off reddish-brown vapour of Br,* which somewhat resembles Cl in smell, but differs by its colour and by turning moist starch orange-red, the colour disappearing by heat. This last test is best performed by dipping the wetted end of a glass rod into some powdered starch, and then moistening the adhering starch by breathing upon it several times. If MnO be mixed with the KBr powder before adding H2SO4, the Br is evolved in greater quantity and more readily.

263. Cl-water, added carefully drop by drop to a little KBr solution whilst shaking it, liberates Br, which gives the solution an orange-red colour; excess of Cl-water must be carefully avoided, as it destroys the colour :

KBr+Cl = Br+ KCl.

:

On warming a part of this solution in a dish, brown fumes are given off which colour orange-red some starch held in the vapour as described in the preceding reaction.

If CS2 be added to another portion and the liquids are then well shaken together, the Br is dissolved by the CS2 and on standing the reddish brown CS, solution sinks to the bottom; on adding a little KHO solution and shaking well, the colour of the Br again disappears from the CS,, owing to the formation of the colourless salts KBr and KBRO:

6Br+6KHO=5KBr+KBrO3+3H2O.

264. A mixture of solid KBr and K2Cr2O, heated with strong H2SO4 as in (260), gives off reddish-brown vapour of Br:

:

6KBr+ K2Cr2O7 +11H2SO4 = 3Br2+ Cr2(SO4)3 + 8KHSO4 +

7H20.

This vapour, if received in a little water, imparts its colour to the water; if the liquid is shaken up with CS, however the Br is dissolved away from the water by CS2, which forms a brown layer at the bottom of the water by shaking

* KBr solution does not so easily give off Br vapour, but the liquid becomes coloured reddish-brown.

well after addition of AmHO or KHO the colour of the solution is destroyed. (Difference from HCl.)

Hydrogen bromide, or hydrobromic acid (HBr), is a gas resembling HC1; its solution differs by evolving Br when heated with MnO2.

265. AgNO3 added to a little KI solution, gives a yellow precipitate (AgI), which is easily coagulated by heating or shaking the liquid: it is insoluble in HNO,, and very slightly soluble in AmHO.

This precipitate differs from AgCl and AgBr by its colour and its slight solubility in AmHO, but more particularly by its behaviour when heated with strong H2SO4. If the precipitate, coagulated by heating or shaking the liquid, is allowed to settle and the liquor decanted, then strong H2SO4 poured in and heated with the precipitate, the acid becomes coloured red, and violet vapours of iodine are seen especially on looking down the tube; the colour is particularly distinct after the tube has been allowed to cool for a short time. This vapour of iodine may be detected when almost invisible by holding in the air of the tube a glass rod previously dipped into freshly-made starch solution, which will become coloured blue.

266. CuSO4 mixed with H2SO, or FeSO4 solution, which reduces it to CuSO4, gives a white precipitate (Cu2I2); warming, or the addition of a drop of AmHO, causes this precipitate to form more quickly :—

2CuSO4+H2SO3 + H2O = Cu2SO4+2H2SO4

2KI+Cu2SO4 = Cu2I2+ K2SO4 •

If this precipitate is allowed to settle and the water poured aff, or if it be filtered, and the precipitate is then warmed with strong H2SO4, the liquid becomes violet and gives off violet fumes of I.

Solutions of chlorides and bromides are not precipitated thus by CuSO4.

267. Solid KI if warmed with strong H2SO4 gives off violet vapour of I,* which, if in large quantity, cools on the sides of the tube to black solid I: this vapour colours starch solution blue, as may be proved by dipping a rod moistened with freshly-made starch solutiont into it, or by pouring the heavy fumes out into a white porcelain dish whose inside has been wetted with starch solution, or by dipping into the vapour a slip of filter paper moistened with starch solution. The I is more readily and plentifully evolved if some MnO, be mixed with the KI before adding the H,SO4.

268. Cl-water added drop by drop (carefully avoiding excess, which would form colourless ICl), sets free I, which dissolves in the water giving a brown liquid; divide this into three parts :

1. Heat one part; violet vapour of I is given off and is best seen by looking down the mouth of the tube at a white surface. The colour is most distinctly seen if, before heating, some strong H2SO4 is added to the liquid. If a glass rod is dipped into starch solution and held in the vapour, the starch becomes blue.

2. To another portion starch solution is added, which gives an intense blue colour often appearing black unless much starch is added and the liquid largely diluted; on warming the solution the colour disappears, but often reappears when the liquid is cooled. To insure the disappearance of the colour on being heated, the solution must be diluted and the starch added in excess.

3. To the third portion add a drop of CS, and shake well; the CS, dissolves the I, and gradually settles to the bottom. having a beautiful violet colour. Add a little KHO solution and shake well, the violet colour of the CS, will disappear owing to the formation of the colourless salts KI and KIO.

Note. The last two methods of testing for the free I are by far the most delicate known; but Cl-water is by no means the best reagent for

* KI solution does not so easily give off purple vapour, but the liquid becomes reddish-brown or red.

The starch solution must always be freshly made, as after being kept it loses the property of yielding a blue colour with iodine.

K

setting I free from its compounds, because if added in excess it combines with the I and prevents its detection. The reagents described in (269) are much better suited for liberating I from its compounds.

2 4

269. If a little KNO, is added to the KI solution, and then any dilute acid (HÃ, HCl or H2SO4) poured in, HNO2 is produced which liberates I. A solution of NO in dilute H2SO4, made by heating Pb(NO3)2 strongly and passing the red fumes into dilute H2SO4, also sets free I from KI. The I thus liberated in solution may be identified by the three methods described in (268), of which the third is the most delicate.

These two reagents for liberating I, do not hinder its detection if they are added in excess, and they are also useful because they do not liberate Br from its compounds.

270. HgCl, gives with KI solution a scarlet precipitate, with KBr or NaCl it gives no precipitate; PbÃ, gives with KI solution a bright yellow precipitate, with KBr or NaCl a white precipitate.

Hydrogen iodide, or hydriodic acid (HI), is a gas resembling HCl and HBr; its solution differs by evolving I when heated with MnO2.

271. A chloride, bromide, or iodide, is easily detected by the foregoing tests. The reactions given in paragraphs (259) and (260) are perhaps the most characteristic tests for a chloride, those described in (263) and (264) for a bromide, whilst an iodide is detected with the greatest ease and certainty by reaction (269).

The colour of the precipitate produced by AgNO, solution in the solution made acid with HNO,, and its behaviour with AmHO, though less reliable than the above tests, serve to indicate which of these acid-radicles is present, AgCl being pure white and very easily soluble in AmHO, AgBr being pale yellow and not readily soluble in AmHO, whilst AgI is primrose-yellow and almost insoluble in AmHO. See also the distinctive action of hot strong H2SO4 on AgI, described in paragraph (265).

272. (s) The detection of a chloride, bromide, and iodide when occurring together.

The method of examination varies according to whether an iodide is present or absent; since a chloride cannot be detected in the presence of an iodide, the latter, if present, must be separated before proceeding to examine for the chloride (see 273). In paragraph (274) is given a process for detecting an iodide and bromide; it is of value only when a chloride has not to be tested for.

273. (s) To a small portion of the solution, made just acid if necessary by addition of dilute H2SO4, add a little cold freshly prepared starch solution and then strong HNO3, or either of the reagents mentioned in (269), drop by drop; a dark blue coloration* shows the presence of an iodide. The remainder of the solution is examined for a chloride and bromide by the directions given below under A or B, according as an iodide is present or absent:

4

A. An Iodide is present.-Add to the solution a reagent consisting of a mixture of CuSO, solution with half its measure of strong H2SO, solution, and warm gently for a short time; Cu2I will be precipitated (266). In order to be sure that the iodide is entirely precipitated, a small quantity of the liquid must be filtered and warmed with a little more of the above reagent: if any further precipitate is caused, the filtered portion is returned to the unfiltered liquid, more of the reagent is added to the whole, which is then warmed for a time and again tested as above, to see if the precipitation is complete; these trials are repeated until a small portion, on filtration, gives no further precipitate when warmed with more of the reagent. The whole is then filtered till clear, pure NaHO is added in excess to the filtrate and the liquid boiled. The precipitate thus produced is filtered off, and the filtrate, now free from iodide, is tested for bromide and chloride by the directions given in column B.

B. No Iodide is present.—Evaporate the solution (made

* Black, unless the starch solution has been added in sufficient quantity to the dilute solution,