dampers, by means of which the furnace of each boiler can be regulated independently of the other, and by which each boiler may be separately detached from communication with the chimney. The letters of reference in the horizontal section, fig. 121., correspond with those in the transverse vertical section, fig. 122., E representing the commencement of the flues, and c their termination.

A longitudinal section of the boiler made by a vertical plane extending from the front to the back is given in fig. 123.,

Fig. 124

F

where F, as before, is the furnace, & the grate-bars sloping downwards from the front to the back, H the fire-bridge, c the commencement of the flues, and a the chimney. An elevation of the front of the boiler is represented in fig. 124., showing two of the fire-doors closed, and the other two removed, displaying the position of the grate-bars in front. Small openings are also provided, closed by proper doors, by which access can be had to the under side of the flues between the foundation timbers of the engine for the purpose of cleaning them.

Each of these boilers can be worked independently of the others. By this means, when at sea, the engine may be worked by any two of the three boilers, while the third is being cleaned and put in order. In all sea-going steamers multiple boilers are at present provided for this purpose.

In the boilers here represented the flues are all upon the same level, winding backwards and forwards without passing one above the other. In other boilers, however, the flues,

GG

after passing backwards and forwards near the bottom of the boiler, turn upwards and pass backwards and forwards through a level of the water nearer its surface, finally terminating in the chimney. More heating surface is thus obtained with the same capacity of boiler.

The most formidable difficulty which has been encountered in the application of the steam-engine to sea-voyages has arisen from the necessity of supplying the boiler with seawater instead of pure fresh water. The sea-water is injected into the condenser for the purpose of condensing the steam, and it is thence, mixed with the condensed steam, conducted as feeding water into the boiler.

(209.) Sea-water holds, as is well known, certain alkaline substances in solution, the principal of which is muriate of soda, or common salt. Ten thousand grains of pure sea-water contain two hundred and twenty grains of common salt, the remaining ingredients being thirty-three grains of sulphate of soda, forty-two grains of muriate of magnesia, and eight grains of muriate of lime. The heat which converts pure water into steam does not at the same time evaporate those salts which the water holds in solution. As a consequence it follows, that as the evaporation in the boiler is continued, the salt, which was held in solution by the water which has been evaporated, remains in the boiler, and enters into solution with the water remaining in it. The quantity of salt contained in sea-water being considerably less than that which water is capable of holding in solution, the process of evaporation for some time is attended with no other effect than to render the water in the boiler a stronger solution of salt. If, however, this process be continued, the quantity of salt retained in the boiler having constantly an increasing proportion to the quantity of water, it must at length render the water in the boiler a saturated solution- that is, a solution containing as much salt as at the actual temperature it is capable of holding in solution. If, therefore, the evaporation be continued beyond this point, the salt disengaged from the water evaporated instead of entering into solution with the water remaining in the boiler will be precipitated in the form of sediment; and if the process be continued in the

same manner, the boiler would at length become a mere salt-pan.

But besides the deposition of salt sediment in a loose form, some of the constituents of sea-water having an attraction for the iron of the boiler, collect upon it in a scale or crust in the same manner as earthy matters held in solution by spring-water are observed to form and become incrusted on the inner surface of land-boilers and of common culinary vessels.

The coating of the inner surface of a boiler by incrustation and the collection of salt sediment in its lower parts, are attended with effects highly injurious to the materials of the boiler. The crust and sediment thus formed within the boiler are almost non-conductors of heat, and placed, as they are, between the water contained in the boiler and the metallic plates which form it, they obstruct the passage of heat from the outer surface of the plates in contact with the fire to the water. The heat, therefore, accumulating in the boiler-plates so as to give them a much higher temperature than the water within the boiler, has the effect of softening them, and by the unequal temperature which will thus be imparted to the lower plates which are incrusted, compared with the higher parts which may not be so, an unequal expansion is produced, by which the joints and seams of the boiler are loosened and opened, and leaks produced.

These injurious effects can only be prevented by either of two methods; first, by so regulating the feed of the boiler that the water it contains shall not be suffered to reach the point of saturation, but shall be so limited in its degree of saltness that no injurious incrustation or deposit shall be formed; secondly, by the adoption of some method by which the boiler may be worked with fresh water. This end can only be attained by condensing the steam by a jet of fresh water, and working the boiler continually by the same water, since a supply of fresh water sufficient for a boiler worked in the ordinary way could never be commanded

at sea.

(210.) The method by which the saltness of the water in the boiler is most commonly prevented from exceeding a certain

limit has been to discharge from the boiler into the sea a certain quantity of over-salted water, and to supply its place by sea-water introduced into the condenser through the injection-cock for the purpose of condensing the steam, this water being mixed with the steam so condensed, and being, therefore, a weaker solution of salt than common sea-water. To effect this, cocks called blow-off cocks, are usually placed in the lower parts of the boiler, where the over-salted, and therefore heavier, parts of the water collect. The pressure of the steam and incumbent weight of the water in the boiler force the lower strata of water out through these cocks; and this process, called blowing out, is, or ought to be, practised at such intervals as will prevent the water from becoming over salted. When the salted water has been blown out in this manner, the level of the water in the boiler is restored by a feed of corresponding quantity.

This process of blowing out, on the due and regular observance of which the preservation and efficiency of the boiler mainly depend, is too often left at the discretion of the engineer, who is, in most cases, not even supplied with the proper means of ascertaining the extent to which the process should be carried. It is commonly required that the engineer should blow out a certain portion of the water in the boiler every two hours, restoring the level by a feed of equivalent amount; but it is evident that the sufficiency of the process founded on such a rule must mainly depend on the supposition that the evaporation proceeds always at the same rate, which is far from being the case with marine boilers. An indicator, by which the saltness of the water in the boiler would always be exhibited, ought to be provided, and the process of blowing out should be regulated by the indications of that instrument. To blow out more frequently than is necessary is attended with a waste of fuel; for hot water is thus discharged into the sea while cold water is introduced in its place, and consequently all the heat necessary to produce the difference of the temperatures of the water blown out and the feed introduced is lost. If, on the other hand, the process of blowing out be observed less frequently than is necessary, then more or less incrustation and deposit

may be produced, and the injurious effects already described

ensue.

As the specific gravity of water holding salt in solution is increased with every increase of the strength of the solution, any form of hydrometer capable of exhibiting a visible indication of the specific gravity of the water contained in the boiler would serve the purpose of an indicator, to show when the process of blowing out is necessary, and when it has been carried to a sufficient extent. The application of such instruments, however, would be attended with some practical difficulties in the case of sea-boilers.

The temperature at which a solution of salt boils under a given pressure varies considerably with the strength of the solution; the more concentrated the solution is, the higher will be its boiling temperature under the same pressure. A comparison, therefore, of a steam-gauge attached to the boiler, and a thermometer immersed in it, showing the pressure and the temperature, would always indicate the saltness of the water; and it would not be difficult so to graduate these instruments as to make them at once show the degree of saltness.

If the application of the thermometer be considered to be attended with practical difficulty, the difference of pressures under which the salt water of the boiler and fresh water of the same temperature boil, might be taken as an indication of the saltness of the water in the boiler, and it would not be difficult to construct upon this principle a self-registering instrument, which would not only indicate but record from hour to hour the degree of saltness of the water. A small vessel of distilled water being immersed in the water of the boiler would always have the temperature of that water, and the steam produced from it communicating with a steam-gauge, the pressure of such steam would be indicated by that gauge, while the pressure of the steam in the boiler under which pressure the salted water boils might be indicated by another gauge. The difference of the pressures indicated by the two gauges would thus become a test by which the saltness of the water in the boiler would be measured. The two pressures might be made to act on opposite ends of the same column of