NoVKMBEB 30, 1S8;3.] 



SCIE.XCE. 



709 



turns the disk always worked with tlie same speed, 

 and if the number of openings were the same for all 

 experiments, we should only have to determine once 

 for all the interval of time which 

 elapses between two images, and 

 we should immediately have the 

 expression of the rapidity: in 

 short, if the successive illumi- 

 nations were separated by one- 

 tenth of a second, and if the in- 

 terAl in long measure between 

 the iina<;es were five-tenths of a 

 metre, it is evident that in one 

 second five metres would be 

 traversed. But the rapidity of 

 the disk varies with the experi- 

 ment : it must, then, be con- 

 trolled. This control can be 

 obtained by nutans of a chrono- 

 graph which shall indicate the 

 interval of time between the 

 various turns of the disk during 

 the experiment. But this meth- 

 od would give two kinds I'f independent indications, 

 — that of the spaces on the pholograpliie plate, and 



that of the times on a revolving cylinder. It seemed 

 tons better to obtain, also on the plate, the indi- 

 cations of the times elapsing between the succes- 

 sive images. This result was obtained in the 

 following manner. In order to know the fre(iuency 

 of rotation of the disk, we have only to photograph 

 the successive positions of a body moving with a 

 uniform and known velocity. Fig. 4 shows, above 

 the head of the walker, an apparatus which answers 

 this purpose, and which we will call b, photographic 

 chronograph. It is a black velvet dial, on which 

 bright nails, arranged in a circle, divide the circum- 

 ference into a certain number of equal parts. A 

 bright needle on the f.iceof this dial is in continual 

 motion, turning at the rate of a revolution a second. 

 It is evident, that, if the disk of the photographic 

 apparatus revolve only once a second, we shall have 

 only one image of the needle on the dial; if the 

 disk make si.x revolutions a second, we shall have 

 six images, etc. Since the velocity of the disk is 

 uniform, the images on the dial are separated by 

 equal distances. These divisions allow us to easily 

 estimate the fraction of a second corresponding to 

 the interval between the images. 



This method will be better comprehended if we 

 consider its application. Fig. 5 represents a runner 

 jumping a bar. The series of photographs com- 

 mences at the moment when the leaper started on 

 the preliminary spurt, and ends when tlie leap is 

 finished, and the fall to the ground has partly de- 

 stroyed the velocity. Let us analyze this figure. 

 We see the subject represented nine times; that is, 

 the disk revolved nine times during the experiment. 

 Each rotation, bringing the opening of the disk in 

 front of the objective, has permitted light to enter 

 for a brief instant, wliich has sufliced each time to 

 give an image. These successive images were pro- 

 duced at different points on the plate, because the 

 leaper himself occupied different posiiions before 

 the screen when eacli of the illuminations took 

 place. The space traversed either on the ground 



or in the air, between successive images, is easily 

 measured by means of the divisions in the planks 



