Feb. II, 1886] 



NA TURE 



351 



ilie solid body. Contact action seems also to have played its 

 part in the researches of M. Lemoine on the dissociation of 

 hydrogen iodide. On the whole, in all those cases where the 

 process of chemical transformation in a gaseous medium offers an 

 uninterrupted character, there is reason to suppose that a 

 contact action has been taking place. But if this supposition 

 proved to be correct, we should be compelled to admit that the 

 chemical transformation, even in its simplest shape in a gaseous 

 medium, is intimately connected with the action of molecular 

 forces — that is, with such actions which do not have the 

 characters of determinated chemical combinations. Molecular 

 forces ought to be taken into account even in the transforma- 

 tions going on in a gaseous medium ; both factors — tbe chemical 

 affinity and the cohesion — appear so intimately connected that it 

 would be impossible to delimitate them : the chemical reaction 

 would appear as a result of both the forces which unite atoms 

 in molecules and those which are at work between the 

 molecules. 



The last issue of the Journal of the Russian Chemical 

 Society (xvii. 7) contains the first part of a most valuable 

 inquiry, by M. Konovaloff, into the part played by contact 

 actions in dissociation. Without undertaking to deal with this 

 immense subject in full, the author, taking advantage of obser- 

 vations he had made together with Prof. Menschutkin during 

 their experiments as to the dissociation of compound ethers, has 

 submitted to a closer investigation the contact phenomena when 

 gases are brought into contact with solids. The want of cohesion 

 between thegaseous molecules, and thegreatdifference of densities 

 of both the gas and the solid, give better conditions under which 

 to study the influence of the solid. Summing up the researches of 

 Sainle-Claire Deville, Wintz, Faraday, Kamsay, Berthelot, and 

 many others who have devoted attention to the subject, the 

 author shows that capillary structure and porosity are not neces- 

 sary conditions in a solid body for producing dissociation : 

 smooth surfaces may also condense vapours and gases, and 

 sometimes retain them with such a force as to make the disen- 

 gagement of the absorbed gas quite correspond to the dissocia- 

 tion of a chemical compound. The character of the surface, 

 having of course a great importance, M. Konovaloff has carried 

 on his experiments so as to study the influence of the character 

 of the surface. The first part of his inquiry contains the experi- 

 ments made as to the dissociation of the tertiary amylacetate, the 

 method of inquiiy being successive determinations of the density 

 of its vapours on W. Meyer's method. The result arrived at is 

 obviously that the structure of the surface of the glass which is 

 brought into contact with amylacetate vapours is of great im- 

 portance ; but it is worthy of notice that the rough surface of 

 the glass-powder condenses the vapour without producing a 

 notable dissociation, while the smootli surface cf the glass- 

 cotton dissociates it. 



THE INSTITUTION OF MECHANICAL 

 ENGINEERS 

 'I'^HIS Institution held its annual meeting at the theatre of 

 ■*■ the Institution of Civil Engineers on Thursday and Friday 

 last under the presidency of Mr. Jeremiah Head, who was 

 re-elected for the ensuing year. 



A paper was read by Mr. J. H. VVicksteed descriptive of an 

 autographic test-recording .apparatus of a very ingenious cha- 

 racter. It is designed to obviate both the labour of observation 

 and that of hand-plotting. But, beside the saving of time and 

 labour, there is the further gain, in obtaining the diagrams auto- 

 graphically, that the progress of the test is continuous ; and as 

 time is a factor in the behaviour of a test-piece, it is important 

 in making tests for comparison that there should be no irregu- 

 larity in this factor. 



The sample is held between an upper and lower gripping-box. 

 The upper box is suspended from the back centre of a steelyard, 

 which, by the adjustment of its poise-weight, weighs whatever 

 pull is put upon the sample. The lower box is connected with 

 a hydraulic cylinder, which puts the pull upon the sample, and 

 extends it until it breaks. Thus while the hydraulic cylinder is 

 doing the mechanical work of breaking the sample, the steelyard 

 is measuring the load it sustains. The object of the indicator is 

 to record simultaneously the amount of the load and the exten- 

 sion due to it. To get this simultaneous record the horizontal 

 ram of the indicator, which carries the tracing pencil, is in fluid 

 connection with the hydraulic cylinder which puts the load upon 



the sample, and the indicator therefore partakes of that load. 

 Round the outer end of the ram is coiled a spiral spring, which 

 is compressed as the pressure on the ram increases, and expands 

 as the fluid pressure on the ram decreases ; the pencil records 

 the point of equilibrium between the two. The friction of the 

 leathers in the hydraulic cylinder and that of the indicator ram 

 are both eliminated from the diagram, the first by putting on to 

 the piston of the hydraulic cylinder a gross pressure equal to the 

 effective pressure on the sample and the friction of the hydraulic 

 leathers, and the second by revolving the indicator ram by belt 

 power and gearing ; the driving power being applied in a plane at 

 right angles to the longitudinal travel of the ram has no effect upon 

 that travel, but entirely overcomes the obstruction which the friction 

 of the leather would otherwise offer to the free travel of the ram, so 

 that the ram becomes sensitive enough to respond to the very 

 smallest want of balance between the opposite forces of the water 

 pressure and the spring. For recording the extension of the 

 sample simultaneously with the load upon it, the metallic paper 

 on which the pencil travels is mounted on a brass barrel like 

 that of an ordinary steam indicator ; and in accordance with the 

 extension of the sample the barrel is made to revolve by means 

 of an arrangement which eliminates any general movement of 

 the sample, recording that only which is due to its extension. 



The author summarises the autobiography of every specimen 

 strained to the breaking-point in the testing machine. Entering 

 the machine in a state of dnternal equilibrivmi, its first stage is 

 what is called in the paper one of unyielding elasticity ; it 

 extends about 1/10,000 of its length per ton of load, but on 

 removal of the load remains unstrained. In its second stage the 

 strains and stresses fluctuate, the bar yielding about 2 per cent, 

 of its length, the strain being beyond recovery. The pencil of 

 the indicator hesitates and almost trembles. There would seem 

 to be a succession of local extensions in the bar, as was lately 

 pointed out by Prof. Kennedy in this journal (Nature, vol. xxxi. 

 p. 504). These local extensions reduce the area locally in a 

 higher ratio than the cohesive force increases ; fracture would at 

 once occur were it not that after a short critical interval the bar 

 sets up increased resistance, thus entering its third stage. Stable 

 equilibrium is restored, but the permanent strain increases in its 

 ratio with every additional ton, and the bar may stretch 20 per 

 cent. During the last stage the equilibrium is again unstable ; 

 the pencil steadily records a rapidly-decreasing resistance, accorn- 

 panied by a local strain which, over the part where it occurs, is 

 very much greater than in any preceding stage. The author 

 concludes by drawing attention to the circumstance that the 

 apparatus records definitely the elastic limit of the material, the 

 diagram traced gives the gross mechanical work put upon the 

 sample, as it enables the local extension about the breaking- 

 point to be separated from the general, thus affording a means of 

 comparing samples of different shapes ; and lastly the apparatus 

 makes its record quite independently of the manipulation of the 

 poise upon the steelyard. 



A paper descriptive of tensile tests of iron and steel bars was 

 read, prepared by the late Mr. Peter D. Bennett. His principal 

 object in making these tests was to ascertain the relative effect 

 produced on the tensile strength of a flat bar of iron or mild 

 steel : (i) by a hole drilled out of the bar to the required size ; 

 (2) by a hole punched \ inch smaller in diameter, and then 

 drilled out to the size of the first hole ; and (3) by a hole 

 punched in the bar to the size of the drilled hole. In each of the 

 former cases the average strength was increased per square inch 

 of the original area across the fracture ; in the third case there 

 was a falling off in strength of nearly 20 per cent, owing to the 

 method of perforation. The results in the first two cases were 

 alike both for iron and mild steel, but in the third case the 

 diminution in strength of mild steel was only 6 per cent. In 

 another series of tests the perforated hole was filled with a rivet 

 put in by a hydraulic machine with a pressure of thirty-one tons 

 on the head, the results being relatively as before. The author 

 considers these results to be due to the fact that in the drilled 

 bar the slightly greater strain indicated was reached only along 

 the transverse diameter of the hole, and that the strain on the 

 metal decreased along the longitudinal diameter of the hole 

 until it was distributed over the whole width of the bar. Thus, 

 at the point where it was most severely .strained the metal would 

 receive some support from the less severely strained parts 

 adjoining. 



The tests go to prove that the elongation of different test- 

 bars, all of the same length, is greatly aflected by their diameter, 

 those of larger diameter elongating more than those of smaller 



