NA TURE 



[September 14, 1905 



In 1903, the Automobile Club tested the dust- 

 raising qualities of a large number of cars. Each 

 car was run at twenty miles per hour over a patch 

 of flour on the cycle track at the Crystal Palace. 

 The flour was kept at a definite thickness, and as 

 each car passed it was photographed. These photo- 

 graphs gave a permanent record of the dust cloud 

 raised by each car, enabling the committee to classify 

 the cars in the order of merit. The records so 

 obtained gave a great deal of useful information, 

 and it was recognised that this method of lt->Ung 

 was far more satisfactory than optical observations, 

 because an observer has a good deal of difficulty in 

 retaining a mental picture of what may be termed 

 a standard car as regards dust. 



^ In a paper read in 1903 before the Automobile 

 Club, Colonel Crompton and Mr. Crawley came to 

 the following conclusions, based on these experi- 

 ments : — Hard tyres are better than soft; narrow 

 tyres are better than broad; neither have a pre- 

 ponderating influence ; flaring mud-guards are prob- 

 ably bad, especially if they come low down ; cars 

 which are low underneath are worse than cars a long 

 way off the ground; but smoothness of bottom-shape 

 and absence of forward coning are infinitely more 

 important. There is strong evidence that it is 

 desirable that the car should slope upwards towards 



1 tricycle 



the back. At the same time, the authors point out 

 that "a low car is not necessarily a dusty one, nor 

 is a high car necessarily dustless." 



Figs. I and 2, which are reproductions of photo- 

 graphs taken during the Crystal Palace tests, show 

 the great difference that alreadv exists between 

 difl'erent makes of car, the first 'being of a com- 

 paratively dustless car, and the second oif a very dusty 

 car, both running at twenty miles per hour, the pace 

 being given by a " speed car " running alongside. 

 It will be seen that it is possible to make cars com- 

 paratively dustless, though the means of doing so 

 are not yet well understood. 



In approaching the problem, it is neccssarv to give 

 up all preconceived ideas, for the practical results 

 by no means always agree with what would 

 theoretically be expected. If a dusty car and a com- 

 paratively dustless car are examined and compared, 

 it is often not at all easy to say whv the one is more 

 dusty than the other. People are apt to have the idea 

 that comparatively small differences in the car body 

 are important. Last year, however, the Automobile 

 Club carried out a series of experiments on different 

 shapes of body, and these showed that the dust raised 

 did not depend to any great extent upon the shape, at 

 least as regards small modifications of the upper 

 NO. 1872, VOL. 72] 



structure. An irregular shape under the chassis is 

 no doubt bad, and the transverse tool-box carried 

 low down at the back of the car, which is frequently 

 seen, is certainly harmful. 



The investigation of the problem by means of a 

 car is difficult, owing to the number of variables. 

 For example, the experimental car may be a 

 " medium "' dust-raiser, and, if any modification is 

 made, the effect mav be masked to a considerable 



. 6.— Dust thrown up by a very so't tyre loaded uitl- 



cxtent by other factors which we mav not be able 

 to control. Therefore it would be better, if possible, 

 to study the problem with a much simpler apparatus, 

 so that any one variable might be taken in turn 

 without difficulty. In the hope of doing something 

 in this way, I have recently carried out experiments 

 on an elementary form of tricvcle, which is shown 

 in Fig. 3. This is drawn over a track laid with 



