November 9, 191 1] 



NATURE 



0/ 



points of the periclinal forms been able to prove that 

 some of the layers have the chromosome numbers 

 proper to S. nigrum, and others those of Lycopersicum. 

 We can scarcely doubt that this remarkable series 

 of observations will pass into the classics of biology. 

 On similar lines Baur proposes to elucidate the old 

 problem of Cytisus Adami and Crataego-mespilus, the 

 two traditional examples of "graft-hybrids." The 

 former, for instance, is regarded as a Laburnum 

 enclosed in a sheath of C. purpureus. On occasion, 

 as when the exterior is wounded, the Laburnum can 

 come out and develop. Baur's idea is doubtless a 

 part of the truth, but I cannot clearly apply it to all 

 the phenomena which Adami presents, especially to 



; the sexual vagaries which it shows in this country at 

 least. In it the Adami flowers have good pollen but 



' no good ovules ; the purpureus flowers have the female 



• parts developed, but the anthers bad ; while the Lahur- 

 ( num flowers are perfect and set seed in plenty. Nor 



I do I clearly understand the origin of the purpureus 

 ' branches. One can scarcely help suspecting that in 



the segregation by which these phenomena are pro- 

 duced there is some complex repulsion between the 



i sex-factors and the factors for colour or form, com- 

 parable with the distinctions now known to exist be- 

 tween the genetic constitution of pollen and ovules of 

 the same individual in several cases, e.g. Stocks and 



t Petunia (Miss Saunders), or CEnothera (de Vries). 



* The only point in this excellent book which calls 

 for criticism is, in my judgment, the rather crude 

 representation of segregation as effected by chromo- 

 somes. These pictures will live in the memory of the 



t reader, and tend to limit his imagination of the possi- 

 bilities more closely than the known facts at present 

 warrant. W. Bateson, 



THE RUSTING OF IRON. 



II The Corrosion of Iron and Steel. By Dr. J. Newton 

 r Friend. Pp. xiv + 300. (London : Longmans, 

 M Green, and Co., 191 1.) Price 6^. net. 



nPHE author gives a concise account of all the 

 *- important work that has been carried out in 

 ii'imection with investigations relating to the causes 

 rorrosion of iron and steel. The book is a model 

 its kind, since the references to original contribu- 

 !s to knowledge are exhaustive and will serve to 

 rt the investigator to the literature of that branch 

 the subject in which he is specially interested. 

 ' r dealing with the action of air, of water and of 

 iin on iron, the various theories which have been 

 meed to explain corrosion are discussed, and the 

 iclusion is drawn that "the most recent experi- 

 ntal results are entirely in favour of the acid theory 

 (orrosion." 



I he action of acids and of alkalis, and the influence 



solutions of salts of various kinds on iron are next 



i^idered, and a short chapter is then devoted to the 



ion of oils on the metal. In dealing with the sub- 



' of the passivity of iron the author has failed to 



' tke clear the fact that the immersion of the metal 



ill chromic acid inust necessarily bring about the 



^ iremoval by oxidation of such impurities as manganese 



' 'Sulphide, which exist on the surface, and form 



NO. 2193, VOL. 88] 



acids on exposure to moist air. The metal by such 

 treatment must in consequence be rendered more re- 

 sistant to corrosion irrespective of any question of 

 passivity. Nor is attention directed to the fact that 

 the surface of iron immersed in chromic acid must 

 necessarily remain bright whenever the acid is suffi- 

 ciently concentrated to dissolve any rust which might 

 be formed, and in this connection it may be noted that 

 ferric hydroxide dissolves readily even in dilute solu- 

 tions of chromic acid. Moreover, H. B. Baker and 

 others have clearly shown that whatever properties 

 are given to iron by immersion in dilute chromic acid 

 immunity from rusting is not one of them, and it is 

 extremely doubtful if previous immersion in chromic 

 acid of any strength is a protection against atmospheric 

 oxidation of iron. 



The last three chapters of the book deal respectively 

 with the influence of chemical composition on the 

 corrodibility of iron, with electrical effects and with 

 the relative rate of corrosion of iron and steel. The 

 first of these might with advantage have been con- 

 sidered at an earlier stage — the author states that it 

 is of paramount importance — since an explanation of 

 that troublesome form of corrosion known as pitting 

 in water-tube and other boilers must be sought mainly 

 in the chemical composition of the iron of w^hich they 

 are made. The chapter on electrical action opens with 

 the unfortunate statement that, "as is well known, 

 when an electric current passes through water, the 

 latter is readily split up into its constituent elements, 

 oxygen and hydrogen." This erroneous inclusion of 

 water amongst electrolytes may produce confusion in 

 the minds of some readers, and is greatly to be 

 regretted. 



The book is well illustrated throughout, and will 

 appeal to the general reader of scientific literature 

 since it contains matters of interest apart from tech- 

 nical detail. For instance, an account is given of the 

 Iron Pillar of Delhi, dating from 912 B.C., as an 

 example of iron which has for centuries resisted atmo- 

 spheric attack. There is, however, no record of the 

 very rapid corrosion of the steel pipe line which con- 

 veys water from Mundaring to the Kalgoorlie 

 Goldfield in Western Australia, and represents an 

 outlay of upwards of 3,000,000/. sterling. Some 

 account of the reports presented to the Government of 

 Western Australia on this matter might with advan- 

 tage have been cited, and the suggested treatment of 

 the water by deaeration and liming, involving an 

 expenditure for machinery of 187,000/., discussed. 



G. T. M. 



ASPECTS OF THE EARTH'S STORY. 

 The Changeful Earth : an Introduction to the Record 

 of the Rocks. (Readable Books in Natural Know- 

 ledge.) By Prof. G. A. J. Cole. Pp. x + 223. (Lon- 

 don : Macmillan and Co., Ltd., 191 1.) Price is. 6d. 

 IT is refreshing to turn from the ordinary text-books 

 of science— useful and necessary as such works 

 undoubtedly are— to a booklet like that now before us. 

 Science manuals in their efforts after inclusion and 

 compression, in order to meet the wants of examina- 

 tion candidates, tend to become dogmatic in their 



