August 19, 1909] 



NATURE 



sent is usually about 20 per cent., and a method of lower- 

 ing the amount which has been employed recently is to 

 admit exhaust gases into the producer in place of the usual 

 water supply. The CO, then takes the place of H.O as 

 an energy absorber, and the chemical composition of the 

 resulting gas is so greatly affected that the hydrogen sinks 

 to less than i per cent. Another recent improvement in 

 the suction gas producer is the regulation of the water 

 admitted, so that the composition of the gas may be the 

 same, no matter whether the " draw " of the engine is 

 vigorous or slight. Without some such device the gas 

 tends to become " wet " at the lower loads, and the pro- 

 ducer to " go dead." To avoid this, the water supply to 

 the producer has to be cut off, or greatly reduced, in pro- 

 portion as the governor is cutting out working strokes in 

 the engine. 



Improvements in Class (c). 



These are exceedingly numerous. The most recent 

 decided change has been the adoption of the sleeve type 

 ■of valve in the Daimler engine. Despite apparent draw- 

 backs from the theoretical point of view, it has lately 

 undergone an extremely severe test under the officials of 

 the Royal Automobile Club, and has emerged victorious. 

 It is understood that a number of other manufacturers are 

 now considering the adoption of the slide type of valve. 



Another innovation is the use of air pressure to force 

 the fuel up from low-lying tanks to the carburettor. This 

 is claimed to be an improvement on the exhaust pressure 

 ■feed, as being less likely to choke. As regards heavy oil 

 engines, the chief improvement to be noticed is the wide- 

 spread use of water injection, which is even more useful 

 here than in a gas engine on account of the very low 

 pre-ignition point of mixtures of oil vapour and air. 



Attempts have continually been made to produce an 

 engine working on the two-cycle principle, and there would 

 seem to be no reason why engines of this kind should not 

 be as practically successful as any built to operate on 

 the four-cycle principle. One would expect that in this 

 way a lighter engine could be built, and lightness is a 

 great asset, particularly in the most recent use to which 

 the internal-combustion engine has been put, viz. aero- 

 plane work. Wonderfully light engines have alreadv been 

 made for this purpose. The 50 horse-power engine on 

 the aeroplane Silver Dart weighs only 220 lb. without oil 

 and water, or 4^4 lb. per horse-power. The 50 horse- 

 power " Wolseley " V-type eight-cylinder engine, built for 

 aeroDlane work, is reported to weigh only 340 lb., or 

 6-S lb. per horse-power, although the engine is fitted with 

 a water-cooling system. The somewhat similar 80 horse- 

 power engine fitted to Mr. Moore Brabazon's aeroplane is 

 reoorted to weigh only jCifi lb., or 4-6 lb. per horse-power; 

 this is also a water-cooled engine. The possibilities of 

 the different types of engine have lately been summarised 

 in the technical Press ' as follows : — 



Type of engine 



Four-cylinder vertical 

 Eight- ,, diagonal ... 

 Diagonal, wiih several cylinders 



grouped on one crank pin 

 Rotary 



From this it appears that engines amply light enough 

 for aeroplane work have already been built, and there is 

 not much scope left for any improvement in this direction 

 for which the two-cycle engine would be useful, although 

 ■there is still room for general improvement by the avoid- 

 ance of the use in the engine parts of all heavy materials 

 of low mechanical strength. On the other hand, the 

 lighter the engine the better, as it means the possibility 

 ■of adding additional accessories that make for constancy 

 of operation, such as duplicate ignition, ample water- 

 cooling arrangements, &c. The aeroplane appears to offer 

 little chance of an " altitude stop " to permit of the engine 

 being adjusted, and constancy of operation is therefore 

 the one essential thing so far as the motor is concerned. 



Carburettors are continually being improved, but the 

 ideal one, which will give a constant mixture at all speeds 



1 Fut^'/nefrr'ft^, April i6. 



NO. 2077, VOL. Si] 



and all loads in all weather conditions, has yet to be in- 

 vented. The high-water mark as regards fuel economy 

 that has so far been reached is the performance of the 

 White and Poppe carburettor in the 1907 Royal Auto- 

 mobile Club trials of commercial vehicles. This carbu- 

 rettor was fitted to a Maudslay car, and showed the very 

 high "figure of merit " of 62 gross ton miles per gallon 

 of petrol, which is nearly twice as good as was obtained 

 from the average car of that time. In the last two years 

 the average has, however, risen appreciably. This, with 

 an assumed road resistance of 50 lb. per ton, would 

 correspond to an efficiency of power transmission between 

 the carburettor and the road wheels of no less than 15 per 

 cent. What the road resistance really was is not known, 

 but now, without doubt, such road resistances ought to 

 be accurately measured and the results applied. 



Conclusion. 

 With such a rapidly moving industry as that of the 

 internal-combustion engine, prophesy is even more unsafe 

 than it usually is. Writers have been bold enough to look 

 forward to solid explosives being employed, but there one 

 is faced with the ditliculty of selecting any form of solid 

 explosive that would have an entirely gaseous exhaust. 

 When the gas turbine has taken practical shape, this con- 

 sideration may be of less importance. Indeed, the com- 

 bination of a solid explosive with a gas turbine promises 

 this advantage, that the difficulty of the initial com- 

 pression would thereby be removed. On the other hand, 

 if we may judge by analogy with the steam turbine, pro- 

 vided that it were possible to keep the exhaust pressure 

 sufficiently low, a high initial pressure would not be 

 essential to economy. Whatever may be the outcome of 

 the present experiments with gas turbines, or of gas pro- 

 ducers, suitable for marine purposes on the one hand or 

 to road transport on the other, one may be certain that 

 the days of the external-combustion engine, the steam 

 engine, are numbered, and that the engineers of the near 

 future will not be satisfied with any less degree of efficiency 

 than that which the internal-combustion engine will afford. 

 One seems to see in the world of engineering the work- 

 ing of a continuous process leading to the supersession of 

 those ideas, which, though old and tried friends, are found 

 to produce less efficient results than those obtainable by 

 more scientific methods. There is no reason that the 

 writer can see to doubt the continued operation of this 

 process during the present rivalry between the steam 

 engine and the internal-combustion engine. 



H. E. WiMPERIS. 



OSMOTIC PHENOMENA AND THEIR 

 MODERN PHYSICAL INTERPRETATION.' 



OSMOTIC pressure is a phenomenon of such import- 

 ance in the theory of solutions, and in the interpreta- 

 tion of all vital processes, and so much valuable work has 

 recently been directed to its elucidation, that, although it 

 is a somewhat thorny and difficult subject, no apology is 

 needed for any serious attempt, however inadequate, at its 

 explanation. 



One of the earliest recorded experiments on osmotic 

 pressure is that of the Abb6 Nollet, who found that a 

 bladder containing alcohol, when immersed in water, 

 absorbed water so greedily as in many cases to burst the 

 bladder. The experiment illustrates in _ an imperfect 

 manner the fundamental property of all animal and vege- 

 table membranes of allowing some substances to pass 

 through them by osmosis more easily than others. In 

 many cases such membranes, while freely permeable to 

 water, are practically impermeable to certain substances 

 in solution, and play the part of sieves in directing and 

 controlling diffusion.' It will readily be understood that 

 results of the greatest importance to biology have been 

 obtained bv studving this property of semipermeahility, 

 as it is called, but the application of natural membranes 

 to the phvsical study of the subject is necessarily limited 

 on account of the difficulty of obtaining sufficiently large 

 and perfect membranes capable of withstanding any con- 

 siderable pressure. 



1 Discourse delivered at the Royal Institution on Friday, February 26, 

 by Prof. H. L. Callendar, F.R.S. 



