States and 

 Canada is very full. A more detailed account of the 

 fixed mechanical kilns which are now assuming so much 

 importance would have been welcome, as they are now 

 becoming serious rivals of the rotary kiln. It has been 

 found possible to include a short account of the high- 

 alumina cements recently introduced, although there 

 is no systematic consideration of their properties. 

 The section on slag cements is misleading. Only the 

 older pozzolanic cements, consisting of mere mixtures 

 of granulated blast-furnace slag and lime, are considered, 

 and the much more valuable " Iron-Portland " or 

 " Blast-Fumace Portland " cements, made by mixing 

 a suitable granulated slag with clinker and grinding 

 together, are not even mentioned. The chemistry and 

 physics of cement do not receive attention, the treat- 

 ment being purely empirical, but within its limits the 

 book gives an excellent survey of an important American 

 industry. 



The Causes and Prevention of Corrosion. By A. A. 



Pollitt. Pp. 230. (London : Ernest Benn, Ltd., 



1923-) 255. net. 

 The literature of corrosion is extensive, but far from 

 satisfactory in its scientific aspects. There is a large 

 collection of facts, but a singular lack of co-ordinating 

 principles. Each experimenter has his own hypothesis, 

 which fits a small group of observations, but usually 

 breaks down when applied to other, nearly related 

 facts. The writer of the present work has prepared 

 a useful survey of the subject, although confining 



