�For Practical _ Workers
��A Foot-Warmer Attachment for a Radiator
THE fact that air currents pass up through the sections of a steam or hot water radiator so that the heat rises before dispersing throughout the room accounts for the floor under the radiator being a poor place to warm the feet. The shelf arrangement shown in the illus- tration will be found very convenient for supporting the feet at a comfortable height near the side of the radiator close to the heat. The device i s made of a board 15 in. long and 6 in. wide, with two rods attached, ha\- ing their upper ends bent into long hooks to en- gage the connec- tions between the sections of the ra- diator at the top. This makes it easy to remove the at- tachment when not required. It may also be used as a shelf on which to set articles to keep them warm. — A. E. Hol.^dav.
���A shelf supported by long hooks attached to radiator connections for a foot-warmer
��How Electricity and Temperature Affect a Watch
EVERY second counts — it counts half a revolution of the balance wheel of your watch, 1,800 revolutions an hour; so, an\thing that will affect that i2-in. coiled spring that go\erns the escapement may make you miss your train — or make your train miss its signals and safety.
The effects of varying temperatures have been compensated for even in watches that need not be \-ery accurate,
��and especially in carefully designed time- pieces. The effects of magnetism are also considered, the influence of which has been minimized in the last few years because of the development of electric generators. But watches still are liable to impairment from electrical influence, even in the slight amount of static electricity created by the friction of steel wheels on steel rails when a train or car starts or stops suddenly.
A simple experiment, showing how a watch may be magnetized with- out actual contact with the magnet- izing source, may be conduc ted with paper, some steel filings and an ordinary magnet. The mag- netic field will be indicated by the lines of force in which the steel fil- ings will arrange themselves when the magnet is held under the paper. With a magnetizing source more powerful and its extent of influence proportionately greater, it may be readily seen how the steel in the watch may be magnetized — the steel parts, of course, retaining the magnet- ism, which, as the parts of the watch assume various relative positions in moving, causes them to be affected so that they are retarded or accelerated. Whereas a watch in perfect working condition may be adjusted to vary only an average of .23 seconds daily, a magnetized watch will var\' from i to 125 seconds an hour. Of course a watch may be demagnetized, but it