part of the operation, three grades of benzene are running at the same time, and also during the "cracking" process, a portion of light product the result of the decomposition of the heavier oils may also be received through the medium of the air condensers.

Whatever form of condensing apparatus may be employed, two important features must receive due consideration. There must be no stint either in length or size of the condensing worm, or in the quantity of water supplied to the tank. The writer has observed that every addition made to the capacity of the condensing arrangement was attended with an improvement in the quality of the oil and an economy of time in employing the still. A six hundred barrel still will require at least a thousand feet of six inch tube, or its equivalent in piping of a smaller diameter.

The condensing tanks are usually built of iron, generally oblong, but sometimes also circular in form, in either case resting upon solid masonry of stone or brick-work; condensing worms are also frequently formed of cast-iron pipe, four to ten inches in diameter, in lengths of 6, 8, or 10 feet, flanged and bolted together. It has been claimed that they are more economical. This is questionable, and in many respects they are vastly inferior to the usual lapwelded wrought-iron pipe.

The supply of water is not second in importance to the capacity of the condensing worm; an abundance of water is especially needful in the summer season, and under no conditions should the distillate be warmer, upon issuing from the tail pipe, than the water supplied to the condenser. A

warm distillate, if in a large body, retains its heat for a long time and rapidly darkens in color. In this condition it is violently attacked in the acid treatment, and the product will generally be found to be "off color." This result, however, can be guarded against by cooling the oil before subjecting it to the acid treatment. It will be found a wise economy in the end to invest liberally in the condensing arrangements.

The terminal portions of the condensing pipes all converge and enter the receiving house within a few inches of each other. Near the extremity of each, a trap (Fig. 32) in the pipe is made, for the purpose of carrying away the

incondensable vapor. This vapor is burned underneath the boilers or stills, effecting thereby a large saving in fuel. The location of some refineries in certain large coal-distributing centres enables their proprietors to employ large quantities of fine refuse coal or coke by mixing the same with the cokey settlings of the residuum. Neither of these could be employed separately as a fuel with advantage, but mixed, as before stated, they form by far the larger proportion of the fuel used. The fire, at certain stages, or when these are not combined to the best advantage, may be urged by a blast of air.

DISTILLATION OF PETROLEUM.

A careful perusal of the chapter on the chemistry of this substance will greatly aid one in comprehending the general principles involved in this, the first process, in refining petroleum. That this peculiar and highly complex body is composed of a series of hydro-carbon compounds, varying in gravity from an incondensable gas to a solid substance, does not appear to admit of a doubt. How this peculiar compound is constituted, or its component parts are held together, is by no means so clear, and the question first propounded by Professor Silliman after his admirable analysis, whether the different distillates obtained by him, each differing from the others, were so many distinct products of the destructive distillation of the oil, or whether they were educts of varying compounds already pre-existing in the oil, has not yet been satisfactorily answered, at least not that I am aware of. Mr. Warren, of Boston, claims to have obtained products which maintain throughout the whole distillation equable boiling points. Granting this to be the case, and his experiments were too accurate to doubt their correctness, we would assume petroleum to be a mechanical mixture of a series of hydrocarbons of a definite composition, and capable of being separated from each other by careful distillation. There is one fact which is matter of common observation among oil men, which would tend to confirm such a view of its constitution. An oil tank containing a large quantity of distilled oil, if allowed to remain undisturbed for several

months, will be found to contain several grades of oil; the upper portion being several degrees lighter than the bottom; the heavier portion obeying the ordinary laws of gravity finding its proper level. It is not necessary for us either to assent to, or to deny, the correctness of this purely mechanical view of the subject, to explain the practical methods of dividing crude oil into marketable products.

The stills having been placed in proper position, and all the connections with the condensing apparatus having been carefully inspected, they are filled to their estimated capacity. To save time the crude-oil pump, especially, should be large. A twelve hundred barrel still should be filled in thirty minutes or less. The fires being kindled, the work of emptying now devolves upon the fireman and the distiller, and the varied experience which they acquire in their respective positions is of immense value. A gentle fire is started, and the distiller, whose position confines him exclusively to the receiving house, is notified by speaking tube or signal, that still No. is "fired up." The light products known under the generic name of benzene soon make their appearance. In this early stage of the process the diversity of methods employed by different refineries. manifests itself. A certain refiner, who may have a local demand for all the light product above a certain gravity, will allow the whole of his benzene run to flow into one benzene tank, from which it is barrelled and sold. Another refiner may have a demand for two or more grades of benzene for specified purposes. With this division of the product in view, he will of course provide himself with

two or more separate tanks, and direct his light distillate into each according to gravity.

The point at which the distiller "cuts off" his benzene and directs the product into the oil tanks, varies according to the kind of oil he manufactures. For an oil of low firetest, say 110°, some refiners cut off as high as 64° Beaumé scale; most generally, however, it is considered safer to continue the run" down to 62°. The benzene distillate between 70° and 62° contains quite a percentage of burning oil. The "run of the still" between these two points is, by some refiners, who are bent upon extracting the largest yield of burning oil, laid aside for separate distillation, the heavier portions being separated and run into the oil. To make an oil to stand 120° fire-test, it is never safe to cut off above 62°. An oil with a fire-test of 150° and "water-white" in color, known in the trade as "headlight," is now much in demand, commanding a higher price. In the manufacture of this variety, the lighter products of distillation are kept out until the Beaumé indicates 56°. In this process a large percentage of oil between 62° and 56° is obtained, which may be returned to the benzene tank, but is generally reserved for the manufacture of the 110° oil. Having disposed of the benzene "runnings" according to the grade of oil to be made, the burning oil is then directed into its special tank. The still-man now urges his still to its utmost, being guided in his directions to his fireman very much by the color of his distillate. The saving of time is all important in this part of the operation, in order to accomplish his run in a specified