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

16 THE CAUSES AND PREVENTION OF FIRE AT SEA.* BY PROFESSOR VIVIAN B, LEWIS, F.I.C., F.C.S. During the first five years of the present century fire at sea resulted in the total loss of 141 vessels, having a total tonnage of 162,594 tons, whilst the total number of fires would approxi- mate to twenty times this number, and it is clear that the financial importance, entirely apart from the dread nature of the subject, demands all the attention we can give it. Three years ago Mr. Edwin O. Sachs, who, as chairman of the British Fire Prevention Committee, has a wide knowlede of fires on shore, read a paper before this society on "Fire Prevention on Board Ship;" though he gave much excellent advice as to precautionary measures, he left the chief causes of fire at sea untouched, and it is in hopes of adding something to your knowledge of this all-important subject that I venture to bring it for- ward again to-night. The conditions existing on board ship differ so widely from those on land that ideas gleaned from fires on shore are somewhat misleading if taken in more than their general application. "In shore fires we are largely dealing with dangers to the structure and fit- tings of dwelling houses, and it is only when we come to warehouse fires that any great analogy can be found with fires at sea. Since the advent of iron and _ steel ships, risk from the structure is. but small, and it is danger from fittings and cargo that becomes the first considera- ' tion, and this entirely alters the ratio of danger existing amongst the various causes of fire. Taking the fire returns for 1904, in London, Liverpool, Manchester, Glas- gow, Leeds, and Sheffield, we find that the chief ascertained causes of fire were :--Matches, 1695; overheating from defective flues and fireplaces, 853; gas, 464; oil lamps, 408; candles, 327. These amount to 80 per cent of the total fires; - whilst these causes probably did not play any important part in 5 per cent' of the fires taking place at sea. In analyzing the statistics of the re- turns of losses by fire at sea, one finds, as might be éxpected, that the general cargo occupies the pride of place, whilst coal, fiber cargoes, like cotton and jute, petroleum and other oils, grain and tim- ber, figure in the order named; but such Statistics are somewhat misleading, as the number of ships carrying general cargo enormously outnumbered all others. Statistics, however, do give us a clue _ "Read at the meetings of the. forty-eighth session of the Institution of Naval Architects. but, al- THE Marine REVIEW as to the class of fires that are most dangerous when once started, if we tabulate the probable causes of fire and the total losses from these causes, we gain a valuable insight into what we may term the danger risk. ; Percentage of Percentage of cause of fibres. total loss. General cargo ...---> 51 a Coal 226 cies ees 10 Ole eae ee eee 7 10 Cotton and jute...... 32 ef 100 100 These figures certainly seem to show that a fire on a ship freighted with coal or oil is the most dangerous when once started, whilst the general cargo comes next, and that fires in a cargo of fibre, although bulking largely in occurence, are the least dangerous to the life of the ship. There are three causes which over- shadow all others in leading to disaster from fire, and these are:--(1) Spon- taneous ignition taking place in a homo- geneous cargo, or in one unit of a mixed cargo. (2) The unwitting ad- mixture in a general cargo of substances liable to raise the temperature to the ignition point by chemical action on each other. . (3) The carriage of sub- stances liable to give off inflammable fumes or vapors, which, igniting some distance from the cargo itself, cause explosion and fire. The best known example of the first. cause is the spontaneous ignition of coal cargoes, and this I dealt with so fully in a paper I had' the honor of reading before this-Institution in 1890, that I cannot do better than refer you to it for further information, although I shall have to revert to the subject in dealing with its prevention, and the methods employed in the extinction of stich fires. All fibrous structures when closely packed in bulk are liable, when slightly moist and subjected to pressure from the weight of material stored above, to undergo heating, and if a trace of oil or fatty matter be present, this heat- ing becomes so excessive as to often lead to spontaneous ignition. Of the ordinary fibers, cotton, jute, flax, ramine, and hemp are more dangerous than animal fibers, like wool and silk, this being due to the fact that the vegetable fibers really consist of microscopic tubes, which contain the necessary amount of ar to set up oxidation processes, all of which emit heat, whilst wool and sik being' practically solid do not themselves contain air, although in pack- ing it is impossible to avoid the presence of air in the spaces between the fibers, and under certain conditions this leads to; eating |<: The extraordinary effect which the smallest trace of animal or vegetable oil has in inciting this action and increasing it to the point of ieni tion of the mass is thoroughly recog- nized, and is due to both oil and fibre. It is well known that some oils haye -- the property of drying more rapidly than others, and that if a powdereq pigment be mixed with boiled linseed oil the paint so formed will harden and dry in a few hours, whilst j¢ ordinary lamp oil be employed no dry- ing action would take place. The dry- ing of oils is due to the power which they possess of taking oxygen from the 'air and becoming converted into resins, the change from the liquid oil to the solid resin causing the hardening. This process, however, like all processes of oxidation, gives rise to heat, which, al- though it escapes our notice when tak- ing place in the drying of paint spread over a considerable surface exposed to air, yet makes itself very manifest when the oil in a fine state of division js spread over the surface of such non- conducting materials as cotton or other fabrics of a similar character, and it only needs a few pounds of such ma- terials to be allowed to collect in a heap for ignition to follow in a comparative- ly short space of time. Jute offers the same dangers as cotton when in bulk, and when oiled, spontaneous ignition sometimes occurs even more easily than with cotton; whilst the same danger is found when the jute is in a thoroughly dry condition, and exposed to moder- ate warmth from a steam pipe or flue, or even the rays of the sun. Even wool when compressed in bales becomes dan- gerous when slightly moistened by rain during making up, although spontaneous ignition does not often result. Not only does danger from spontane- ous ignition exist in the fibrous struc- tures when in the raw condition, but fabrics made from them, if oiled or greased, are actively dangerous in even small quantity, a package of oiled silk umbrella cases having been known to ignite a general cargo. Oils and° fats vary very much in their power of heat- ing when spread on the surface of vege- table fibers, linseed oil being the most dangerous, whilst mineral oils are not only themselves practically harmless, but when mixed with animal oils retard their action. Another danger with such cargoes of fibrous material is that they are often very dusty, and the dust is so inflammable that when mixed with air a spark from the impact of the iron bands on the bale or the nails in 4 stevedore's boot striking on an angle iron is enough to cause ignition, and @ flame will run through the dusty aif as easily as through an explosive gaseous mixture.