78 THE SCIENTIFIC MONTHLY
Fireproof construction is a safe protection against fires starting within the building, but not against major fires advancing upon the building from the outside. This device is the most expensive yet suggested, and to be effective it must apply to all buildings in a city. Berlin and Paris are pretty well protected in this way, but Tokyo, with its frail house construction and paper partitions, all of highly combustible material, would find this method of protection from fire of prohibitive expense.
Automatic water sprinklers have proved effective, reducing insurance costs by all the way from one third to nine tenths under different con- ditions. In the Baltimore fire, O^Neil's four-story department store stood in the direct path of the fire, but the curtains of Water automat- ically thrown over its windows prevented the fire entering these and though the fire entered in three places beneath the roof the automatic sprinklers within extinguished it in each place. Indeed it was O'NeiPs store, thus protected, which helped turn the fire eastward toward the little stream on whose banks it was finally checked. Water sprinklers are effective and are usually able to confine any internal fire to the room in which it arises, and similarly water curtains, outside windows and doors have proved efiScient against fires approaching from without, in the few instances in which they have been tried. But the damage from water is a serious consideration and therefore automatic water sprinklers have by no means solved the problem of fire prevention.
There is need of experimental study of the whole problem of fire prevention — including fireproof construction and immediate extinction of fires at their inception. This subject has never had any real study worthy of the name, in this country or in any other. Roughly speaking, America's annual fire loss is a quarter of a billion dollars,^ yet the total expenditure in the study of problems of fire prevention has been far less than this sum.
Why is this ? Why has this subject been so neglected ? The answer is very evident. The economic system of fire insurance, so greatly de- veloped, has removed the individual motive for fire prevention, leaving only the communal motive to urge such protection. Individual security, leached through fire insurance, has made men thoughtless and careless about the loss to the community. It is even true that the collection of insurance serves as a motive for incendiary fires and this to an extent that is an important increment in each year's loss by fire.
The chief need is a government bureau to study experimentally prob- lems of fire prevention in all its aspects. Prevention of forest fires
1 Including Canada. This figure, of course, varies widely from year to year and considerably from decade to decade. If one of the main theses of this paper is correct and all fires in buildings could readily and inexpensively be sup- pressed before damage occurs, then a large portion of the cost of maintaining our enormous fire insurance system should be included as part of the "fire waste."
�� �
