mercury contained in a siphon tube, and the difference of the levels of the two surfaces of the mercury would thus become a measure of the saltness of the water in the boiler. A selfregistering instrument founded on this principle formed part of the self-registering steam-log which I proposed to introduce into steam-vessels some time since.

(211.) The Messrs. Seaward of Limehouse have adopted, in some of their recently constructed engines, a method of indicating the saltness of the water, and of measuring the quantity of salted water or brine discharged, by blowing out. A glassgauge, similar in form to that already described in land engines (156.), is provided to indicate the position of the surface of the water in the boiler. In this gauge two hydrometer balls are provided, the weight of which in proportion to their magnitude is such that they would both sink to the bottom in a solution of salt of the same strength as common sea-water. When the quantity of salt exceeds

parts of the whole weight of the water, the lighter of the two balls will float to the top; and when the strength is further increased until the proportion of salt exceeds parts of the whole, then the heavier ball will float to the top. The actual quantity of salt held in solution by sea-water in its ordinary state is part of its whole weight; and when by evaporation the proportion of salt in solution has become parts of the whole, then a deposition of salt commences. With an indicator such as that above described, the ascent of the lighter hydrometer ball gives notice of the necessity for blowing out, and the ascent of the heavier may be considered as indicating the approach of an injurious state of saltness in the boiler.

The ordinary method of blowing out the salted water from a boiler is by a pipe having a cock in it leading from the boiler through the bottom of the ship, or at a point low down at its side. Whenever the engineer considers that the water in the boiler has become so salted that the process of blowing out should commence, he opens the cock communicating by this pipe with the sea, and suffers an indefinite and uncertain quantity of water to escape. In this way he discharges, according to the magnitude of the boiler, from two to six tons

of water, and repeats this at intervals of from two to four hours, as he may consider to be sufficient. If, by observing this process, he prevents the boiler from getting incrusted during the voyage, he considers his duty to be effectually discharged, forgetting that he may have blown out many times more water than is necessary for the preservation of the boiler, and thereby produced a corresponding and unnecessary waste of fuel. In order to limit the quantity of water discharged, Messrs. Seaward have adopted the following method. In fig. 125. is represented a transverse

section of a part of a steam-vessel; w is the water-line of the boiler, B is the mouth of a blow-off pipe, placed near the bottom of the boiler. This pipe rises to A, and turning in the horizontal direction, A c is conducted to a tank T, which contains exactly a ton of water. This pipe communicates with the tank by a cock D, governed by a lever H. When this lever is moved to D', the cock D is open, and when it is moved to K, the cock D is closed. From the same tank there proceeds another pipe E, which issues from the side of the

vessel into the sea governed by a cock F, which is likewise put in connection with the lever н, so that it shall be opened when the lever н is drawn to the position F', the cock D' being closed in all positions of the lever between K and F'. Thus, whenever the cock F communicating with the sea is open, the cock D communicating with the boiler is closed, and vice versâ, both cocks being closed when the lever is in the intermediate position K. By this arrangement the boiler cannot, by any neglect in blowing off, be left in communication with the sea, nor can more than a ton of water be discharged except by the immediate act of the engineer. The injurious consequences are thus prevented which sometimes ensue when the blow-off cocks are left open by any neglect on the part of the engineer. When it is necessary to blow off, the engineer moves the lever H, to the position D'. steam in the boiler on the surface the salted water or brine up the pipe open cock c into the tank, and this continues until the tank is filled when that takes place, the lever is moved from the position D' to the position F', by which the cock D is closed, and the cock F opened. The water in the tank flows through the pipe E into the sea, air being admitted through the valve v, placed at the top of the tank, opening inwards. A second ton of brine is discharged by moving the lever back to the position D', and subsequently returning it to the position r′; and in this way the brine is discharged ton by ton, until the supply of water from the feed which replaces it has caused both the balls in the indicator to sink to the bottom.

The pressure of the

of the water w forces BA, and through the

(212.) A different method of preserving the requisite freshness of the water in the boiler has been adopted by Messrs. Maudslay and Field, and introduced with success into the Great Western and other steam-vessels. Pumps called brine-pumps are put into communication with the lower part of the boiler, and so constructed as to draw the brine therefrom, and drive it into the sea. These brine-pumps are worked by the engine, and their operation is constant. The feed-pumps are likewise worked by the engine, and they bear such a proportion to the brine-pumps that the quantity of salt discharged in a given time in the brine is equal to the quantity of salt

introduced in solution by the water of the feed-pumps. By this means the same actual quantity of salt is constantly maintained in the boiler, and consequently the strength of the solution remains invariable. If the brine discharged by the brine-pumps contains parts of salt while the water introduced by the feed-pumps contains only part, then it is evident that five cubic feet of the feeding water will contain no more salt than is contained in one cubic foot of brine. Under such circumstances the brine-pumps would be so constructed as to discharge of the water introduced by the feed-pumps, so that of all the water introduced into the boiler would be evaporated, and rendered available for working the engine.

To save the heat of the brine, a method has been adopted in the marine engines constructed by Messrs. Maudslay and Field similar to one which has been long practised in steamboilers, and in various apparatus for the warming of buildings. The current of heated brine is conducted from the boiler through a tube which is contained in another, through which the feed is introduced. The warm current of brine, therefore, as it passes out, imparts a considerable portion of its heat to the cold feed which comes in; and it is found that by this expedient the brine discharged into the sea may be reduced to a temperature of about 100°.

This expedient is so effectual that when the apparatus is properly constructed, and kept in a state of efficiency, it may be regarded as nearly a perfect preventive against the incrustation, and the deposition of salt in the boilers, and is not attended with any considerable waste of fuel.

(213.) About the year 1776, Mr. Watt invented a tubular condenser, with a view to condense the steam drawn off from the cylinder without the process of injection. This apparatus consisted of a number of small tubes connecting the top and bottom of the condenser, arranged in a manner not very different from that of the tubes which traverse the boiler of a locomotive engine. These tubes were continually surrounded by cold water, and the steam, as it escaped from the cylinder passing through them, was condensed by their cold surfaces, and collected in the form of water in a reservoir below, from

whence it was drawn off by a pump in the same manner as in engines which condensed by injection. One of the advantages proposed by this expedient was, that no atmospheric air would be introduced into the condenser, as is always the case when condensation by injection is practised. Cold water, which is injected, has always combined with it more or less common air. When this water is mixed with the condensed steam, the elevation of its temperature disengages the air combined with it, and this air circulating to the cylinder, vitiates the vacuum. One of the purposes for which the air-pump in condensing steam-engines was provided, and from which it took its name, was to draw off this air. If, however, a tubular condenser could be made to act with the necessary efficiency, no injection water would be introduced for condensation, and the pump would have no other duty except to remove the small quantity of water produced by the condensed steam. That water being subsequently carried back to the boiler by the feed-pumps, a constant system of circulation would be maintained, and the boiler would never require any fresh supply of water, except what might be necessary to make good the waste by leakage and other causes.

This contrivance has been of late years revived by Mr. Samuel Hall of Basford, near Nottingham, with a view to supersede in marine engines the necessity of using sea-water in the boilers. Mr. Hall proposes to make marine boilers with fresh water to condense the steam without injection, by a tubulated condenser, and to provide by the distillation of sea-water the small quantity of fresh water which would be necessary to make good the waste. These condensers have been introduced into several steam-vessels: in some they have been continued, and in others abandoned, and various opinions are entertained of their efficacy. I have not been able to obtain the results of any satisfactory experiments on them, and cannot therefore form a judgment of their usefulness. Mr. Watt abandoned these condensers from finding that the condensation of the steam was not sufficiently sudden, and that consequently at the commencement of the stroke the piston was subject to a resistance which inju