Apparently little was known about spontaneous combustion in early times, and even to-day it is but slightly understood; yet there can be no doubt that many conflagrations whose origins still lack satisfactory explanation. might rightly be attributed to this cause. When we speak of spontaneous combustion we usually refer to that peculiar form wherein ignition takes place without evidence of the employment of any external agent. For example—there is a striking difference between the spontaneous bursting into flames of a pile of coal and the ignition of that same coal under a boiler. It is the former type of combustion which we wish to consider in this paper. When ignition has taken place combustion follows the same laws whether of spontaneous or other origin and these laws always represent more or less complex chemical actions. Almost all chemical actions are attended with the liberation of considerable heat and the quantity of heat set free is constant whether the action be relatively slow as in the decaying of a piece of wood or whether it be rapid as in the combusion of a similar piece of wood in a furnace, the products of combustion being identical in either case. Combustion, as we usually understand it, is the union of a substance with oxygen, but in chemistry we have numerous instances of combustion where oxygen is not involved at all. The velocity of combustion increases with marvellous rapidity with rise of temperature, a rise of 200 degrees Fahrenheit increasing the velocity one thousand fold. But since this paper has to deal not so much with the results as with the causes of fires, we will turn now to a consideration of a few of the more common forms in which spontaneous combustion manifests itself.


Of these, perhaps one of the commonest is the spontaneous ignition of a piece of cotton waste or an old rag, saturated with oil and left lying in a warm place. Only certain oils possess the property of spontaneously igniting and of those the most dangerous is linseed oil. the boiled variety being much worse than the raw, due to slight changes in its chemical structure which have been effected by the boiling and due also to the presence of certain metallic, salts (such as lead acetate, litharge and-manganese dioxide) which tend to accelerate the process of “drying,” by acting as oxygen carriers. Other oils which are subject to spontaneous ignition are hempseed oil, poppyseed oil, nut oil, and in fact any rapidly drying vegetable oil, the action of so-called “drying” being the cause of the ignition. Mineral oils are not generally considered to be dangerous in this respect, and sometimes a mineral oil is added to one of the above rapidly drying oils to lessen the danger of spontaneous ignition if left exposed to the air absorbed in a cloth or anything of a spongy structure which furnishes a relatively large area of exposed surface. When spontaneous combustion results from an oil-soaked rag (or an oil mop for instance) having been left carelessly in a warm room or in the sun, it is caused by the rapid oxidation of the oil spread out over the large surface afforded by the folds of the cloth —the greater the exposed surface the more rapid the oxidation—and as the oil continues to combine with more and more oxygen from the air the temperature gradually rises until sufficient heat has been developed to cause the mass to burst into flames. An excellent example of this rapid oxidation may be seen in linoleum works where linseed oil is made to drip over the surfaces of vertically suspended sheets of specially prepared cheesecloth. In so doing the oil becomes oxidized by contact with the air and gradually builds up a tough rubbery mass on both sides of the cheese-cloth. This substance is called linoxyn and is used as a binder in the manufacture of linoleum. It is also ground with cork dust to produce cork carpet. The oxidation of the oil is regulated so as to prevent the action becoming sufficiently rapid to cause spontaneous combustion. The reason why a can of linseed oil does not burst into flames when left uncovered is that the small surface exposed to the air cannot develop heat with sufficient rapidity to raise the whole body of the oil in the can to the temperature of ignition. Oxidation, however, does take place as may be seen on the surface of the oil after having stood exposed for some time. This action of drying renders these oils very valuable as paint oils, but it also introduces a grave danger of fires being produced due to the spontaneous combustion of the oils. It is interesting to note that paint oils do not dry in the sense that water or gasoline will evaporate or dry up, but their drying consists in the oils absorbing oxygen from the air, as above mentioned, to produce a substance of a solid nature in place of the liquid oil. Thus they gain in weight on drying while water and gasoline lose in weight and, in fact, finally disappear altogether. Great care should be exercised in paint and oil works where such oils are liable to leak out and come in contact with old rags, waste or other absorbent material. Numerous instances might be cited of fires having started in new houses just after the floors had been oiled—the origin in each case having proved to be the mop or cloth which had been used to apply the oil. The dangers resulting from the spontaneous combustion of oils are to be found in many industries where drying oils are employed. In tanneries the oils used for dressing the leather frequently oxidize with sufficient rapidity to take fire. In fact, freshly oiled leather is liable to ignite if left in a warm place, as near a steam pipe or heater. Here we have practically the same reaction as with the linseed oil on the mop. Another instance worthy of note is in cloth mills where oil is added to the wool before spinning, and particularly in the preparation of shoddy. If such oil treated goods are piled where ventilation is poor, the oxidation of the oils may raise the temperature sufficiently to cause the pile to ignite.

Charring of Woods.

In the charring of certain kinds of wood in the preparation of charcoal there is considerable danger of spontaneous combustion. It is a well known fact that charcoal, being of an exceedingly porous nature, has an enormous capacity for absorbing certain gases. Good charcoal made from boxwood will absorb On to 170 times its own volume of ammonia, 68 times its volume of carbon dioxide and 8 to 18 times its volume of oxygen. Freshly made dogwood charcoal absorbs oxygen so fast that it often bursts into flames spontaneously. For this reason, after charring, it has to be set away for a couple of weeks to permit it to slowly absorb moisture and air before being ground to be used in the manufacture of the higher grades of gunpowder. The slaking of lime is a process well known to everyone and yet few realize that here also lies a danger of spontaneous combustion. When water is added to quick lime it chemically unites with the lime with the evolution of large quantities of heat. If the action takes place in an enclosed space (a boxcar for instance or a bin! where the heat cannot dissipate readily, a temperature as high as 750 degrees Fahrenheit may be produced, thereby causing any wood or other readily combustible material close at hand to ignite. Few people realize that lime shipped in a car with a leaky roof may cause the car to burst into flames if the lime be very pure or “fat.” as it is frequently expressed. Farmers have been known to have their wagon-boxes charred when driving home in a shower of rain with a load of lime on. Many of us have not actually experienced such phenomena, but there can be no doubt as to the authenticity of these facts.

Combustion in Coal.

In the purification of coal gas by means of hydrated ferric oxide (a special oxide of iron) the ferric oxide removes certain sulphurous impurities (such as sulphuretted hydrogen) and gradually changes over to iron sulphide. This iron sulphide is then exposed to the atmosphere where it oxidizes, liberating sulphur and changing back to ferric oxide. During this chemical action considerable heat is generated and if the action is permitted to go on too rapidly, the mass may ignite spontaneously. Here again we have an excellent example of spontaneous combustion caused by simple exposure of the substance to the at mosphere. In industries where steam power is extensively developed, considerable difficulty is sometimes experienced in the storage of bituminous or soft coal. Certain varieties of soft coals are much given to spontaneous igniting when stored in large quantities. The usual cause of such ignition is the presence of iron sulphide in the coal. Some coals contain as high as two to five per rent, of this impurity which may occur in either of two forms, namely, pyritc, being a yellow variety, exceedingly hard and not a particularly dangerous impurity; and marcasite, a white sulphide which is much more prone to oxidation, especially in a damp atmosphere. When soft roal containing sulphide impurities is stored in large quantities and more or less exposed to dampness, the iron sulphide slowly oxidizes, yielding iron sulphate and sulphuric acid with the liberation of ronsiderable heat, If tltis heat be not carried away by suitable ventilation, a high temperature is created and ignition of the coal results. Such fires are very difficult to extinguish and necessitate the moving of a large portion of the coal, at the same time thoroughly quenching it with a hose. Not only is sulphur (present in the sulphides of iron) objectionable from the standpoint of spontaneous combustion, but it also renders roal very undesirab’e for domestic purposes, the sulphur being given off as in oxide, or burning, and eventually producing sulphuric acid which is injurious to.articles of the notisehold, especially leather and to surrounding plant life. It also causes corrosion of the grate liars and furnace fittings. Some cases of spontaneous combustion in coal have occurred which could not be attributed to the presence of pyrites, in these instances the cause has usually been assigned to the power wiiich some coals possess, of absorbing and condensing in their pores the gases of the atmosphere, thus producing heat and ignition, as already mentioned in the case of freshly ignited charcoal. Then, too, any organic matter other than the coal itself, which may he present, seems to suffer oxidation particularly when warmed in contact with air, and this also augments the tendency to spontaneously ignite. It is worthy of note that while moisture favors the oxidation of iron sulphides in coal, it retards the action of coal in absorbing oxygen, so that it is necessary to know the chemical properties of the coal in order to determine the cause of its spontaneous ignition. In either case the danger is lessened by good ventilation whereby the heat may he removed before the temperature of ignition is reached. On the other hand, no ventilation is better than a limited access of air which often favors spontaneous combustion. Acetylene, which is the gas given off when calcium comes in contact with water, is in some respects subject to spontaneous combustion. The pure gas is not dangerous, but commercial carbide usually contains impurities which give rise to the production of other gases besides acetylene. Of these, the most dangerous are phosphuretted hydrogen and siliciurettcd hydrogen, the one a phosphorous compound and the other a compound of silicon and hydrogen. Moth gases are poisonous and both are liable to burst into flame on coming into contact with air. This fact may account for some of the explosions so frequent in acetylene generators even in the absence of a jet or any external means of igniting the gas. The wdiite haze which is often seen in rooms lighted by acetylene is attributed to the formation of phosphorous pentoxide—a white powder which is formed on burning phosphuretted hydrogen and which settles very slowly. Pure acetylene has a pleasant ethereal smell, the impurities producing the unpleasant odor with which most of us arc familiar. Mixtures of acetylene and air are much more readily ignited than are mixtures of coal gas or water, gas and air. A glowing cigar is sufficient to ignite acetylene air mixtures. This gas is not particularly dangerous if handled at atmospheric pressure, but when compressed or liquified it is more dangerous and a sudden jar or a detonation nearby is often sufficient to cause it to explode. In the storage of calcium carbide great care must be observed to prevent contact with moisture or damp air which will liberate acetylene. If the liberated gas cannot escape readily high pressures may result and finally serious explosions. About onetenth as much acetylene as coal gas is sufficient to form an explosive mixture with air and the explosive range of such a mixture is very wide, ranging from :t per cent, to 82 per cent, acetylene.

Storage of Chemicals.

In warehouses where chemicals are stored, certain precautions should be observed in order to prevent accidental contact of chemicals which spontaneously ignite when mixed. For instance, potassium or sodium chlorate should not be stored where there is any possibility of sulphuric acid leaking out and coming in contact with them, for sulphuric acid has a great affinity for chlorates, giving rise to considerable heat — in many cases sufficient to ignite wood. A fire occurred some time ago from this cause the sulphuric acid was stored in carboys on a loft under which were piled several sacks of chlorate. Some of the carboys became cracked and the acid, running down over the chlorate, set fire to the warehouse, resulting in considerable loss of chemicals, Concentrated nitric acid, if brought into contact with dry organic matter, produces rapid oxidation and may raise the temrcrature of the mass to the ignition point. Dry straw, sawdust and packing materials in general are particularly subject to ignition on contact with nitric acid and great care should be taken in packing this acid to prevent spilling. When we recall that nitric acid is an important constituent of picric acid, nitrobenzol, gun-cotton, nitroglycerine, dynamite and numerous other high explosives, we begin to realize that it is a dangerous chemical with which we are dealing. The chlorates resemble nitric acid and nitrates somewhat, but are more susceptible to friction. Potassium chlorate may be ignited by stepping on some of the powder mixed with dust or other finely divided combustible. Sodium and barium peroxides may effect ignition in the same manner. In textile works where the processes of mercerizing and weighting of fabrics are carried on, certain heavily weighted silks, when very dry and closely packed, are subject to spontaneous combustion. The cause of this is not thoroughly understood, but when we reflect that weighting is often done by the addition of such substances as tannic acid and tannates, together with iron salts, which are very susceptible to oxidation, and consequent heating, we can at least approach an explanation of this phenomenon in that the iron salts and tannates take up oxygen from the air with such rapidity that sufficient heat is generated to ignite the fabric. Black silks arc more readily ignited than colored silks, due to their larger content of oxidizable coloring matter. An unweighted silk is not dangerous in this respect. In the dyeing of fabrics where potassium chlorate is used as the oxidizing agent there is considerable danger of fire due to the rapid oxidation of the aniline dye of the chlorate. YVe have already seen that chlorate in the presence of a strong acid is very liable to produce fire not only by the heat evolved during the action, but also by the liberation of chlorine peroxide, which is a highly explosive gas and a powerful oxidizing agent. Bacteria or micro-organisms arc sometimes responsible for fires, the heat being produced by fermentation or putrefaction. Although the temperatures produced by such actions are not high, yet if the heat cannot escape, a gradual rise in temperature results, which in turn accelerates the fermentative action and may cause the ignition point to be reached. An excellent example of such heating is afforded in heaps of stable manure or in tanner’s refuse. Here the heat cannot escape and the centre of the mass may reach a relatively high temperature. Any form of decaying organic, matter may give rise to a similar effect in the absence of ventilation. Such heat, of course, could never be obtained if the mass were loosely piled so that air could freely circulate. On a larger scale this heating action is frequently observed in hay stacks when the hay has been put away green or imperfectly dried. The micro-organisms so plentiful in grass and hay, act upon the fibres of the hay, producing oxidation and consequent heating (the initial heat developed by the organisms is not high, for most bacteria are destroyed below 212 degrees Fahrenheit.) Here again a free access of air would prevent undue heating by allowing the heat to escape as fast as it is generated; or if the hay were pressed into bales and the air thoroughly squeezed out from the interstices, heating would be prevented. It is the presence of a limited amount of air that causes the heating. On the other hand, it is not always desirable to entirely prevent fermentation for, by this means, the aroma, flavor and color of the hay are improved. The addition of salt is frequently the only precaution a farmer takes when storing away his winter supply of’ hay. The salt retards the action of the bacteria and renders the hay more palatable. Filer. -tric sparks caused by friction may ignite volatile solvents such as ether, naphtha or benzine, as in dry cleaning establishments. F’ires have actually been known to originate from the friction of rubbing benzine-soaked goods in a cleaning works, although such occurrences are rare. Some years ago a forty gallon tank of naphtha became ignited in a cleaning works in France. One of the men threw a bottle of liquid ammonia into the tank and the result was marvellous, the fire being almost instantly smothered by the ammonia fumes. In this case water would have been of little use. Another similar fire occurred in a ladies’ hairdressing parlor, where petroleum spirits was being used. The eminent scientist, Lord Kelvin, attributed this fire to the ignition of the spirit vapors by an electric spark from the brushing of the hair—no other source of fire being present. Finely divided particles of any combustible substance suspended in the air are subject to rapid combustion if ignited from an open flame, a hot bearing, or other cause, as in flour mills, cork grinding works or coal mines, where fire damp is not always responsible for the many disastrous explosions, which occur from time to time. Such fires might often be prevented or rendered less violent by keeping the atmosphere damp in order that the dust might tend to settle. In some works solutions of calcium chloride (a byproduct from the ammonia-soda process) are sprayed on the floors. The calcium chloride, being of a deliquescent nature, absorbs moisture from the air and tends to keep the floor damp, thus causing dust particles to adhereand settle more readily. This solution is sometimes used on the streets to keep down dust.

Location of Fire at Glens Falls, N. Y., Showing Position of Hydrants and Mains.

Combustibles of Vegetable Origin.

Combustibles of vegetable origin are much more prone to ignition than those of animal’ origin, since they all contain cellulose in ont form or another. Chemically speaking, cotton, linen and flax are almost identical with wood, all being made up largely of cellulose. Frequently it is desirable to render such substances fireproof where woven into fabrics. It is not practicable to make them thoroughly fireproof for a continued exposure to intense heat unless the materials consist of asbestos, but by treating them with solutions of certain salts as, for instance, alum, ammonium sulphate or phosphate, borax, water glass (being a soluble silicate of sodium), calcium chloride of sodium tungstate, and allowing those salts to crystalize in the pores of the fabrics, temporary fireproofing may be accomplished. The effect is simply to produce ou the fabrics a coating which can resist a short application of intense heat without igniting, or which will liberate volumes of gases (ammonia or steam) incapable of supporting combustion. These gases produce a smothering effect on the flames in proximity to the fabrics and prevent their taking fire. A mixture of three parts borax, two and one-half parts cpsom salts and twenty parts water, makes an excellent fireproofing solution, but must be applied as soon as prepared, the borax and cpsom salts being first dissolved in separate portions of water. Most of the above mentioned salts may also be employed in the fireproofing of wood. To quench a fire, one of two things is necessary. Either the temperature of the burning substance must be reduced below the ignition point or the air or gas which support the combustion must be excluded. In practice both methods arc employed, the former by a liberal application of cold water and the latter by the use of sand, textile fabrics (a coat or a quilt is often used in checking domestic fires), saline solutions or chemicals. Of the saline solutions, probably the most used is a solution of sodium bicarbonate (baking soda) in water, to which sulphuric acid may be added, by inverting the glass jar within the container. The acid reacts with the bicarbonate to liberate carbon dioxide—an inert gas which will not support combustion—and this gas dissolves in the water and is forced out through the nozzle into the fire, where it has a slight smothering effect, while the water acts in the usual way. Chemical fire engines operate on this principle. A very popular form of chemical apparatus is seen in the volatile non-combustible liquid called carbon “pyrene” extinguisher, which contains a heavy tetrachloride. This liquid is applied by forcing it through a nozzle by means of a piston and is unsurpassed for quenching burning gasoline, naphtha or benzine. The vapors of carbon tetrachloride being inert, smother the flames where water would be of little or no avail. Extinguishing powders, consisting of a mixture of saltpetre, sulphur, charcoal and resins or similar substances, act by giving off dense fumes, which fill the area surrounding the fire, and prevent access of air. Obviously such powders may be utilized effectively only within an enclosed space.

Front View of Ruins of Moynehan Building at Glens Falls, N. Y.




Spontaneous combustion is the subject of an interesting bulletin issued by State Fire Marshal Walter H. Bennett, of Springfield, Ill., who, in it, says: “Spontaneous combustion is responsible for many fires in the State of Illinois, according to reports received by this department, and it is undoubtedly true that a large per cent, of the 2,000 fires reported as being of unknown origin are in reality due to the same cause. For every substance there is a particular temperature at which it catches fire spontaneously. This temperature, of course, varies widely with the different substances. In some cases the heat generated by the slow combustion of a substance may raise its temperature sufficiently high to cause ignition, which is then called spontaneous combustion. Thus, greasy rags, sawdust with linseed oil, or wet hay piled in heaps might ignite spontaneously. Greasy rags cause many fires from spontaneous combustion. Vegetable oils, especially linseed oil, are liable to cause such fires. They take oxygen from the air so rapidly that the heat from tbe chemical union will, if not carried away, by a movement of air, rise to a point at which the fibers of the rag will first char and then take fire. It is especially dangerous to put greasy rags in closets or cupboards, because they frequently cause spontaneous combustion fires. In all combustions we have at least two substances acting chemically upon one another, which substances are generally spoken of as combustible bodies and supporters of combustion. Illuminating gas is a combustible substance, and oxygen is a supporter of combustion, but these terms are only relatively correct. Combustion may be slow or rapid, accompanied by flame or not. The appearance of flame indicates the presence of a gas or vapor which has been raised to a temperature at which it becomes luminous. The total amount of heat evolved during the rapid combustion of a substance is the same as that generated by the same substance undergoing slow combustion, but the intensity of the heat depends upon the time required for oxidation. A piece of wood in a forest may require years to combine with oxygen, or it may burn up in a few minutes in a stove, and yet the total amount of heat generated in each case is the same, though we can nteasure it in the first instance only by the most delicate instruments, while in the second it is very intense. The reaction of oxygen in the sulphur and iron found in soft coal frequently causes sufficient heat to cause serious fires. If this heat cannot be radiated away and the oxidation and increase of temperature goes on until a kindling temperature is reached, then thp gases from the coal are ignited. Moisture plays an important part in such fires. Barns are frequently destroyed by the heating of newmown hay. Dry hay does not ignite spontaneously, but hay that is not sufficiently cured frequently heats to such an extent that it takes fire and thereby much valuable property is destroyed.”