Acid, the Sulphur contained in it being thereby converted into Sulphuric Anhydride, and combining with the lead to form the insoluble Lead Sulphate, from which the amount of the lead is estimated, by the fact that pure Lead Sulphate contains 68.3 per cent. of Lead. The salts of Iron and Copper have to be dissolved out separately with this method.

In the Volumetric method the ore is treated as in the Gravimetric process, Lead Sulphate being formed as before. The Lead Sulphate is dissolved in Ammonium Acetate, and treated with Ammonium Molybdate, the White Lead Molybdenate being formed, which is carefully estimated in a specially graduated burette.

Copper. In the dry or fusion process for assaying Copper, a flux is employed, consisting of Borax, Soda, and Bi-Tartrate of Potash, the process being very much the same as with Lead.

The principal wet method, with Copper, is the Electrolytic, the ore being dissolved in a solution of Sulphuric Acid, and subjected to the action of an electric current, the Electrodes being Platinum, the increased weight of the Cathode, the negative Electrode, measuring the weight of Copper in the solution, and therefore in the sample, if all the Copper has been brought down. A favourite form for the Electrodes is, two concentric cylinders, the inner one being the Cathode. Care must be taken, in using the Electrolytic method, to eliminate the Silver, Arsenic, and Cadmium, which would come down with the Copper.

There are two Volumetric methods for assaying Copper. In one, Potassium Cyanide is added to an Ammoniacal salt of Copper, the insoluble salt Cyanide of Copper being formed, which is estimated in the usual way. The ore is treated with a mixture of Nitric and Sulphuric Acids, and then with Ammonia.

In the other method, Potassium Iodide is added to Cupric Acetate, prepared from the ore, Cupric Iodide being formed, and it is then treated with Hyposulphite of Sodium, Hydriodic Acid being formed. The quantity of Copper in the sample is estimated from the quantity of Iodine liberated; one atom being liberated for each atom of Copper.

Tin.—A certain quantity of the ore is taken,

and crushed sufficiently to pass through a 40 mesh sieve, and concentrated, the concentrates being roasted in an iron roasting dish in a muffle furnace. The product is treated with aqua regia; the product of this, after filtering, is ground to an 80 mesh sieve, and is then assayed with a flux, by heat in a muffle furnace, Cyanide of Potassium being used for a flux by some assayers; a mixture of the Bicarbonates of Potash, and Soda, together with some Borax, and some Charcoal, by others; and Sodium Carbonate, with Lime, by others. The process is similar to that with the other metals.

Iron.-Iron may be assayed by the fusion method, and by two wet methods. In both wet methods, the iron in the ore is all reduced to the Ferrous state, and in one is then treated with Permanganate, in the other with Bichromate of Potassium, the Iron being thereby raised to the Ferric state, the volume of the standard solution required to bring all up being the measure of the Iron in the ore. With Permanganate of Potash, the salt being of a deep red colour, the Iron solution turns yellow, as the Permanganate is added, till all the Iron has been raised to the Ferric state, when the first drop in excess causes some of it to turn pink. With the Bichromate, the solution turns green, and the end of the process is known by placing a drop of the solution on a white tile, and adding Ferrocyanide to it. As long as any Ferrous salt remains, the solution turns blue. In all cases the metal, or the salt of the metal, is carefully weighed in specially delicate balances, from which even the air is excluded, and which are kept scrupulously free from dust. The percentage of metal in the ore is found by a simple arithmetical calculation, from the weight of ore in the sample, and the weight of the metal or the salt of the metal finally obtained. But it is necessary to be careful, as the sampling proceeds, to see that credit is given for all the metal obtained. Sometimes minute pellets of the metal are obtained in the crushing process, before any particular sieve. These are carefully collected, weighed, and credited, not to the final sample, but to the sample and weight at which they occurred. Thus, if say a pennyweight of metal were obtained at the sample going through the sieve at which the sample

was reduced from to, the pennyweight would be credited as a pennyweight in so many pounds the weight of the sample at that stage.

Assembling. This is a comparatively recent practice in its application to engineers' work, being only possible when an interchangeable system is adopted. The term signifies that all the parts of which a motor or mechanism is built up are brought together finally without any correction by hand fitting; that any parts which are identical in shape and dimensions may be taken at random from a pile, and put together without the assistance of cutting or of scraping tools, differing therein from mechanisms which require the numerous corrections of the fitter.

This ideal is not always fully realised even in shops where parts are nominally assembled, but it is absolutely so in large numbers of mechanisms. Without it, cheap production, and the replacement of worn and broken parts as required by customers, would not be possible.

In order to the realisation of a perfect system of assembling, the machinery employed in the production of the parts must be so designed that the tools and appliances used shall both cut and size, that is embody the dimensions as well as the formation of all similar pieces. If this is not done on one machine, it must be on another at a later stage, i.e. a grinder succeeding the lathe, or planer. A familiar illustration is afforded by the Automatic Screw Machines, in the use of Box Tools, which size as well as shape, and the cutting of screw threads by fixed dies.

And then, to prevent loss of time when the parts come into the hands of the assemblers, the separate pieces are all gauged as they leave the machines. Fixed gauges of various kinds are used, and lads or girls frequently handle them. Even here things are often so arranged that the pieces automatically size themselves, according as they fit, or do not fit gauges, which lie in a course along which they are compelled to travel.

Assembling in its strict and absolute sense is only practicable with the smaller mechanisms, of which Small Arms afford the best illustration.

As dimensions increase, the devices used for small work are no longer practicable, neither can the effects of spring, of temperature, and other variables be eliminated. In all machines of medium and large sizes some fitting and adjustments become more or less necessary. Then it is a question whether the work is sufficiently often repeated to make it worth while incurring the expense of working very closely to absolute gauged dimensions, when mutual fitting might be equally well adapted to the case. That is a question which is answered differently by different firms.

The operation of assembling is done on special benches set apart for the work, and it may happen that the assemblers will put together the whole of a small mechanism, or a separate section of a mechanism will be entrusted to each worker, which is more desirable and economical in some instances. Where parts are put into stock largely, the complete fittings may only be made up as required, and the assemblers kept busy on detached portions, which they assemble to the exclusion of everything else. Special devices are employed in connection with assembling, such as boards or trays carrying the detached components of the mechanism, for convenience of picking them out, and special stands to rest the partly completed pieces upon, holding them steady while the assembler is at work. Devices such as indicators and various gauges are also employed, for getting correct distances and positions of related parts, not because the latter are not machined correctly, but because such adjustments are unavoidable in many mechanisms, though not necessarily involving cutting or working by tools. When, however, the latter practice is involved, we step from assembling at once into fitting. Fig. 150, Plate XII., is an example of a modern assembling shop.

Assistant Cylinder.-A small cylinder designed by Mr Joy to relieve the strain on the eccentrics, and the valve gears of heavy marine engines. The dead weight of the slide valve, its inertia, and the friction of the valve on its seating, make up a total which stresses the eccentrics and gears severely, and produces tensile and compressive strains in the valve rod. The assistant cylinder sets up forces equal and

opposite in character to those caused by the weight, inertia, and friction of the valve. It is a small cylinder bolted on top of the steam chest. The slide valve rod passes up into it through a gland and serves as a valve for admitting steam to the lower side of the piston. The piston is hollow, and holes in it in communication with slots in the cylinder valves receive steam from below, which is passed to the upper side of the piston to overcome the tension in the rod due to the inertia, and to start the valve on its downward stroke. The volume of steam thus admitted is adjusted to meet any conditions. See also Balance Cylinder.

A-Standard.-See A-Frame.
Astatic Needle.

See Galvanometer.

Astragal is a moulding or bead used in architecture. It is either applied to the top of a shaft where the capital commences or to the base. It is semicircular in form, with a plain surface, although there are Roman and Greek examples in which the surface is carved into beads or leaves. See Architecture.

Astronomical Instruments. The scope of this work does not include the description of these instruments. But there is one branch of the subject with which the engineer has become identified, namely, the construction of big telescopes, of which some account should be given, and that of dividing engines for the accurate division of circles and lines. These will be treated under Dividing Engines, and Telescopes.

Atlantic Liners. In sketching the growth of the Atlantic liners from the standpoint of the engineer we outline the history of ocean steamships in general. There are no liners so large or so swift as those which cross the Atlantic, and all pioneer work has been done in that service. That ocean was the first to be crossed by steam, and always on its waves the newest designs have been put to the test, for more than sixty years past. We do not propose to give a detailed account of historical matters that are familiar to most people, but to indicate the epochmaking engineering developments of that history. For the story of the Atlantic liners is one of epochs in engine and boiler design, in auxiliary machinery, in the design and construction of hulls, in the displacement of old materials by

new, in safety, speed, dimensions, and accommodation. This is the point of view from which this article is written.

In 1838 the Royal William, the Sirius, and the Great Western made their historic passages across the Atlantic. The Royal William was the first to be divided into watertight bulkheads, of which she had four. The beginning of the Transatlantic Companies dates from 1840, when the Cunard vessels began to carry the mails between Liverpool, Halifax, and Boston, backed up by a subsidy of £60,000 a year, and up to the present time this line has never lost a passenger's life. The first vessels of this fleet were of wood, and propelled by engines of 750 HP., the boilers burnt 37 tons of coal a day, the ships occupied from 13 to 15 days in the passage, and carried 115 first-class passengers only. Eight years elapsed before any steamship carried steerage passengers, or before any rivals ventured to compete with the Cunard vessels. In ten years the fleet had grown both in numbers and size, but paddles had not as yet been supplanted by screw propellers. The Inman Line now came on the Atlantic with the City of Glasgow, the first iron vessel which proved successful. For the Great Britain, though built of iron, and driven by a screw, and the longest vessel then in existence, 1843, was wrecked after having made but two voyages. The Cunard Co. built their first iron steamer, the Persia, in 1855. During that period of fifteen years, the power for propulsion had grown from 750 HP. on the Britannia to 3,600 on the Persia, and the consumption of coal per day had risen from 37 tons to 160 tons. The Cunard Co. adopted the screw propeller first on the China (1862). The last paddle steamer of this line, the Scotia, was built in this year also.

The period between 1850 and 1860 was a remarkable epoch, rendered so by the building of the Great Eastern (1858), only eighteen years after the Britannia began running. Forty-one years passed before the length of the Great Eastern was exceeded, by the second Oceanic. The vessel was before her due time. The engine-power was not sufficient, and her coal consumption was exorbitant, burning fully two and a half times the amount that is burnt on a