December 



1915] 



NATURE 



3«i 



steels met with in practice. The same difficulty occurs 

 in regard to Prof. Tschischewski's paper, which does 

 not, therefore, set at rest the vexed question whether 

 the very minute quantities of nitrogen which are 

 found in industrial steels — and particularly in those 

 made by the Bessemer process, are really responsible 

 for the injurious etTects which are sometimes ascribed 

 to them. It was therefore satisfactory to hear that 

 the whole question of the influence of nitrogen was to 

 be placed on a more satisfactory basis by systematic 

 research under the auspices of Dr. J. E. Stead. Mean- 

 while, the results of the Russian investigator furnish 

 the best available data in regard to nitrogen in steel. 

 In order to introduce nitrogen into steel. Prof. Tschi- 

 schewski found it necessary to expose the heated 

 metal to ammonia vapours, since free nitrogen appar- 

 ently does not combine with iron except, possibly, at 

 very high temperatures. Incidentally, this result is of 

 importance from the point of view of experiments on 

 iron and its alloys at moderately high temperatures, 

 since it indicates that an atmosphere of pure nitrogen 

 would be without action on the material. 



Interesting from what is, at first sight, an entirely 

 difTerent point of view, are the results obtained by 

 Mr. Byrom in his observations of the carburising action 

 of blast-furnace gases at temperatures not exceeding 

 500° C. Hitherto it has been generally held that the 

 carburisation of iron does not take place until the 

 temperature of Ac, — about 700° C. — is exceeded. The 

 explanation given has been that since iron carbide is 

 not soluble in alpha iron, which can alone exist at 

 such lower temperatures, carburisation would not 

 occur, or, if it did, must remain strictly confined to 

 the surface. Mr. Byrom, however, with the co- 

 op>eration of Dr. Stead, has shown that in the stream 

 of gases which come from a blast-furnace, iron be- 

 comes rapidly carburised, and is, in fact, converted 

 into a carbide of iron so rich in carbon that the 

 presence of a carbide, Fe.C, in addition to the well- 

 known Fe,C is suspected. There is, however, a 

 simultaneous increase in the sulphur-content of the 

 material. 



The contradiction between these observations and 

 the previously accepted views is not so great as at 

 first sight appears. An examination of the iron after 

 partial carburisation bv these gases at once shows 

 that there really is no diffusion of the carbides through 

 the iron, but that the carbide is formed in situ bv 

 the interaction of the gas, which has diffused through 

 the iron as such, and the iron immediately in contact 

 with it. It is, further, very doubtful indeed, if it has 

 not actually been disproved, whether any such car- 

 burising action would occur if the blast-furnace gases 

 were replaced by pure carbon monoxide ; it is almost 

 certain that some of the other gases present in the 

 blast-furnace, such as the carbon oxv-sulphide which 

 Mr. Byrom suggests, play an important part in the 

 reaction as "catalytic agents." 



This consideration also affects a theoretical bearing 

 to which attention was directed by Prof. Carpenter 

 in the discussion on Mr. Byrom 's paper. The point 

 in question is that of the supposed " meta-stable " 

 character of iron carbide. This conception of the 

 nature of the carbide, and its tendency to dissociate 

 into free carbon and iron, lies at the base of the 

 widely accepted equilibrium diagram of the iron- 

 carbon system. If. however, it could be shown that 

 iron and carbon could unite to form iron carbide at these 

 moderate temperatures, then the idea that the carbide 

 is meta-stable at these temperatures would have to 

 be abandoned. As was pointed out by Dr. Rosenhain, 

 however, the results obtained by Mr. Bvrom do not 

 at all establish such a fact; all that thev do establish 

 IS that in the complex system consisting of iron, 

 carbon monoxide, and a number of other gases, iron 

 NO. 2405, VOL. 96] 



carbide can be formed at temperatures in the neigh- 

 bourhood of 500° C. But it is quite possible, and 

 even probable, that the equilibrium ranges of such 

 a substance as iron carbide may be very considerably 

 altered in the presence of three or more components. 

 The fact that the presence of another component does 

 frequently alter or depress the lines or surfaces of 

 equilibrium in a thermal diagram is well known, and 

 it may be that what is frequently termed " catalytic 

 action " is simply due to such an effect. Therefore, 

 although interesting and important in themselves, 

 Mr. Byrom 's results appear to leave the question of 

 the stability or* otherwise of iron carbide much where 

 it was before. 



The two papers on the nitrogenisation and the 

 carburisation of iron brought out one common feature 

 of great interest. In both it was shown that the 

 gases penetrated along the boundaries of the crystals 

 of the metal and from these spread into the masses 

 of the crystals along the cleavage planes. Both the 

 carbide and the nitride of iron exhibit the distribution 

 typical of such action most clearly — so much so that 

 attack by means of gases would seem to offer possibili- 

 ties for the study of the crystallographic data of the 

 material by rendering visible certain cleavage planes. 



The phenomenon is, however, interesting in itself, 

 and demands explanation. Such an explanation was 

 offered in the discussion, on the basis of the 

 "amorphous cement" hypothesis of Dr. Rosenhain. 

 If the crystals are held together by thin layers of 

 amorphous metal, then these layers would naturally 

 serve as channels for the diffusion of gases. Liquid 

 metals are well known to possess greater powers of 

 dissolving gases than the solid material, and the 

 " amorphous " condition is at all events closely akin 

 to_ the liquid phase, even if it is not entirely identical 

 with it. Even iron carbide would be soluble in the 

 amorphous layers, and its distribution along the 

 crystal boundaries would thus be readily explained. 

 On the cleavages also, traces of amorphous metal 

 might well be left as residues of the amorphous metal 

 which would — according to the view^s of Beilby and 

 Rosenhain — be formed on those planes whenever the 

 metal was "wrought." It will be seen that these 

 results of experiments on the action of gases throw 

 an unexpected sidelight on a theory which Is still the 

 subject of controversy; they even suggest the possi- 

 bility of employing gases as " reagents " for the detec- 

 tion and location of any amorphous metal which may 

 be present in a specimen. 



Finally, it should perhaps be stated — although it 

 is obvious enough — that the results contained in 

 these papers are of some considerable practical im- 

 portance, but in this place they have been regarded 

 rather from the point of view of scientific interest. 



SCIENCE AND CIVILISATION.^ 



'T^HE San Francisco meeting has been appointed 



^ with the double purpose of encouraging the 



development of science in the Pacific region and of 



uniting with other organisations in celebrating the 



completion of the Panama Canal. There could 



scarcely be a better illustration of the relations of 



science to civilisation than the canal supplies. This 



great waterway has been constructed, not so much by 



the potency of our national wealth in gold, not so 



much by the wonderful engineering and administra- 



j five ability which we all delight to honour, as by the 



! victory of pure and applied science over the sources 



' of malarial and yellow fever infection. Three centuries 



j 1 Address (slightly «brideed) to the American A«sociaii n for the Advance- 

 ment of Science, delivered at the San Francisco meeting, August 2, by the 

 • President, Dr. W. W. Campbell. 



