The following text may have been generated by Optical Character Recognition, with varying degrees of accuracy. Reader beware!

position are caused by the insufficient adherence of the cement to the rough and awkward sides of large sawdust particles. The addition of gelatinous bodies, such as precipitated mag- nesium silicate, has been tried and even patented, under the supposition that they would close up the pores, but they only make the mixture hard to lay down and are not very effective. The impregnation of the composi- tion with resinous bodies, after it has been laid down and allowed to dry, has been tried; but as these do not penetrate very far the surface is soon worn 'off; leaving the oe as por- ous as before: 'Of late years a number. of sabe stances have been introduced which are laid on the deck itself before the composition is laid down, and which prevent the magnesium chloride solu- tion, even when it does leak through, from getting at and corroding the deck. One method employed is to apply a thick coat of a hot bituminous mixture to the deck, and, while it is still -soft, to strew its surface with granite chips. The sheathing composi- tion is then laid on this, and is keyed onto the smooth surface of the bitu- men by the pieces of granite. Another method consists in the application of a substance which is painted on the aeck and which, when attacked by magnesium chloride, forms a hard, in- soluble compound. The idea is that the greater the quantity of magnesium chloride which leaks through onto such a coating, the harder and more resisting will it become. The writer's experience leads him to believe that such coatings are ef- fective, provided that they are some- what elastic and can be applied in thin layers. It will be-readily under- stood that a thick layer of a protective coat laid between two bodies of wide- ly differing physical properties, such as steel and magnesia cement, is bound to 'be-subjected to great strains, due to the difference in the elasticity and the co-efficients of expansion of the two bodies. In one ship examined by the writer the protective layer was found ground to powder after having _ beén laid only nine months. Oxychloride Compositions Another drawback to the use of oxychloride compositions is caused by the fact that, process and even for some time after, they expand considerably. As the sub- stance is firmly held to the deck by its own adhesion, and usually also by special "cleats", it follows that a con- siderable strain is set up in the ma- terial. If, for any reason, the adhesion to the deck is not good, this strain results in the lifting and cracking of 486 | during the hardening THE MARINE REVIEW the composition--circumstances which kelp the admission of water, and con- sequent corrosion of the deck, very greatly. Some firms adopt measures tc minimize the danger of expansion, but these, in all fairness to the firms in question, cannot be dealt with here. There can be no doubt that some compositions as laid down are unsatis- factory and even dangerous; but, given rigid adherence to certain conditions, they are as safe as, or even safer than, wood or other deck sheathings. A few of these conditions are:. (a) The limiting of the use of oxychloride compositions to floors not exposed to the weather. (b) The careful supervision and preparation of the binding mixture. (c) The use of the proper quality and texture of sawdust. (d) The use of an effective anti- (A) (a) Deck sheathing composition Steel deck (c) Nani instrument Movable point Ce)" Battery De The Dotted Lines Show hee Path of the Cur- rent in the Composition c corrosive' mixture between the deck and the composition. It 18.08. im- portance that this mixture should ad- here well to the deck, and that the composition should adhere well to the mixture. (e) The use of a auteur pane of cleats of ample area to hold the composition down. (f) The~ careful scrutiny of the anticorrosive coat for flaws and non- adhering patches. No coating should be used which will not stand a con- siderable amount of foot traffic over it without coming off. Some of the substances laid down at present have the appearance and wearing properties of whitewash, and they are almost useless by the time the sheathing composition is laid on top of them. If the Board of Trade atid the -in- surance corporations, instead of look- ing on these substances with a dubious eye, were to issue two or three definite regulations as to their manufacture and present simple tests which sam- _the current registered will be December, 1914 a ples of the material actually being laid down must pass, the unpleasant feeling of uncertainty concerning them would soon be a thing of the past. It is often of great importance to know whether a composition when laid is actually holding to the deck, and the writer has found the following method to afford ready means of tell- ing whether this is so. Provided that the composition is mixed evenly and is of uniform thick- ness, the layer applied to the deck offers a fairly high and uniform resist- ance to the passage of an electric' current. If, therefore, one pole of a steady source of current be earthed, that is, connected to the steel deck, and the other pole be connected to a small metal disc which is laid on the top of the sheathing composition, the amount of current which passes through the composition at any part of the deck will be constant, and can be measured on an ammeter or other suitable instrument included in the eiremit. The 'diagram. (A) im Fig. tT; it. which the thickness of the steel deck and the composition are purpose- ly exaggerated, shows the necessary connections, and the dotted line the path of the current through the com- position. If the composition is not adhering to the deck, the current will be com- pelled to take the path shown by the dotted lines in diagram '(B); and. as it has much further to travel through the composition, the resistance to 'its passage will be greater, and hence less.' By moving the small disc about on the cleaned surface of the composition, therefore, and watching the measuring itistrument, it is quite easy to tell where the composition is adhering and::where this is not the case. In practice it is convenient to use a megger and bridge ohmmeter' to measure the resistance of the sheath- ing composition. Instead of a disc, a. heavily weighted brass ring, sup- ported on three short feet with nat- row ends (about % inch in diameter) is used to make contact on top of the composition. An arrangement of this kind is essential to ensure that the actual area of contact is the same at each measurement and is not altered by slight local unevenness 'of the sur- faces Non-Conducting Substances Needless to say, this method is of no value when highly non-conducting substances, such as bitumen, are used as a protective layer. The compositions of the second class which employ bituminous materials to bind together the sawdust instead of the oxychloride cement, have cer-