another, it seems v e r y difficult to avoid the subject of a l u m i n u m for engine construction, because, even if we d e m a n d w a t e r cooling as the only fully s a t i s f a c t o r y m e t h o d for motor-cars, we shall not evade the question of a l u m i n u m cylinders with castiron liners. T h e saving in w e i g h t b e t w e e n a l u m i n u m cylinders with liners and cast-iron cylinders and jackets is so m u c h in f a v o r of a l u m i n u m t h a t those w h o disbelieve in it will have to establish some c o m p r o m i s e b y a d o p t i n g a lighter s y s t e m of j a c k e t i n g than can ever be achieved in cast iron. In this there will be no novelty, as the s h e e t - m e t a l w a t e r jackets s u r r o u n d i n g iron cylinders are very old in motor construction, and at one time a h n o s t all the best continental machines had these jackets. In all p r o b a b i l i t y t h e y would not have been a b a n d o n e d had the earlier m e t h o d s of g e t t i n g w a t e r - t i g h t joints for these jackets been so successful as those adopted at a later period.
Cracking Hydrocarbon Vapors by Electricity. ANON. (Electriced World, vol. 7o, No. 7, P. 295, August 18, I 9 1 7 . ) - - A process that aims to obtain by means of an electric discharge a greater yield of fixed gases from hydrocarbon vapors is the development of J. G. Davidson and R. W. Ford, of Vancouver, B. C., Canada. Patent No. 1,229,o42 has been granted covering the discovery that by passing such vapors through an electric brush discharge field the amount of noncondensible gases is increased. In actual tests it is claimed that by this process the amount of C H , in the gas has been increased from ahout 25 per cent. to approximately 4o per cent., while the amounts of C, H2n were increased from between 7 per cent. to ro per cent. to between 2o per cent. and 23 per cent. Ordinarily the gas is co~lducted through a plurality of conduits formed as vertical pipes, connected at their lower ends to a supply header and at their upper ends to an outlet header, the discharge electrode being formed as wires hung axially in the pipes from an insulated support. The pipes are grounded and the electrodes are connected by a wire to a mechanical rectifier of the usual rotary type included in the hightension circuit of a step-up transformer, the rectifier connected to ground. M a r k i n g Materials. L . B . BAKER. (Scientific Americalz, vol. cxvii, No. I I , p. I95 , September 15, I 9 1 7 . ) - - I f a diamond be drawn across a piece of glass, its path is marked by a scratch, the deepness of which depends upon the amount of pressure exerted in the operation. This scratch, as we all know, is an actual rupture or tearing of the surface of the glass. Again, if a piece of lead be rubbed over a rough surface of steel a m a r k is made which, contrary to the conditions in the previous example, is composed of the marl~ing material. Finally, if a liquid or semi-liquid composed of dye or coloring matters in suspension, or chemicals capable of reacting to produce color, be placed on a smooth surface and allowed to dry, we have
lJ. F. I.
the production of a kind of marking different from that in either of the above examples. At first sight all three of these methods seem extremely simple in principle; and, indeed, the use of a hard writing object upon a material softer than itself, or the application of dyes and colors, does not offer many perplexing problems for the physicist to explain. But the use of a material which by friction leaves on the writing surface a m a r k composed of its own substance is not quite so simple. In the first place, disregarding all writing materials which inav be made of a sticky composition, it is extremely difficult, if not ~npossible, to write upon a hard surface, if that surface be smooth or polished. EiAdently, then, one of the prerequisites is a writing surface to a greater or less degree rough. A consideration of equal importance is the nature of the writing substance itself. This nmst possess the quality of yielding up very minute particles of itself when it is being drawn across the uneven surface of the material written upon. T h a t all soft substances do not possess this property will be evident if an ordinary rubber eraser, appreciably soft to the touch, be drawn over a rough surface of steel. It leaves a m a r k far less visible than that produced under the same conditions of application by a piece of lead, far harder to the touch than the rubber. Many similar examples might easily be quoted; even in the case of the ordinary pencil, we are familiar with the apparent greater hardness of the graphite core of the pencil as compared with the paper. A closer examination will disclose the fact that the property really essential to a successful marking material concerns itself with the quantity of coherence existing between the minute particles of the substance, and the relative ease or difficulty with which they detach themselves f r o m each other under the action of abrasion. F o r instance, the individual particles which go to make up the structure of a diamond are extremely hard themselves, and they are united to each other so strongly that ordinary abrasion on no known substance has the power to tear them apart. On the contrary, the core of a pencil is composed of a material the small individual units of which, although harder to the touch than the paper, are so loosely bound together that the abrasion caused by writing detaches them from their mass. U n d e r the microscope the surface of ordinary paper is seen to be very rough. The various fibres of which the sheet is made appear on the surface as a tangle of thread-like strands. Because these fibres are more or less rounded in cross-section they are slight depressions or pockets between adiacent fibres. It is these pockets which fill up with the pencil substance and make what we all recognize as the pencil mark. The pencil m a r k is not a continuous black line, but consists almost entirely of isolated pockets, as described above, each containing a little deposit of graphite. Furthermore, very little mark, if any, is left upon the fibre itself.