is no undoing of work. We knew an instance in which a machine was designed for bending bars, the preliminary part being to straighten them before the curve was put in; when the bar came out of the machine it was curved, but if not to the required curve, the operation could not be repeated without first straightening the bar again, before any additional curving could be put in. This was a case in which the action was not continuous; the first part of the machine undid what had been done by the latter part. Hence the machine, when made, turned out a complete failure, for want of being designed in accordance with correct principle.

Position.-As soon as we are satisfied that the principle is correct, the important point of position comes to the front. To some extent it may be included in principle, which we have defined to be "taking care that no natural law is violated," for the foundation of position is due to the natural law or axiom that. no two things can be in the same place at the same time. This is more important to bear in mind when designing moving machinery. A drawing only shows one position of a moving part; hence, unless there is great care on the part of the draughtsman in following out the motion of each part to its full extent, he is liable to have some parts foul one another when they have moved into a new position not shown on the drawing. To some extent

this may be avoided by making diagrains showing successive phases of the moving parts, but this involves a great deal of work, and is often neglected. It is desirable, when possible, to show the parts in the worst possible position in relation to each other. Very frequently the extreme phase of one part does not coincide with the extreme phase of another, as in the steam engine the valve is not at full stroke at the same time as the piston. To get over this difficulty, it is very customary to make a working drawing disjointed-that is to say, the various parts are drawn as if they were all in their worst position at once, it being understood that this is only a convention, in order that the various parts may be in their proper position in all phases of the machine, for if everything clears in its worst positions it naturally clears in every other position. In spite of all drawings, however, errors will arise, and the constant care of the draughtsman should always be directed to the point technically called clearance.

It should also be borne in mind that not only have the parts to work when they are together, but that they have to be put together; there should be no collars, for instance, on a shaft to prevent a wheel being slipped endways. Putting work together is naturally an extension of position in relation to moving parts, for the parts are moved, although but seldom, viz., when the

machine is put together or pulled to pieces. Whilst considering this point it is well to mention bolts and nuts; the designer should take care that they are arranged where they can be put in, and where the nuts can be put on and taken off. This is a small point, but is one frequently overlooked by draughtsmen, who often show work as if the bolts and other minor parts had grown, it being impossible to get the work together as shown on their drawing.

Strength. After principle and position, strength may be considered the next factor. This is a matter primarily of calculation, and secondly of precedent, it being well known that if a part of a machine was strong enough in some other machine, or under similar circumstances, it is strong enough for the case under consideration. The strength of materials, and parts of machines, would require a volume to itself, and it is not our intention to dwell upon it. It is important that the designer should be a good calculator, well up in theory, and also that he should have had a good deal of practice, or, in other words, a good store of precedents, which is only theory embodied in practice. Particular care should be taken that no sudden changes occur in the dimensions of bars, axles, and other parts, and also that there should be no internal sharp corners to any of the castings. These are great pitfalls to the unwary, and care should be taken in avoiding them.

Proportions.-The vital points of principle, position and strength being supposed to be settled, there remain what may be considered the luxurious points of pleasing proportions and of ornamentation. Elegance of proportion is more a matter of fancy than anything else, and can only be acquired by practice. There are generally various ways of drawing and arranging the parts of a machine without violating any of the vital points we have already dwelt upon. The draughtsman will do well to think to himself as he goes along: Does this look well? Will it look better in any other way?

Ornamentation in machinery is generally the addition of a few mouldings, scrolls, and such like, which, as in proportion, is a matter of taste or feeling. Care should be taken to avoid contradiction in ornament; there should be a similarity of style about the mouldings, the scrolls, and the shapes of the subordinate parts of the machinery.

It must be understood that all the points into which this article is divided must be considered by the designer simultaneously; they cannot be separated when designing, as we have done here for the convenience of treatment, and in relation to the importance of the divisions.

FOAMING IN BOILERS.

The causes are dirty water; trying to evaporate more water than the size and construction of

the boiler is intended for; taking the steam too low down; insufficient steam room; imperfect construction of boiler, and too small a steam pipe.

Take a kettle of dirty water and place it on a fire and allow it to boil and watch it foam, and it will be the same in a boiler.

Too little attention is paid to boilers with regard to their evaporating power. Where the boiler is large enough for the water to circulate, and there is surface enough to give off the steam, foaming never occurs. As the particles of steam have to escape to the surface o. the water in the boiler, unless that is in proportion to the amount of steam to be generated, it will be delivered with such violence that the water will be mixed with it and cause what is called foaming.

A high pressure insures tranquillity at the surface, and the steam itself being more dense it comes away in a more compact form, and the ebullition at the surface is no greater than at a lower pressure. When a boiler foams we close the throttle to check the flow, and that keeps up the pressure and lessens the sudden delivery.

Too many flues in a boiler obstruct the passage of the steam from the lower part of the boiler on its was to the surface; this is a fault in construction, but nearly all foaming arises from dirty water, or from trying to evaporate too much water without heating surface or steam room