long period of time that it was entirely negligible compared to the errors in weighing. A series of experiments were, therefore, carried out to determine the accuracy of Babington's observations. Experimental. At the outset we were convinced that there was nothing to be gained by working with a large volume of solution and weighing with approximate accuracy. Accordingly 2-inch crystallising dishes were used as containers for 10 cc, portions of the several solutions which were weighed on a sensitive balance. To prevent evaporation of the solutions during the weighing periods, the edge of each container was carefully ground and fitted with a ground glass plate. Before carrying out the earlier experiments the inside of the dishes was given a thin coat of paraffin to guard against "creeping" of the solutions; this was later discontinued since it was found that, in general, the rate of evaporation was more uniform from dishes that were not paraffined. The first evaporations were carried out on a stationary platform in the centre of a room ten feet square. The dishes were set the same distance apart on the circumference of a circle of 10-inch radius. It was hoped to secure uniform exposure in this way; but the results were not entirely satisfactory and a rotating platform was later employed as subsequently described. The first experiments were carried out with salt solutions containing 10 grms. of salt per 100 cubic centimetres which was the concentration used by Babington. Ten cubic centimetres of each were placed in dishes which were arranged on stationary platform as above described. The loss on evaporation was determined in a series of experiments, the results of which are given in Table I. a Evaporation from a Stationary Platform. Loss on evaporation in grms. II III IV V 20 hrs. 24 hrs. 16 hrs. 18 hrs. 12 hrs. 1569 1269 0446 0·658 0.973 1586 1278 0'419 0653 0.963 1609 1311 0414 0684 0043 Copper sulphate 1540 1235 0411 0620 Pot. ferrocyanide 1539 1216 0.391 0'597 Pot, bicarbonate I'495 1202 0380 0.624 Pot. nitrate 1'441 1'441 1'132 1'057 Pot. tartrate ... Pot. carbonate Pot. sulphate ... 1390 1135 I'022 0.563 O'555 0.488 0.568 0'509 Sodium carbonate 1327 From the above table it will be seen that, contrary to Babington's observations, all solutions evaporated more slowly than water. However, the results are not entirely satisfactory as different samples of the pure solvent showed considerable variation in the rate of evaporation. Moreover, if the 10 per cent salt solutions are arranged in the order of evaporation rate, this order is not always the same. Such a variation can be due only to experimental errors. The most likely source of error seemed to be the failure to get uniform conditions of exposure. To obviate this it was decided to carry out the evaporations on a rotating disk that was 16 inches in diameter. This disk was turned once every half minute by 10 per cent Copper sulphate 0'962 0'903 10 per cent Copper sulphate 10 per cent Copper sulphate The above results show two things; first, that 10 per cent copper sulphate evaporates more slowly than water instead of faster, as Babington found; and second, that a clean dish makes a better container than a paraffined one. The use of the latter is most likely to give erroneous results with saturated solutions, which tend to creep up the sides and deposit crystals if the dish is not thoroughly clean. This effect is shown in a striking way in the experiments recorded in Table IV. 1 results are not comparable. It is altogether probable that Babington's failure to take this into account resulted in the erroneous conclusion that certain saturated solutions evaporate as fast as, or faster, than the pure solvent. From the above considerations it follows that if the satruated solutions do not creep during the process of evaporation, their rate of evaporation would be much smaller than the pure solvent. This is well illustrated in the experiments recorded in Table V. In this experiment particular care was taken to clean the containers and there was no creeping, so that all crystals deposited on the bottom of the dish instead of on the sides. Under these conditions the solutions evaporated much slower than water as the results show. It is interesting to note that the solutions used were the same as Babington found to evaporate more rapidly than water. Sample TABLE V. Ostwald (Zeit. phys. Chem., 1888, ii., 78) has shown that when CrO, is dissolved in water there is formed dichromic acid, H,Cr,O,, instead of chromic acid, H,CrO,, as might be expected from analogy with the action of SO, and water. Το verify this, two solutions of CrO, in water were prepared the first contained 1 mole of CrO, per litre; and the second one-half mole of Cro, per litre. The rate of evaporation of these two solutions was compared with the rate of evaporation of M/2 sulphuric acid. The time of exposure was 14 hours. The results are given in Table VII. : I Water Saturated potassium chlorate Saturated copper sulphate Saturated potassium ferrocyanide The A fairly accurate comparison of the relative lowering of the vapour pressure of water by different salts is obtained by comparing the loss on evaporation of solutions of suitable concentration. In the following experiment half-molar solutions of several salts were employed. results are given in Table VI. In the first column the salts are arranged in the order in which they decrease the rate of evaporation, in the second is recorded the loss in weight after 14 hours' exposure; in the third, the approximate degree of ionisation, a determined from conductivity; and in the fourth, the approximate value of (the ratio of the number of particles after dissociation to the number of particles before dissociation) calculated from the expression i=1+(f—1) where is the number of parts into which a molecule dissociates. TABLE VI. a 0.696 23.8 124 0.679 7413 174 0.075 79'9 1.80 0.665 40'0 1.80 0'633 60'0 2.20 M/2 Copper sulphate As the above results show the loss by evaporation of equimolecular solutions is least with salts that give the greatest number of ions, which is exactly what one might expect. The conformity of the results to the theory suggests the possibility of using the rate of evaporation as a means of determining the molecular weight of certain From the above results it is evident that molar Cro, which corresponds to M/2 H2Cr2O, evaporates at almost the same rate as M/2 H2SO.; while half-molar CrO,, which would correspond to M/2 H2CrO4, evaporates considerably faster than M/2 H2SO, showing that it is not M/2 H2CrO, but M/4 H,Cr2O,. This confirms Ostwald's conclusion that CrO, reacts with water to form chiefly dichromic acid rather than chromic acid. The fact that M/2 sulphuric acid evaporates somewhat faster than the same concentration of dichromic acid is due to the fact that the degree of ionisation of the latter is greater than the former in M/2 solutions (Walden, Zeit. phys. Chem., 1888, ii., 71). From the above series of experiments it is proven that Babington's conclusion is erroneous, namely, that certain salt solutions evaporate more rapidly than water under constant conditions of exposure at ordinary temperature and pressure. As has been shown, the chief source of error in Babington's work probably is the failure to maintain constant conditions of exposure. The factor which is most difficult to maintain constant is the extent of surface exposed, particularly if the containers are not uniformly clean. This source of error is magnified with saturated solutions which will creep badly if the sides of the container are not thoroughly clean. By carrying out the evaporations with clean glass containers on a rotating platform results can be duplicated consistently to within 1 per cent. All the results obtained under these conditions are readily explained in terms of the modern theory of solutions. Journal of Physical Chemistry, May, 1920. CORRESPONDENCE. VALENCY AND IONISATION. To the Editor of the Chemical News. SIR,-In the CHEMICAL NEWS of October 11, 1918, cxvii., p. 319, an article of mine appeared on Valency. Recently I have seen a criticism directed against the idea, of what might be termed a "crystal molecule." So far as I am able to judge, the ideas advanced in my article are in agreement with certain parts of Langmuir's atomic theory; and Prof. Bragg in Nature, July 22, 1920, speaks of the diamond as a vast molecule. The idea that atoms become in a sense ionised in solids, crystals in particular-because the evidence by means of X-rays gives direct support to the idea is a very beautiful one, and it is one of the salient points in Langmuir's theory. This may remind the readers of the CHEMICAL NEWS of my definition of cohesion (and valency) on p. 320 (loc. cit.). My object in writing is to point out that the ideas I have brought together are worthy of fresh perusal in the light of Langmuir's 1919-20 theory (I say this, as I am not claiming anything as my own), since every idea, as small as it may seem in itself, is a step in advance if it leads in the right direction. See also, Langmuir, CHEMICAL NEWS, July 16, 1920, p. 29.—I am, &c., NOTES. F. H. LORING. IRON AND STEEL INSTITUTE.-Autumn Meeting, 1920. The Autumn Meeting will be held in Cardiff, at the South Wales Institute of Engineers, on September 21 to 24. The following is a list of the papers which it is expected will be submitted at the meeting : E. D. Campbell and B. A. Soule: "Notes as to R. H. Greaves and J. J. A. Jones: "Temper K. Honda and T. Murakami: "Graphitisation in K. Honda and S. Saito: "On the Formation of Spheroidal Cementite." A. Hutchinson and E. Bury: "The Rough Cleaning of Blast-Furnace Gas at Skinningrove by the Lodge Electrostatic Process." A. McWilliam: "Indian Iron Making at Mirjati, Chota Nagpur." W. Rosenhain and D. Hanson: line Fracture in Mild Steel." "Inter-Crystal J. H. Whiteley: "Experiments on the Deoxidation of Steel with Hydrogen." Arrangements have been made for excursions to several works, &c., in the district. Further particulars can be obtained from Mr. G. C. LLOYD, Secretary, 28, Victoria Street, S. W. 1. CITY AND GUILDS OF LONDON INSTITUTE.-The Council have awarded diplomas of "Associate" to the following students who completed a full course of instruction at the close of the recent session :-Allen, R. H., Ascanio, W. de, Ayerst, W. G., Baillie, R., Baines, T. N., Bedford, L.H., Beresford, D., Bishop, H., Budgett, F. la T., Blair, T. S., Bowden, W. Á., Buckell, J. H. W., Burkitt, S. T., Butler, F. W., Carter, R. H., Chapman, A. R., Chilton, O. H., Clifford, S., Coombes, L. P., Cousinery, E. de, Cowie, G. C., Cramer, W., Cronin, H. F., Cushny, J., Davis, G. W., Davis, J. S., Day, A., Dean, A. J., Dovaston, G. E., Dovaston, W. A., Downes, E., Edmunds, F. T. (Henrici Medal), Falkner, V. M., Fenton, J. M., Firnberg, L. B., Fraser, R., Gabel, R., Glasspoole, G. H., Goodman, R. T. W., Gordon, L. V., Gowring, H. J., Green, E. R., Grose, J. W., Hall, T. F. B., Harris, H. A., Hasselt, M. Van, Haward, F. B., Heading, F. R., Highett, R. F., Hills, B. A. C., Hinton, L. T., Hodge, H. W., Holley, E. G. (Siemens Medal), Holstrom, J. E., Hurley, T. F., James, R. T., Johns, J. P., Jones, E. M. R., Jones, E. R., King, G. B., Lee, W. Y., Lobjoit, W. H., Macculloch, A., Macgregor, J. C., Macqueen, E. N., Martin, G. W., Maughan, J. D., May, A. E., McLeod, R. C., McVie, A., Mooney, A. McL., Morphy, D. W., Moul, F. D., Nachshen, M., Naish, H. F., Naylor, E. J., Paget, A. L., Pearlman, A., Peters, H., Pomary y Tenaud, P. L., Ringang, R. D., Robins, B. G., Rowell, K., Ryves, F. D. M., Shipman, E. M., Sich, W. E. E., Simonson, W. F. (Bramwell Medal), Stavridi, A., Stern, M., Stevens, S. G., Sturt, C. G., Swann, A. H., Townshend, B. W. O., Turner, E. H., Tutleman, B., Vereker, H. C., Webster, E. H., Whyte, D., Wilson-Jones, R. A., Wood, L. J. ASSOCIATED SCIENTIFIC AND TECHNICAL SOCIETIES OF SOUTH AFRICA.-A movement which has been proceeding actively for some time in Johannesburg, writes H.M. Senior Trade Commissioner in South Africa recently, culminated in the formation of a body to be known as the Associated Scientific and Technical Societies of South Africa. The new organisation will bring together eight or ten of the chief technical societies of South Africa into a common institution. The necessary funds are being obtained by a loan from the Chamber of Mines, by contributions from the capital funds. of each society, and by an appeal to each of their members. TREATMENT OF OVERSEAS AGENTS IN THE DOMINIONS.-During the difficult times that accompanied and have followed the war, United Kingdom manufacturers and merchants have no doubt received many complaints from their customers and agents all over the Empire. Many of these complaints are due to causes over which firms here have no control, but one that is frequently made might, it is felt, to some extent be remedied, and that is, failure to notify agents and customers of the state of affairs in the industry at home, of ability to supply, of failure to execute orders, and of the reasons for such failure. Statements have been received from a number of agents, either direct or through His Majesty's Trade Commissioners, to the effect that not only do their principals not volunteer information as to the reasons for high prices and late deliveries, but that when asked for reasons which the agents can advance to their customers, their principals etiher ignore the request or reply in a brief and entirely unsatisfactory letter. An instance may be cited of a firm in one of the Dominions who took a great deal of trouble to secure a municipal contract and were assured that, if the goods could be delivered, the contract would be given to them. They cabled and wrote, but received no reply from their principals, who thereby jeopardised not only their own position with respect to further orders from that municipality, but also that of all the other firms represented by that agent. On the other hand, goods sometimes arrive almost before the 'agents know that they are coming at all. Overseas agents realise a good deal of the difficulties which exist at home and are prepared to take them into account. His Majesty's Trade Commissioners also lose no opportunity of meeting criticisms in this way, but it is felt that this plea does not altogether meet the case, and that irritation is sometimes caused which could be avoided, and which must be detrimental to the strong desire to purchase British goods which undoubtedly exists throughout the Empire. It is pointed out that manufacturers should remember that it is their agents, and not they, who have to live alongside and endeavour to do business with disappointed customers, and that failure to keep the agent properly informed may lead to the latter going to useless trouble and expense in obtaining orders which cannot be executed. A little more information from home and a little clearer indication of the possibilities of doing business would go a long way towards smoothing matters, and would tend to remove the impression that in more important things also, overseas agents and customers are not receiving due consideration. OWING NOTICE. WING to the greatly increased cost of printing and paper, and the advance in the postal rates, we have been compelled to raise the price of the CHEMICAL LADY CHEMIST, B.Sc., with four years experience of Iron and Steel analysis requires post-Box 810, c/o Scorт & Son, 63, Ludgate Hill, London, E.C.4. £0 16 0 £0 8 0 80 SUPER-GROUND Silica Flour, of all grades, from 10 mesh to 200 mesh; also "extra fine" and "superfine" (air-flown) Flour Silica, of high quality, perfectly milled, of even texture, and correctly graded. Speciality-120-mesh grade, guaranteed to contain not more than 0.5% residue when passed through Woolley's interwoven silk 120-mesh. General grinding undertaken for the trade.-THE STAR QUARRIes, Ltd., Experts in Fine Grinding, 5a, St. Werburgh Street, Chester. THE CHEMICAL NEWS. VOL. CXXI., No. 3149. EDITORIAL. THE following papers have been received for publication, and will be inserted as soon as space permits : "On Sodium Phosphide." By E. Tomkinson and G. Barker. A preliminary account of the reaction under various conditions, the product, and its reaction with water. Received August 9. ISOMERISM, TAUTOMERISM, AND PSEUDOMERISM OF ORGANIC COMPOUNDS AND THEIR RELATION. By INGO W. D. HACKH. THE purpose of this paper is the formulation of a system and classification of isomeric organic compounds which enables a clear cut distinct definition of the terms isomerism, tautomerism, metamerism, desmotropism, and pseudomerism. All these phenomena are caused by differences in the structure of the organic compound; they belong therefore to structural phenomena and their classification must be based upon their structural differences. While for practical reasons it may be advantageous to base a division of these cases upon dynamical grounds, that is upon the ability to transform one form into the other, it is nevertheless a question of structure. There are six factors to consider, two of which (K and N) are self-evident and included to show the connection with polymerism, homologous and isologous series. The relation among these six factors and the corresponding category are shown in the following key to isomeric phenomena in which N is the number of atoms in the organic molecule; K the kind of atom in the organic molecule; P the position or arrangement of atoms; T the chemical type of compound; V the valency of the principal atom, L the linkage or connections between the atoms. The first group of cases-isology, homology, and polymerism-does not come properly under the heading of isomerism, but is related to it. The second group has as its characteristic the difference in the position of the atoms and is thus isomerism proper, with the subdivisions of structure isomerism, cis-trans isomerism, and stereo isomerism. The third group has as earmark a difference in the type of compound and is known as tautomerism in general, but may be divided into metamerism, desmotropism, and tautomerism. The fourth group is marked by a difference in the valency of the principal atom and is designated pseudomerism. Isomerism needs no explanation or definition. It includes all the cases described as (a) chain isomerism (e.g., propane and isopropane); (b) place isomerism (e.g., 1-chloropropane and 2-chloropropane); (c) meta-isomerism (e.g., 1-butylene and 2-butylene or 3-pentanone, 2pentanone, and 1-pentanone) which should not be mistaken for metamerism; (d) ring isomerism (the o-, m-, and p-, the v-, s-, a- positions); (e) sidechain isomerism (e.g., propylbenzene and isopropylbenzene); and (f) mixed isomerism (e.g., a-chlorotoluene and 2- or O- chlorotoluene). All these cases can be properly termed structure isomerism, for they are based a different two-dimensional arrangement of the atoms in the organic molecule. In the second series of isomerism, sometimes called geometrical or cis-trans isomerism, there is a double bond either between (a) two carbon atoms, (b) two nitrogen atoms, or (c) a C and N atom. To this group belong all cis-trans forms of aliphatic and aromatic compounds as well as the syn- and antiforms. In the third and last series of isomerism radicals show on 1 S S d d S d d same or different. Pseudomerism when S where s same, d=different, and |