Cottonseed-oil was the only drying-oil that showed a marked exception; the fatty acid from it exhibited a very slight increase of weight. That is probably the reason why this oil can play a double role, as a drying and a non-drying oil, for it is used to adulterate linseed-oil on the one hand and olive-oil on the other. Contact with precipitated lead imparts to oil the property of absorbing oxygen rapidly. In his study of the oxidation of the oils, Cloez has shown that it is always attended with the total disappearance of the glycerine, and in Livache's experiments it was noticed that the glycerine was modified by the precipitated lead. If glycerine is mixed with precipitated lead in a tight bottle free from air, the lead soon vanishes, being oxidized at the expense of a portion of the glycerine, and then dissolved in it. These facts indicate that a rapidly drying oil can be obtained by simply treating linseed-oil for some time with red-lead or litharge, although the product thus obtained always remains greasy, and does not dry as well and quickly as boiled linseed-oil. In the arts, advantage may be taken of this action of lead toward drying-oils, as for example to prove the presence of cottonseed-oil in linseed-oil as well as in olive-oil. Probably boiling may be dispensed with by substituting mere contact of the oils with precipitated lead, or solutions of lead and strips of zinc on which the lead may be deposited in a fine state of division.

Oils prepared in this way are always of a lighter colour, and retain a greater degree of fluidity. Perhaps the bad smell of boiling-oils, and the great danger of their taking fire in the operation, can be avoided by this treatment. (Oil, Paint, and Drug Rep.)

All the mineral oils -and also sperm-, lard- and neatsfoot-oils-appear to reach a nearly uniform coefficient of friction at very greatly different degrees of heat in the bearings. Several kinds of the best mineral oils, and sperm- and lard-oils, show a uniform coefficient of friction at the following degrees of heat:-

Temperature at which the coefficient of friction is the same.

Lubrication seems to be effective in inverse ratio to viscosity-that is, the most fluid oil that will stay in its place fa the best to use. Lard-oil heated to 130°; F. (54 1/2° C.) lubricates as well as sperm at 70° F. (21° C), or tl)e best mineral oil at 50° F. (10° C). But of course it is a great waste of machinery to work oil of any kind up to an excessive heat; and there must be the least wear in the use of oil that shows the least coefficient of friction at the lowest degree of heat.

- The quantity of oil used is a matter of much less importance than the quality. The mill that saves gallons of oil at the cost of tons of coal, or dollars of repairs, plays a losing game. Waite's experiments on Very heavy bearings at Manchester go far to prove that a 'considerable quantity of thin fine oil keeps the bearings much cooler, and required loss power, than a smaller quantity of thick viscous oil. Here let it be observed that a superstition which prevails in favour of using castor-oil to cool a hot bearing is without any warrant. No vegetable oil is fit to use as a lubricant; and castor-oil is the worst of all, because the most viscous. If used, it will surely set the mill on fire, as it did in the only case of which we have a record.

Cast iron holds oil better than any other metal or any alloy, and is the best metal to use for light bearings, perhaps for heavy.

It has been proved by Waite's experiments that a highly polished bearing is more liable to friction than a surface finery lined by filing. The lines left by the file serve as reservoirs for the oil, while the high polish leaves no room for the particles between the metal surfaces.

So far as laboratory experiments may serve as a guide in practice, it therefore appears that fine mineral oils may be made to serve all the purposes of a cotton-mill, and such is the practice in some of the mills that show the very best results in point of economy.

Next, that the best animal oil to mix with a fine mineral oil, in order to give it more body, Is sperm-oil; this, again, accords with the practice of many of the mills in which the greatest economy is attained.

Lard- and neatsfoot-oil are used to give body to mineral oil in some of the best mills; but the results of work seem not to warrant this practice, unless there is some peculiarity in the machinery that makes it more difficult to keep a less viscous or tenacious oil on the bearings.

All the mixed oils sold under fancy names we believe must of necessity consist of certain proportions of the oils, heretofore named, as none of the vegetable or fish oils is fit to be used, and there are no other animal oils that can be had in any quantity.

. It appears that all varieties of mineral oils are or have been used in print cloth-mills, and are all removed in the process of bleaching, as practised in printworks.

All mineral oils stain more or less, and give more or less difficulty to the bleacher when dropped upon thick cloth, or cloth of a close texture. On this point we have been able to establish no positive rule; but as very many kinds are and have been used in mills working on such cloths, and are removed, we are inclined to the belief that this question is not of as great importance as it has been assumed to be. (Scient. Amer.)

A writer in Iron says:- One . of the first effects of friction is to engender heat. It is a singular circumstance that whereas lubricants are intended to prevent friction, many of them are capable, under certain circumstances, of exciting a degree of heat which is fraught with danger. This action takes place, not when the lubricants are in contact with metallic surfaces, but when they are absorbed into certain fibrous or cellular bodies of a vegetable nature. Wood, in the form of sawdust, when impregnated to a moderate extent with one or other of a certain class of oils, is capable of developing spontaneous combustion. The action is still more rapid when the material consists of cotton rags to which the dangerous class of oil has been applied. The phenomenon is evidently one of oxidation, for in the case of sawdust, if the saturation be very complete, combustion does not take place, whereas if the soaking be slight, there is a probability that combustion will ensue.