been determined by Professor W. H. Miller, of Cambridge. The truth is, that the mineral in question was submitted to this eminent authority, but Professor Miller was unable to separate a crystal from the aggregated masses in a condition fit for accurate measurement. Professor Church's own observations had led him to conclude that the crystalline form of the mineral was oblique, but his observations having been made with the inferior instrument, the microscopic goniometer, may almost be said to require confirmation--a duty which Professor Maskelyne, at the meeting, volunteered to undertake.

The hydrated phosphate of calcium and aluminium, although containing the same amount (12 per cent.) of water, is believed by Mr. Church to differ from the mineral described by M. Damour as occurring in the diamond

sands of Bahia.

Lastly, with respect to the double arseniate of copper and lead, it should have been described, not as a variety of olivenite, but as a distinct species belonging to the olivenI am, &c.

ite group. November 13.

THE REPORTter.

The Past and Present History of Alum.

To the Editor of the CHEMICAL NEWS. SIR,-In your last issue I find mention of my " Dictionary of Chemistry" by Mr. J. Carter Bell, F.C.S., relative to the article "Alum." Mr. Bell remarks :-" This is a most unfair description of such important alum works as Mr. Spence's are, and I am surprised to find such a statement in such a compendious dictionary as Dr. Muspratt's professes to be." Your correspondent, Mr. Bell, applies the word professes to a book that all the first savans of the world have eulogised, and which Baron Dumas, of Paris, calls "the great chemical work of the age.' As it is the first and only time the information in the Dictionary has been called into question, I feel bound to notice Mr. Bell's statement, and I beg leave to say that at the time each monograph appeared, it contained "a complete position of the state of the chemical manufactures, with the latest and most improved processes" (vide Preface). Your readers must bear in mind "Alum" nearly fourteen years since!

"

A set of six cylindrical cells of Groves' battery were thus, with the same materials, amalgamated, equipped, and primed for action in a quarter of an hour.

No friction is needed; the plates should be well drained from excess of mercury, lest they become brittle, though this danger is lessened by the rapidity of the process. The rationale of my preference for HCl and HgCl, is sufficiently obvious. I am, &c. B. W. GIBSONE, F.C.S. November 7.

MISCELLANEOUS.

selected for the ensuing year :-President.-Major-Gen. Ed.
The Royal Society.-The following are the officers
Sabine, R.A., D.C.L., LL.D. Treasurer.-William Allen
Miller, M.D., LL.D. Secretaries.-William Sharpey, M.D.,
LL.D., George Gabriel Stokes, Esq., M.A., D.C.L.
Foreign Secretary.-Professor William Hallows Miller,
M.A. Other Members of the Council.-John Frederic
Bateman, Esq., Lionel Smith Beale, Esq., M.B., William
Bowman, Esq., Commander F. J. Owen Evans, R.N.,
Edward Frankland, Esq., Ph.D., Francis Galton, Esq.,
John Peter Gassiott, Esq., John Edward Gray, Esq.,
Ph.D.; Thomas Archer Hirst, Esq., Ph.D., Sir Henry
Holland, Bart., M.D., D.C.L., William Odling, Esq.,
M.B., Sir John Rennie, Knt., Professor Warington W.
Smyth, William Spottiswoode, Esq., M.A., Paul E.
Count de Strzlecki, C.B., D.C.L., Vice-Chancellor Sir
W. P. Wood, D.C.L.

Reaction of Metallic Thallium on a Few

Solutions of Metals.-Thallium immersed in solutions of sulphate, nitrate, and acetate of copper, deposited in AgONO, deposited silver; thallium in AuOSO, deposited flakes (similar to iron) metallic copper; thallium in gold; thallium in HgOSO, (sulphate of the red oxide) ex-deposited mercury; thallium in PbOA deposited lead. The gold appeared in flakes, partly floating and partly at bottom; mercury in small globules; lead as a crystallised mass on top of the thallium-very similar to thallium on zinc. When thallium is put into a solution of the nitrate of cobalt, the metal is covered with a blue coating, which from it becoming green on exposure to the air, appears to be similar to that basic salts of cobalt precipitate which is thrown down by KO.—Walter C. Reid.

was written

Is it likely, Sir, that "Alum" should be the only incomplete article in the book? Why Mr. Bell has thought fit now, at this late date, to find fault with it is best known to himself. If I should re-write " Alum" for a "next edition of the Dictionary, a better description will be given," if requisite. I am, &c.

SHERIDAN MUSPRATT, M.D., Prof. of Chemistry. College of Chemistry, November 13.

Ready Mode of Amalgamating Zinc Plates.

To the Editor of the CHEMICAL NEWS. SIR,-I venture to send you a method of almost instantaneously amalgamating corroded zinc battery plates, which occurred to me recently, after some twenty years' trial of different plans; perhaps economy of time in even humble matters of detail may be worth record where the process is of repeated occurrence.

The following treatment in the case of thickly oxidised plates will yield in speed and effectiveness to few:-Place in a flat dish two ounces of common hydrochloric acid, one drachm of a saturated solution of bichloride of mercury (corrosive sublimate), and half an ounce of the latter metal; lay the zinc, without previous scouring, in the liquid mixture, and gently smear the mercury over surface of the plate with a toothbrush; the mercury will readily and thoroughly adhere to each portion of the surface as the oxide is rapidly dissolved by the HCl.

As a means of comparing speed, in seventy seconds I completely coated inside and out a cylindrical plate of forty square inch surface, whose interior was rather inaccessible and very corroded.

Clay and Glycerine for Modelling. We read in Cosmos that a mixture of clay and glycerine, which keeps its plasticity for any length of time at all temperatures, has been found very useful by modellers. The clay must be well dried before it is mixed with the glycerine. It is said that the mixture can be used in place of wax for the most delicate work.

ANSWERS TO CORRESPONDENTS.

All Editorial Communications are to be addressed to the EDITOR, and Advertisements and Business Communications to the PUBLISHER, at the Office, 1. Wine Office Court, Fleet Street, London, EC. Private letters for the Editor must be so marked.

adopt the views of the writers. Our intention to give both sides of a In publishing letters from our Correspondents we do not thereby question will frequently oblige us to publish opinions with which we do not agree.

Dr. Muspratt.-Received, with enclosures.

Jeremias. We must decline our correspondent's last communication. Mr. Walker.-The patents have come to hand. Ninety-nine.-Both errors are so obvious as to correct themselves. As regards the second pointed out, it must have occurred to every reader but our acute correspondent that the decimal point was in the wrong place, and that it should have been 56.

Books Received,-"Pereira's Manual of Materia Medica," edited by Dr. Farre.

On the Application of Spectrum Analysis to the Detec-
tion of Chlorine, Bromine, and Iodine, by ALEXANDER
MITSCHERLICH*.

THE difficulty of recognising small amounts of chlorine,
bromine, and iodine in a mixture of haloid salts is well
known; and it is found impossible to detect mere traces
of these bodies in such mixtures by any hitherto
known method. I have, however, discovered a means of
recognising the smallest amounts of these substances by
the use of spectrum apparatus.

be added to the solution. The greater part of the iodine and bromine will be found in the precipitate, which may be tested as before described.

The following results will serve as an illustration of the accuracy and precision of the process. To a pound of common salt which contained no bromine five milligrammes of bromide of sodium were added, and to the solution one decigramme of nitrate of silver. The precipitate was tested in the way just described, and after the spectrum of chloride of copper had been observed for some time, the spectrum of bromide of copper was distinctly visible for five minutes. A further addition of nitrate of silver to the solution gave a precipitate which showed the spectrum for six minutes.

Similar experiments made with iodine compounds gave equally conclusive results, and proved that a ten-millionth part of iodine or bromine may be detected in chloride of

sodium.

The residue of six and a-half pounds of sea water taken Heligoland [precipitated with silver?] showed the spectrum of the bromide for seven minutes. Iodine could not be recognised, probably because the quantity of water was too small.

The haloid salts of copper are the most difficult to de-
compose by heat, and they are therefore to be preferred
for spectrum investigations, which are best made in the
following way-The substance to be examined, well
dried, is intimately mixed with half its weight of sul-off
phate of ammonia and one-tenth its weight of oxide of
copper. This mixture is placed in a globular enlarge-
ment of a combustion-tube, one end of which is connected
with a hydrogen gasometer, the opposite end being open.
A stream of hydrogen is passed through the tube, and
heat gradually applied to the mixture. The hydrogen
being ignited, the first appearance seen in the spectrum
apparatus is a brightness in the green in which, how-
ever, no definite spectrum can be perceived; but after-
wards the spectrum of the haloid salt of copper is dis-
tinctly visible.†

When present in small amount the chlorine compound
is best recognised by the lines at 105 and 109, and by
the brightness near 85 and 87; the bromine compound
is detected by the brightness at 85, 88, and 92; and
iodide of copper by the brilliancy at 96, 99, and 1024.
By this method, and without further trouble, per
cent. chlorine, 1 per cent. bromine, and 1 per cent. of
iodine are easily recognised, and a practised observer
may detect much smaller quantities. One disadvantage
of this method is that at a high temperature the sulphate
of ammonia decomposes, and the spectrum of ammonia,
which renders the others obscure, is obtained. With some
compounds the addition of this salt is unnecessary, as in
the cases of the chlorides of silver and mercury.

The use of sulphate of ammonia, indeed, renders it
difficult to recognise the haloids when mixed with each
other in very small proportions, and it is better to pre-
cipitate the haloids first by a silver salt, mix the dried
precipitate intimately mixed with twice its weight of
oxide of copper, and employ this mixture in the hydrogen
apparatus before described. By this method as little as
th per cent. of chlorine, 4th per cent. of bromine, and
4th per cent. of iodine in the silver precipitate can be
recognised.

The spectra of the haloid salts appear consecutively, that of the chloride first, then that of the bromide, and lastly that of the iodide of copper. Their appearance in this order depends on the different volatility of these salts. Chloride of copper volatilises considerably below a red hoat, the bromide somewhere near redness, and the iodide at a low red heat. The slower the volatilisation is conducted, the more certain are the results of the analysis.

When only traces of iodine and bromine compounds are present in a large excess of a chlorine compound, about the tenth of a gramme of nitrate of silver should

Abstract from Poggendorf's Annalen, No. 8, 1865, p. 629.

f Drawings of the spectrum given by the several haloid salts are given with Poggendorff's Annalen, No. 4, 1864.

VOL. XII. No. 312.-NOVEMBER 24, 1865.

A small quantity of water from the Dead Sea showed a large proportion of bromine, but no iodine. The mother liquor from some salt works showed much bromine, but

no iodine.

Where organic matters are to be examined for the haloids a tube with two bulbs must be connected with the hydrogen apparatus. The bulb nearest the flame is is placed in the other. Both being heated, the products filled with oxide of copper only, and the organic matter of the heated organic matter over the oxide of copper reduced by the hydrogen in the second bulb. this way the smallest amount of chlorine, bromine, or large excess of the other. be recognised; and, indeed, traces of one in

iodine

a

may

In

The author thought it might be possible to arrive at something like quantitative results by noting the time which the several copper compounds took to volatilise; but he found it impossible to avoid large sources of error.

Researches on the Influence of the Electro-negative Elements on the Spectra of Metals, by M. E. DIACON.‡ THE method of analysis founded by Kirchhoff and Bunsen on spectrum observations have given results so remarkable that its utility in chemical researches is not to be contested. Nevertheless, the principle on which it rests is only true under certain determined conditions. From my own observations and those of A. Mitscherlich it seems certain that the different compounds of a metal do not exhibit an identical spectrum, and in this paper I propose to collect the experiments which show the influence of the electro-negative element on the radiations emitted by different salts of the same metal. In a series of spectrum researches made on a great number of mineral waters, I remarked some appearances which led me to study attentively the spectrum of barium in the presence of salts of lime or strontia. Under such conditions I not only observed faint lines not indicated for either of these metals, but was further struck by the remarkable change which the spectrum of barium underwent, when a pearl of chloride of calcium was introduced into the same flame.

This phenomenon, consisting in the much greater brilliancy acquired by certain green lines of barium, appeared Abridged from Annales de Chimie et de Physique, Sept., 1865.

244

Influence of Electro-negative Elements on the Spectra of Metals. {CHEMICAL NEWS,

to me so strange at the time-when it was fully admitted that in the case of a mixture the spectra were simply superposed without influencing each other-that I called in other observers. The observation of a green band, pale and ill-defined, which the spectrum of calcium sometimes presents, and which coincides with the second of the green lines of barium, led me to ascribe it to absorption; the diminished intensity of this line might have produced the sensation of greater brilliancy in the other lines. But all the experiments only gave me negative results; the phenomenon was only produced when the two pearls were in the same flame, and very near to each other. Hence it seemed necessary to admit that the two bodies might act on each other and modify their spectra.

The appearances noticed above not appearing when even volatile compounds of calcium were substituted for the chloride, I soon became certain that they did not belong to the metals themselves. There remained then the action which the chloride of calcium or the chlorine disengaged during the rapid oxidation of the metal in the flame might have on the barium. Experiments which I made with quite a different object set the ques

tion at rest.

Many metals whose chlorides are volatile give no spectra in a Bunsen's flame. It is reasonable to believe that the rapid decomposition these chlorides undergo in an oxidising flame is the principal cause of this, and that it would be otherwise, with many of them at least, if the temperature sufficed. This I hoped to realise by placing them in a chlorinising flame obtained by the combustion of hydrogen in chlorine. The apparatus adopted after some trials is the following:-A box, blackened in the inside, 60 c.c. high, and 15 c.c. wide and deep, receives the blowpipe; two openings in the sides, to be closed at will by glass, serve one for the observations, the other for the introduction of the matters to be experimented upon. An inclined tile of refractory earth forms the roof of the box, and directs the products of combustion into a second box partly superposed, and of about the same dimensions as the first. This second box contains a number of laths placed horizontally, and carrying slaked lime, so as to leave a free passage for the gas. A long wooden exit tube placed below the laths carries the gas not absorbed by the lime into the chimney, a gas lamp above the opening of the tube acting as an aspirator. The rest of the apparatus is composed of a blowpipe and two gasometers placed outside the laboratory, one of glass for chlorine the other for hydrogen. The gases are conducted by lead pipes, and afterwards by caoutchouc tubes to the blowpipe; the chlorine in a small tube placed within a larger, which carries the hydrogen.

The easy reducibility of chloride of platinum allows wires of this metal to be used as supports. The matters experimented upon are placed in the interior of the flame, so that they may be in the presence of an excess of chlorine, and beyond the influence of external air. Lastly, the spectroscope is directed to that part of the flame situated immediately above the wire support. Under these circumstances a great number of metals give spectra of very variable persistence.

Chloride of copper gives a magnificent spectrum; that of chloride of bismuth is also very brilliant. Chloride of lead shows a great number of brilliant lines, the positions of which seem to differ from those of the metal in the gas flame. With tin, and more especially with antimony, it is often impossible to distinguish the lines which characterise them, on account of the diffused light

Nov. 24, 1865.

produced by the decomposition of the salts. The same is the case with the chlorides of platinum and silver. Chloride of gold, however, gives a well-defined spectrum formed of a series of beautiful green lines. With chloride of palladium bright blue lines are seen, but only for a very short time.

Chloride of manganese gives a well-defined spectrum, like that of the salts of the metal in the gas flame. Chloride of iron shows a great number of lines bathed by the light due to the incandescence of solid particles carried by the vapours. The chlorides of cobalt and nickel present the same phenomena, but with less intensity, so that, notwithstanding the great number of lines which constitute them, it is easy to distinguish the spectra of the two salts, the red lines being more numerous in nickel, the yellowish-green lines more numerous with cobalt.

Uranium, cerium, chromium, glucinum, aluminium, and magnesium gave only negative results; to obtain the spectra of these metals it is necessary to experiment with anhydrous chlorides.

Experiments made in the way described showed that certain chlorides decomposed less rapidly in a chlorinising flame, and, therefore, gave spectra; but the experiment with chloride of barium led to results of special interest. That salt gave a spectrum quite different from that of the metal; the large number of lines which characterise the latter had disappeared, and were replaced by a small number of very bright lines. I immediately examined the chloride of strontia, and was no less surprised at perceiving neither the blue nor the orange line, seen so brilliantly in the gas flame. The spectrum of chloride of calcium showed differences no less striking and real. With lithium nothing peculiar was seen. The chlorides of sodium and potassium appeared not to give spectra. Chloride of thallium appeared to present no change, even in the intensity of the characteristic green line.

Modifications so profound can only be explained by admitting that each compound has a peculiar spectrum, but the specific nature of metallic spectra has been so clearly laid down in principle by Bunsen and Kirchhoff that it seemed necessary to support by new experiments a fact so contrary to the results obtained by these authors. If chlorides have peculiar spectra, it must also be the case with bromides and iodides. It was hoped that similar experiments to those made with chlorides would demonstrate this; but attempts to obtain a flame with bromine and iodine and hydrogen did not succeed.§

The author now fell back on the first experiment mentioned in this paper. The appearances observed when baryta was heated in the presence of a pearl of chloride of calcium he assured himself were due to the superposition of the spectra of oxide and of chloride of barium. The rapid decomposition of the chloride of calcium surrounding the baryta with an atmosphere of chlorine chlorinised a part, and, consequently, determined the appearance of lines characteristic of chloride of barium. A pearl of chloride of barium by itself presented the same appearances, but with less intensity at the moment it was introduced into the gas flame.

A comparative study of the spectra given by chlorides in chlorine, and by oxides in a blow-pipe flame fed with air or even pure oxygen, soon demonstrated that the lines

memoir of M. Mitscherlich (see CHEMICAL NEWS, vol. vii, p. 99) and
repeated that gentleman's experiment, but came to the conclusion
that the spectra obtained by his method were simply those of the
observation.
metal, and that the lines specially belonging to the chlorides escaped

At this stage of his investigation the author met with the

"On some of the most important Chemical Discoveries made all loose and useless fibres. One of the most perfect