must be considered in fractional distillation for testing creosote samples.

The specific gravity of the tar usually varies between 1.10 and 1.20, though that from Scotch cannel coal is sometimes as low as .95. The specific gravity of the fractions does not increase regularly with the increasing temperature of the distillation, though the tendency is in that direction.

PRODUCTS OF NEWCASTLE COAL WHEN CARBONIZED FOR THE MANUFACTURE OF COKE

The principal products which can be separated as they come from the still by distillation and filtration :

The creosote oil. The oldest specimens of treated timber in the waters of the Gulf of Mexico were treated with creosote oil imported from England. It contained a high percentage of solid naphthalene, as evidenced by the solid contents of barrels in which it was shipped. It is probable that the amount of naphthalene present in the present day creosote is not so high as formerly, but records of tests by distillation of these old oils are, no doubt, of little value to us to-day, as the methods of sampling and analysis are not now the same.

Creosote is being used in the preservation of timber from two radically different causes of destruction, and by all means they should be considered independently-the decay of timber caused *Temperature by thermometer of still above and close to the surface of the

tar.

by the vegetable organisms, the bacteria and fungus on one hand and on the other the animal organisms, such as the teredo and limnoria. It has been ascertained that antiseptics prevent the vegetable decay, though the knowledge on that subject is rather limited with regard to creosoting, and experiments with the different ingredients of creosote seem to show that the phenols and crysols, though they are of the greatest antiseptic value, do not preserve timber as well as some of the other portions.

Creosote certainly is not poisonous to any great extent to the teredo. Instances of attack of well-treated wood, in which a fullgrown teredo passes from untreated wood and through treated wood, are frequently seen. They evidently do not bore for food, but for a habitation, the débris carried out by the siphon probably not having any marked effect on the borer itself. It is reasonably certain that creosote prevents attack to a large extent by filling up the pores of the wood and thus preventing the lodgment of the teredo in the first place. This protection would then be mechanical. Creosote is no doubt odious to the teredo, for on striking a creosoted strata they usually quickly turn from it back into the untreated wood, and if that is lacking they will soon quit boring and die. It is possible that the action of the creosote is entirely mechanical, inasmuch as impregnated wood, having the pores completely filled, is in such a condition that the teredo can not bore in it for any great length of time. In any event, protection of marine work and prevention of decay are different problems and should be studied as such. A case on this point recently came to the notice of the writer through a treatment made with large percentage of water in the creosote. After the evaporation of the water the wood was left dry and porous, but still contained much creosote resisting decay but failing on exposure to marine borers.

From the evidence at hand it seems that a creosote having a high percentage of solids at the ordinary temperature would resist destruction in sea water longer than the more liquid creosotes. In all of the cases of well protected piling in long use on the Gulf Coast the creosote used had a high percentage of naphthalene, so much that when the barrel in which it was imported fell to pieces the oil was solid enough to stand without support. In the absence of better information should we not conclude that such an oil is the proper one to use at the present time in similar positions? The anthracene is less volatile than the naphthalene fractions and should stay in the piling longer on exposure. There is quite

a loss of creosote from the outer portion of pile, caused by the wash of the waves, and this loss is just at the point of greatest activity of the teredo. As stated previously, the loss from the finer grained woods is less than from the coarse grained, since the effect of capillarity in the smaller cells tends to retain the creosote. If the creosote were solid the capillary action would not be active and, consequently, the oil would not come to the surface of the wood to be washed out.

On account of the solidification of the oils and also the separation into strata of the different oils, the oil in the tank should be kept hot and liquid. After some months without heat or stirring it will be found that the portion on the bottom of the tank is hard and solid-in some cases so hard and solid that it must be broken out with a pick. Table A shows the variation of distillation of sample taken from various depths in a large storage tank.

TABLE A.-Comparative analysis, showing the separation of Creosote Oil in a Storage Tank.

A. Analysis at time tank was filled, oil thoroughly mixed.

B. Four months later, 6 feet below surface.

C. Four months later, 20 feet below surface.

D. Adhering to sides of tank near top after tank was partly emptied.

Not able to obtain sample of hard oil in bottom of tank. Analysis by same person and same apparatus. 500 c. c. Side neck distilling flask with thermometer at the outlet tube. Distillation at the rate of 1 to 3 drops per second.

Most storage tanks are open at the top, and, for prevention of fire, the top is kept covered with a few inches of water to which is added from time to time the quantity caught by rains, usually more than the portion lost by evaporation. Ordinarily water does not mix with the creosote, owing to the higher specific gravity of the latter, but it will be found that after some months the water from the top of the tank has gone down into the creosote for quite a distance, so that water will be drawn off with the upper portion

of the oil. This proportion of water is rather difficult to separate from the creosote, as ordinarily it will not settle on top to be drawn off. Probably after several months time there has been formed an emulsion which tends to keep the mixture stable.

The presence of large percentages of water in the creosote will probably account for more failures in marine work than any one thing. At nearly every step in the treatment process the tendency is to increase the percentage of water, and most people connected with the business have such unbounded faith in the separation by settling that they neglect to examine the oil to see if it really does separate. Clean water will separate, and does so very rapidly, but water charged with saps and resins, mixed with the oil hot and stirred by being pumped back and forth betweeen tanks and cylinders, will not separate unless some special means are employed. Usually, the plant has a chemist employed to ascertain the condition of the creosote, but very often he does not keep up with the changes closely enough and the result is that a treatment is made with a large quantity of water in the creosote. It is not unusual to find

10 per cent, and the writer has found more than 20 per cent on several occasions.

The presence of water does not necessarily render the treatment valueless, but aside from the fact that the quantity of oil specified is not injected, the treatment approaches the open cell process, the value of which as a protection has not been fully determined in all cases of exposure. Very recently two cases of failure of creosote as a protection from teredo caused by a high content of water came within the notice of the writer. The first one was of piling treated some twenty-five years ago, which within a few years, were attacked and would have been destroyed had not the additional protection of vitrified pipe been applied. The writer saw some of these piling removed and sawed recently, and the section showed such little signs of creosote that he did not know at the time that they had ever been creosoted. The wood was scarcely discolored. The creosote must have contained fully 50 per cent water at the time of injection.

In the second case the piling, after treatment, did not show the presence of creosote to any marked degree. The sections after months of exposure showed a light brown color, and after two years were attacked and practically destroyed by teredo. Instances of this sort caused primarily by lack of proper inspection cause the failures, and the wrong interpretation of these failures has tended

to make engineers doubtful to a great extent of the efficiency of the protection afforded by creosote.

Speaking from a practical standpoint, it is more important to ascertain that the creosote used is creosote than to spend time and money buying a creosote from some certain place containing a stated proportion of the various elements. The light oils so often prohibited in specifications are more valuable than water, and yet in practice under these same specifications the water is often injected.

Many engineers have an idea that creosoted material should appear black and slick. Usually, these same engineers limit the amount of free carbon in the creosote and would offer strenuous objections to the addition of any quantity of tar or pitch; yet it is the free carbon, tar, and pitch which makes the timber black. Commercial creosote, which has not long been exposed to the sun and air and has not been used in such a manner to become oxidized, will not leave the timber "black and gummy," as so often wished for, especially when applied to dressed surfaces. The color and grain of the timber will often he apparent and yet it is not detrimental to the timber and is not indicative of faulty treatment. If pure dead oil of coal tar, free from adulterations" is specified, one should not expect to find the timber "black and gummy" on its being withdrawn from the cylinder. It may become so when treated with commercial oil and exposed to sun and water.

The question of the value of naphthalene as a preservative from both the effects of decay and attacks of marine life is certainly an open one. Of late years many have questioned the value formerly placed on it, and no doubt there has been too much reliance placed on it, especially in marine work. It is true that the examples of well protected marine timbers we have, especially on the Gulf Coast, all originally contained large percentages of naphthalene, but certainly most of it has evaporated and at present only traces are found. It is possible that the very loss of it has been the means of protection, in that its action prevented attack by marine borers, yet it is certain that these timbers have long been without any large quantity and yet have been preserved. There will be found, after years of exposure, much of the anthracene portion of the creosote, due to its higher distillation point and consequent slower evaporation. Indications therefore point to that portion of the creosote as the most valuable, which opinion seems to be increasing in popularity.