Nov. 7, 1889] 



NATURE 



13 



which they could lose by burning and regain by the pro- 

 cess they called " revivification." " Hardness [in metals] 

 is caused by the jeiunenese of the spirit and their imparity 

 with the tangible parts," said Francis Bacon ; ' while, 

 according to Stahl,'- steel was merely iron possessing, in 

 virtue of its phlogiston, the characteristics of a metal in a 

 higher degree ; and this view prevails in the writings 

 of Henckel, Newmann, Cramer, Gellert, Rinman, and 

 Macquer. This opinion survived with wonderful per- 

 sistence, but it did not influence the teaching of Rdau- 

 niur,"* who, in 1722, was, so far as I know, the first to 

 suggest a physical theory which has been in any way 

 justified by modern research. He assumed that when 

 steel was heated "sulphurs and salts" were driven out 

 from the molecules, which he represents diagram- 

 matically, into the interstitial space between them. The 

 quenching of the steel and its sudden cooling prevented 

 the sulphurs and salts from returning into the molecules, 

 which were thus firmly cemented by the matter between 

 them, and hard rigid steel was the result. In tempering, 

 the sulphurs and salts partially returned into the mole- 

 cules, and the metal became proportionately soft. I have 

 elsewhere shown ^ that he used the Torricellian vacuum to 

 demonstrate that the hardening of steel is not accom- 

 panied by the evolution of gas, and he concluded that 

 " since the hardening of steel is neither due to the 

 intervention of a new substance nor to the expulsion of 

 air, it only remains to seek its cause in the changes 

 occurring in its structure." Notwithstanding this, the 

 phlogistic school were not daunted, and this brings me 

 to the work of Torbern Bergman, the great Professor at 

 the University of Upsala, who in 1781 showed'' that steel 

 mainly differs from iron by containing about ^-^ per cent, 

 of plumbago, while iron does not. Read in connec- 

 tion with modern research, his work seems wonderfully 

 advanced. He was so forcibly impressed by the fact that 

 the great difference in the mechanical properties of different 

 specimens of iron is due to the presence of small quantities 

 of impurity, and that the properties of iron do not vary, as 

 he says, unless by chance the iron has gathered foreign 

 matter, " tiist forte peregrimein paiillo iibcrius itihcerat 

 metallian." We find, even, the dawn of the view that under 

 the influence of small quantities of foreign matter iron is, 

 as he calls it, polymorphous, and plays the part of many 

 metals. " Adeo ui Jure did qiieat, polymorphuDi ferrum 

 pluriidu simtil inctalloruni vices sustinere" ^ Unfortun- 

 ately he confounded the plumbago or carbon he had 

 isolated with phlogiston, as did Rinman in 1782, which 

 was strange, because, in 1774, the latter physicist had 

 shown that a drop of nitric acid simply whitens wrought 

 iron, but leaves a black stain on steel. Bergman tenaciously 

 held to the phlogistic theory in relation to steel ; it was 

 inevitable that he should. The true nature of oxidation 

 had been explained ; no wonder that the defenders of the 

 phlogistic theory shotild seek to support their case by 

 appealing to the subtle and obscure changes produced in 

 iron by such apparently slight causes. Bergman's view 

 was, however, combated by Vandermonde, Berthollet, and 

 Monge,^ who showed in a report communicated to the 

 Acaddmie des Sciences, in 1786, that the difference be- 

 tween the main varieties of iron is determined by varia- 

 tion in the amount of carbon, and further that steel must 

 contain a certain quantity of carbon in order that it might 

 possess definite qualities. Bergman died in 1784, and the 

 report to which 1 have referred is full of respect for " this 



' " Sylva Sylvarum," 2ncl edition, 1628, p. 215. 



2 ■' Fundamenta Chemise," Part 3, p. 451, quoted by Guyton de Morveau 

 in the article "Acier,"'" Encyo. Mdthodique," p. 421 (Paris, 1786). 



3 " L'art de convertir le fer forge en acier," p. 321 et seq. (Paris, 1722). 



4 Proc. Inst. Mech. Engineers, October i88i, p. 706. 



5 " Opuscula Physica et Chemica," vol. iii. " De Analyst Ferri " (Upsala, 

 1783). A dissertation delivered June 9, 1781. 



6 -'De Analysi Ferri," p. 4. 

 ^ " Histoire de I'Academie R05 



'32- 



•jyale des Sciences," 1786 (printed 1788), p. 



grand chemist," as its authors call him, "whom science 

 had lost too soon." 



Kirwan's essay on phlogiston,' in which Bergman's 

 views were defended, elicited a reply from Lavoisier him- 

 self, and brought down the French school in strength to 

 contest almost the last position occupied by the believers 

 in phlogiston."-* 



An entire lecture might be profitably devoted to 

 Bergman's work. His was almost the first calorimetric re- 

 search, and is specially interesting when taken in connec- 

 tion with the calorimetric investigations of Lavoisier and 

 Laplace in 1780, and it is impossible to read it without 

 feeling that in paying the just tribute to Lavoisier's genius 

 Bergman has been overlooked. He desired to ascertain 

 whether pure iron, steel, and cast iron contain the same 

 amount of heat. He therefore attacked the materials with 

 a solvent, and noted the heat evolved. He says the 

 solvent breaks up the assemblage of the aggregation of 

 molecules and forms other unions. If the new body 

 demands more heat than the body which has been dis- 

 united, then the thermometer will fall. If, on the other 

 hand, the degree of heat required is less, the environ- 

 ment will be heated, which will result in the rise of the 

 thermometer. The modern development is that, when a 

 chemical compound is formed, heat is evolved and energy 

 is lost, but if one substance, say a metal, simply dissolves 

 another, the solution is attended with absorption of heat, 

 and the product when attacked by a suitable solvent 

 should evolve practically the same amount of heat, but 

 certainly not less than would be evolved by the individual 

 metals present in solution.'' This is specially interesting 

 from its relation to the calorimetric work of Lavoisier 

 and Laplace in 1780 and of Lavoisier in 1782, which led 

 the latter to explain the nature of oxidation, and to show 

 that a metal could be as truly "calcined" or oxidized 

 by the action of a solution as by the action of air at an 

 elevated temperature. Now that the importance of thermo- 

 chemistry is beginning to be recognized in relation to 

 industrial chemistry and metallurgy, it is to be hoped that 

 Bergman's merits will be more fully considered. We are, 

 however, mainly concerned with the fact that he taught 

 us that the difference between iron and steel consists in 

 the ,-77 to \\, per cent, of carbon which steel contains. It 

 was only natural that Black, writing in 1796, should have 

 attributed the hardening of steel to the "extrication of 

 latent heat " ; " the abatement of the hardness by the 

 temper" being due, he says, "to the restoration of a part 

 of that heat."' Black failed to see that the work of 

 Bergman had entirely changed the situation. The next 

 step was made in France. It was considered necessary 

 to establish the fact that carbon is really the element 

 which gives steel ii;s characteristic properties, and with 

 this object in view, Clouet,-^ in 1798, melted a little 

 crucible of iron, weighing 57 '8 grammes, containing a 

 diamond, weighing 0907 gramme, and obtained a fused 

 mass of steel (Fig. i). 



His experiment was repeated by many observers, but 

 the results were open to doubt from the fact that furnace 

 gases could always obtain access to the iron, and might, 

 as well as the diamond, have yielded carbon to the metal. 



' R. Kirwan, " Essay on Phlogiston and the Constitution of Acids," 

 p. 134(1787). 



^ " Essni surle Phlogistique," traduit de TAnglois de M. Kirwan, avec des 

 notes de MM. de Morveati, Lavois.er, de la Place, Monge, Berthollet, et de 

 Fourcroy (Paris, 1788). 



3 See French translation of Bergman's work (Paris, 1783), p. 72. The 

 question is, however, so imp-^rtant that I append the original Latin text : — 

 " Menstruo laxatur compages molecul.irum, et nova f ^rmantur c ;r.nubi.i, 

 quae, si majorem, quam diruta, figimt materia- caloris quantitatem, in vicinia 

 calor ad rcstituendum sequilibriura diminuatur oportet, et thcrrnometri 

 hydrargyrum ideo subsidet : si minorem, differentia liberatur et vicinlam 

 calefacit, undeetiam adscendit thermometri liquor; si denique mva conniibia 

 eamdem pra;cise quantitatem rostulait, (luod raro acc.dit, nulla in thermo- 

 metro videbiiur variatio." — Torberni Bergman, _" Opuscula Physica et 

 Chemica," vol. iii. p. 58, 1783 (" Oe Analysi p'erri ''). 



■4 •' Lectures on the Elements of Chemistry," vol. ii p. 505 (1803). 



S Experiment de.scribed by Guyt;n de Mcrveau, Ann. de Chim., xxxi. 

 '799. P- 328. 



