16

THE experiments herein described include the sulphate and chloride baths as these have been used by most investigators. Other electrolytes have been tried without much success, such as ferrous fluorsilicate, but it was considered best to restrict the experiments to the more promising methods and, by comparing them under certain conditions, to decide which would be the most practical for the deposition of metallic iron.

of

The first solution tried was made by dissolving 200 grms. FeSO4-(NH4)2SO4.6H2O in one litre. This gives about 29 grms. of iron per litre, or a concentration of 2.9 per cent of iron. This is the solution as used by Amberg (Zeit. Electrochem., 1908, xiv., 326), and probably by Burgess (Trans. Am. Electrochem. Soc., 1904, v., 201; Electrochem. and Met. Ind., 1904, ii., 183) and Hambuechen. The cathodes were clean sheet iron, and the anodes consisted of wrought iron bars. The experiments were carried on at room temperature and the results are given in Table I.

TABLE I.-Iron.

{° July 8, 1910

FeSO4.7H2O, 21 grms. Na2SO4.10H20 per 100 cc. as described by Kern (Trans. Am. Electrochem. Soc., 1908, xiii., 103).

A current density of 1 ampère per sq. dcm. was used. The deposit was dark grey and formed large nodules. At the end of fifteen hours the electrolyte was in bad condition owing to the formation of a heavy precipitate.

The net results of the experiments with different sulphate solutions were the conclusion that a solution containing 200 grms. per litre of Mohr's salt was the best as regards character of deposit, permanent good condition of electrolyte, and energy efficiency. Attention was next turned to the electrolysis of chloride solutions. The first solution used contained 100 grms. FeCl2.4 H2O and 50 grms. NH4Cl per litre. The cathodes, as before, being of clean sheet iron, and the anodes wrought iron bar. The electrodes were 1 ins. apart, and the bath maintained at room temperature. The current efficiency as determined by a copper coulombmeter was found to be 96.5 per cent, | and the following data was obtained :— Time of run 21 hours.

Average current during run = 0.237.
Average voltage during run=0'42.

Weight of iron actually deposited, 5'14 grms.
Weight of iron theoretically deposited, 5.32 grms.
Current density during run-o'304 ampères sq. dcm.
No. watt-hours = 0.237 × 21 × 42 = 2'14.

I watt-hour = 2'4 grms. iron.

I kw.-hour = 5.28 lbs.

The current was varied temporarily to obtain the fol

lowing readings.

0.260

0.66

6'04

2'42

2.42

5'55

0'44

8.26

I'98

5:05

The current efficiency was determined by inserting a copper coulombmeter in series, and gave an efficiency of 95.5 per cent.

The last column in this table shows the effect of increased current density on the energy efficiency. The current was changed temporarily to obtain these figures. In computing the figures for lbs. per kw.-hour, the current efficiency was assumed to be the same throughout.

The deposit obtained was silvery white and of a fine crystalline structure. After a while it became badly pitted. The electrolyte kept in condition for a long time; the basic precipitate which formed by oxidation of the air was allowed to collect in the bottom of the vessel. The electrodes were ins. apart, and the current density at the anode was about double that at the cathodes.

Another run was made with the solution and all conditions kept the same as the above except the current density, which was raised to I ampère per sq. dcm. The deposit was crystalline and silvery white like that obtained at lower current densities, but showed a greater tendency to form nodules. The current efficiency was nearly 100 per cent, but the voltage drop was 1.2, giving 1.705 lbs. per kw.-hour.

A third run was made with ferrous ammonium sulphate solution, with the addition of a small quantity of glycerine. The deposit in this case was smoother and more solid.

Ryss (Zeit. Elektrochem., 1906, xii., 697), in his investigations, used a solution containing 200 grms. ferrous ammonium sulphate, 50 grms. MgSO4, and 4 grms. NaHCO3 per litre. Ryss claimed that such a solution gave a smoother and thicker deposit of iron than could be obtained with ordinary ferrous ammonium sulphate.

A trial failed to show any improvement in the character of the deposits using such additions, but it did show that the bath had a much greater tendency to form basic precipitate, and the time in which it could be maintained in good condition was less than that possible with the simple bath.

The next solution tried was one containing 30 grms.

the energy efficiency at any given current density is much higher for the chloride solutions. Comparing the two chloride solutions, we see that the more concentrated solution gives a higher energy efficiency at any given current density.

The physical character of the deposit from the chloride solution was somewhat different from that deposited from the sulphate solutions, being more crystalline than the latter excepting that to which glycerine had been added. The next electrolyte tried was the ferrous sodium chloride solutions described by E. F. Kern (Trans. Am. Electrochem. Soc., 1908, xiii., 103). This was made by dissolving 285 grms. FeCl2.4H2O and 102 grms. NaCl per litre, giving 8 per cent Fe and 4 per cent Na. The solution may also be made by dissolving iron in the equivalent weight of HCI. The cathode used was thin sheet iron plate 3 ins. × 3 ins. The anodes were bars of wrought iron ins. x 2 ins. Xin. The electrolysis was carried on in a beaker placed on an asbestos pad with a hole cut in it, and heated by placing an incandescent lamp underneath. This kept the temperature between 50° and 70° C.

3

Previous runs having all shown a high current efficiency of over 95 per cent, it was not considered necessary to use a copper coulombmeter in this run. The object of the

run was to:

(1) Compare the voltage required by this process with the others described.

(2) To find the effect of heating the electrolyte on the energy required.

(3) To find the effect of heating on the physical character of the deposit.

(4) To secure a thick deposit in a short time.

The cathode area was 13.15 square inches.

The electrolyte was kept in condition by adding a little HCI to the bath every day to dissolve the basic precipitate which formed. Water was fed to the bath automatically to make up for evaporation.

The following readings were taken :

At the meeting of this Society on April 5th at Johannesburg, Dr. Moir demonstrated a new test for hydrocyanic acid of his invention. He said: "The principle of the test is the oxidising action produced when a cupric salt is added to prussic acid (owing to its reduction to cuprous cyanide). stance is added, and the most sensitive one I have found An organic compound which oxidises to a coloured subis hydrocorulignone. The test is sensitive to something like I part in 5,000,000 of water, but unfortunately the reagent is rather unstable; however, other oxidisable substances of the same class will answer although not so sensitive, and some of these, like benzidine, keep much better. The reactions in the case of the very sensitive reagent (copper acetate, acetic acid, and hydrocœrulignone)

are :

The above tables show that the voltage decreases with increased temperature; and that the voltage decreases as the run proceeds.

Comparison of this process with the others tried shows the best energy efficiency for the ferrous sodium chloride electrolyte working at 50-70° C.

This bath also gave the best and thickest deposits, the ir on being grey and non-crystalline. Journal of Industrial and Engineering Chemistry, ii., No. 6.

Aloïnose or Sugar of Aloïn.-E. Léger.-Aloïnose is formed by the action of alcohol on barbaloïn at 90°. It has a sweet taste, reduces Fehling's solution, gives the furfurol reaction with aniline acetate, and also G. Bertrand's reaction (violet coloration) with HCl and orcin. [ap]=-57.3° to. -58.5. From its reactions it appears o be a pentose.-Comptes Rendus, cl., No. 16.

"Damping of Sound by Frothy Liquids." By A. MALLOCK, F.R.S.

The object of the note is (1) to explain the well-known fact that a vessel which, when empty or filled with a homogeneous liquid, gives a musical note when struck, ceases to do so when the liquid contains bubbles of gas; (2) to call attention to the fundamental difference between the damping of waves propagated through a gas containing spheres of liquid (e.g., rain or fog), and that which occurs in a liquid containing bubbles of gas.

The damping of sound waves by fog has received considerable attention, and it has been shown that, although presence of liquid particles does cause a certain amount of dissipation, loss of energy is small, and this agrees with

observation.

On the other hand, with a liquid containing bubbles, damping of vibration is excessive, practically the whole of the wave energy being dissipated in a few wave-lengths or periods.

When spheres of liquids are disseminated in a gas, compressions and dilatations take place in the latter much as if only the gas were present, and the increased dissipation is due to slight modification in the motion of the gas, brought about by the liquid spheres,

It is pointed out that in a mixture of liquid and gas (in the form of bubbles) the velocity of wave propagation is less than that in either constituent alone, and has a minimum when the proportion of gas to liquid has a certain value. If the mixture consists of air and water, minimum velocity of propagation is reached when the volumes of air and water are nearly equal, and is then about one-fourteenth of the velocity of sound in air.

When the volume ratio of gas to liquid exceeds a certain limit, depending on the order in which the bubbles are arranged, the latter cannot remain spherical, and the mix. ture then becomes a froth, or collection of air cells separated by thin liquid walls.

It is shown by experiment that such froth is a very effective agent in damping vibrations.

"Dispersion of Light by Potassium Vapour." By Prof. P. V. BEVAN.

The work described in this paper was an attempt to measure quantitatively the amount of dispersion in the vapour of potassium. Dispersion takes place in the vapour chiefly on account of the red absorption lines, but also to a smaller amount due to other lines of the principal series. Measurements were made showing dispersion affected by the first six pairs of lines of the principal series, and an effect could just be detected at the next pair of lines. The dispersion was found to fit a Sellmeier formula, and values for four of the constants for this formula were obtained.

On theoretical grounds we can conclude, from the relative values of the constants of the dispersion formula that the number of atoms taking part in the absorption of the light after the first pair of lines must be only a small fraction of the total number present in the vapour, and that this fraction decreases with the number of lines in the series. It is suggested that the explanation of series lines must therefore be looked for in systems which are not the atom pure and simple, but probably atoms to which a corpuscle or more than one corpuscle become attached. Several types of quasi compounds may thus be formed in a way suggested by Sir J. J. Thomson, and the periods commonly associated with the atoms may be the periods of these systems. Our conception of the atom may be thus considerably simplified, as the number of degrees of freedom for each individual atom may be diminished very largely if

this view be the true one.

"Additional Refractive Indices of Quartz, Vitreous Silica, Calcite, and Fluorite." By J. W. GIFford. "Absorption Spectra of Sulphur Vapour at different Temperatures and Pressures, and their Relation to the Molecular Complexity of this Element." By J. IVON GRAHAM, Royal College of Science for Ireland.

The absorption spectra were photographed at temperatures varying from 530° C. up to 900° C., at atmospheric pressure, and at constant temperatures, under pressures between atmospheric and 10 mm. of mercury. The photographs at constant pressure with the above variation of temperature show the presence of two distinct absorption spectra; these are attributed to the intra-molecular vibrations of the S8 and S2 complexes respectively, the former producing a series of absorption bands between n(=λ-1) 2000 and n 2600, with mean position of maximum absorption about n 2500, whilst the relatively lighter S2 molecular system, by taking up vibrations of greater frequencies, produces a series of bands lying between 2900 and n 3820, with mean position of maximum absorption about n 3750.

Since only two distinct spectra are evident, it is concluded that the equation S8=4S2 represents the sole reaction that occurs in the dissociation of sulphur vapour on heating from its boiling-point up to 900° C. The inter

The position of maximum absorption of each band is towards the more refrangible edge, whilst the individual bands of each series appear to become stronger, also in the more refrangible direction. The two series of bands are shown mapped in oscillation frequencies, the similarity between the series being much more evident when illustrated in this manner.

Reproductions of photographs also accompany the paper.

66

Series of Model Experiments." By T. H. HAVELOCK, Wave-making Resistance of Ships: a Study of certain

M.A., D.Sc.

In a previous paper (Proc., A, 1909, lxxxii., 276) the author discussed the variation of the wave-making resistance of a ship with its speed, and a formula was obtained by specifying the action of a ship in terms of a simple equivalent pressure distribution travelling over the surface. The present investigation is a more systematic study of some of the coefficients of the formula, the experimental data being taken from published records of tank experiments with models. The discussion is limited to types for which the resistance-velocity curve shows clearly the humps and hollows which are usually ascribed to interference of wave systems originating at the bow and stern; the tabulated results form a numerical study of the latter theory, and exhibit the variation of the coefficients of the simple equivalent pressure system with the displacement of the model, the proportion of parallel middle body, and various coefficients of fineness. Without attempting to express the coefficients by empirical formulæ, sufficient information is available to allow of an approximate estimate of their values in similar models; this is illustrated by the Turbinia, and the result is discussed in relation to the published record of trials of that vessel. The characteristic interference effect mentioned above appears to occur specially in rather full-ended models, with fairly high cylindrical coefficients; in this case it is permissible to regard the equivalent pressure system as having two parts associated with the bow and the stern respectively. An examination of models with finer ends suggests that this simple interference theory is inadequate in certain cases; the study of a modified type of pressure distribution is indicated.

"Blood Volume of Mammals as Determined by Experi ments upon Rabbits, Guinea-pigs, and Mice; and its Relationship to the Body Weight and to the Surface Area expressed in a Formula." By GEORGES DREYER, M.A., M.D., and WILLIAM RAY, B.Sc. (Oxon.), M.B. (Adelaide).

"Autotoxæmia and Infection." By E. C. HORT, F.R.C.P.Ed.

The object of this communication is to show that fever, loss of weight, and changes in the antitryptic values of the blood serum, three phenomena common to bacterial and protozoan infection in man, can be reproduced in animals by the subcutaneous injection of small quantities of distilled water. Elaborate controls were set up throughout, and absence of sepsis repeatedly proved by autopsy and by microscopical sections.

Fever.-Sixty guinea-pigs received single injections of boiled distilled water in quantities varying from 1 to 10 cc. Fever resulted in fifty. Thirty guinea-pigs received multiple injections, always followed by fever except when the injections were too closely crowded, or too large, when sub-normal temperatures resulted. Twelve rabbits received single injections varying from 10 to 60 cc. Fever followed in all. Ten rabbits received multiple injections. All showed fever. Fever after each injection was always rapid in onset, abrupt and fugitive in both guinea-pig and

rabbit. By appropriate spacing of injections continuous fever can be produced, ceasing with the injections. Hypersensitisation was frequently observed. Establishment of a constant between weight of injection, animal injected, and degree of fever induced has been so far impossible.

Weight. The effect of small single injections was inconclusive. Multiple injections always produced marked loss, recovery ensuing on discontinuance of injections if few in number.

Antitryptic Values.-Multiple injections produced marked rise in values, strikingly parallel in gross effect to the rise produced in the same species of animal by single injections of dipthteria toxin, or of emulsions of living bacteria.

From these experiments it would appear that, however great the share taken by bacteria and protozoa in initiating the disease-complex of infection, the net result is, perhaps to a large extent, a state of true auto-intoxication.

The results obtained suggest that such auto-intoxication is in part directly due to absorption of derivatives of the infected cells themselves, and only indirectly to the absorption of bacterial products.

CHEMICAL SOCIETY.
Ordinary Meeting, June 2nd, 1910.

Prof. HAROLD B. DIXON, M.A., Ph.D., F.R.S., President, in the Chair.

(Concluded from p. 10).

162. "The Direct Union of Carbon and Hydrogen. Synthesis of Methane." Part II. By WILLIAM ARTHUR BONE and HUBERT FRANK Coward.

In continuation of their previous experiments (Trans., 1908, xciii., 1975), the authors are now able to report a not less than 95 per cent yield of methane from less than 0.03 grm. of highly purified sugar charcoal (containing not more than o'06 per cent of ash and less than o'1 per cent of hydrogen) by maintaining it in a steady current of pure dry hydrogen at 1150° for from seventeen to twenty-five hours continuously, thus completing the proof of the direct synthesis of methane from its elements at this temperature. 163. "The Existence of Racemic Compounds in Solution." By ALBERT ERNEST DUNSTAN and FERDINAND BERNARD

THOLE.

In continuation of their previous work (Trans., 1908, xciii., 1815) the authors have investigated the viscosityconcentration relationships for various solutions of active and racemic compounds. The sec.-octyl hydrogen phthalates (d-, -, and inactive forms) were examined in benzene solution, and were found to give identical curves. Hence the inactive ester does not exist in solution as a racemic compound.

no difficulty has been experienced in separating the osyritrin from them.

Recently a considerable consignment of the leaves and stalks of Osyris abyssinica, Höchst, has been received for examination from the Transvaal. The material as received gave a much deeper coloured extract than Cape sumach, and produced a darker leather of rather poor quality. The aqueous extract, on keeping, deposited a certain amount of solid matter, but this did not appear at all similar to the crystalline deposits previously obtained from Osyris compressa. The new material was therefore examined more closely.

The Tannins.-Analysis showed that the leaves contained about 23 per cent of tannin.

The leaves were extracted with water and the extract evaporated to a small bulk, when there separated a quantity of solid matter consisting of colouring principle and the phlobaphen of the tannin. The aqueous solution was fractionally precipitated with lead acetate, and the last fraction washed and decomposed with hydrogen sulphide. After removal of the lead sulphide and the excess of hydrogen sulphide, the solution was evaporated to dryness. It yielded a reddish brown tannin of the catechol series. This tannin is much darker than that previously obtained from Osyris compressa, and it also shows more readiness to lose water and form the phlobaphen.

Exraction of the Colouring Matter. The coarselyground leaves were thoroughly extracted in a Soxhlet apparatus with 90 per cent alcohol, and the residue, after removing the spirit, was poured into water. The green mass which separated was filtered off, and, after drying, was extracted separately with light petroleum and alcohol. The alcoholic extract set to a viscid semi-solid mass on concentration. By pouring into excess of boiling water it was possible to separate a large quantity of dark brown insoluble matter of a phlobaphen nature, and on cooling the aqueous solution, a dirty brown solid deposit was obtained, which, after repeated precipitation with water from alcoholic solution, yielded a mass of small crystals. These were purified by several crystallisations from dilute alcohol, and finally from boiling water.

A further quantity of the same substance was obtained by deposition from the main filtrate containing the tannin. melting at 189-190°, after softening at 185–186° (osyritrin The substance formed small greenish yellow needles, melts at 1860). After being dried at 160°, it gave on analysis:

C=53'17, H = 4'51. C27H28016 requires C = 53'28, H 4'60 per cent.

=

After being dried at 160° : Found, H2O = 7'90. C27H28016,3H20 requires H20=8.15 per cent. On hydrolysis with dilute acid the substance gives an insoluble colouring matter and a sugar. The latter is dextrorotatory, and gives an osazone Estimation of the colouring at 204-205°. C15H1007 gave 45'45. The equation C15H10O7+2C6H12O6 requires C27H28016 + 3H20 C15H100745'61 per cent. The colouring matter has all the properties of quercetin, and the penta-acetyl compound, prepared in the usual manner, melted at 190°, and gave on analysis :

matter

AULD.

=

=

C=58 48, H=3'93. Calc. C=58.59, H 3'90 per cent. The glucoside is therefore identical with osyritrin. 165. "The Relative Influence of the Ketonic and Ethenoid Linkings on Refractive Power." By IDA SMEDLEY. The molecular refractions of the following unsaturated ketones were determined in chloroform solution :Ma. Ms. My. 48.52 5133 53'79 70 48 75'30 80.2. 64'19 70 76

Benzylideneacetone Benzylideneacetophenone

84.04 92'17 87.59 98.17

121'5