452 On the Permanence of Impressions on the Retina, 



Latin word TOT, or again a number composed of figures 

 under the same conditions, such as the number 808, the in- 

 strument will reproduce each of these figures with a complete 

 identity, and we shall thus have a new and very curious means 

 of making an object in rapid motion appear to be unmoved. 



It is clear that with this system the number of the slits can 

 only be two ; for when the first point of the figure, starting 

 from its coincidence with a slit, shall have accomplished a re- 

 volution, the slit will have performed a half- revolution, so 

 that, for a second coincidence then to take place, the succeed- 

 ing slit must be situated opposite to the first. 



In order to ascertain whether this system is also the equi- 

 valent of another, either of a contrary or of the same direction, 

 let us make, in the formulae (8.) and (9.), V''^=2V',j; the two 



will then give -y^ =0, and consequently V^=0. This other 



system would then be that in which the black disc would turn 

 in any direction before an immoveable transparent disc upon 

 which should be traced a regular figure. Then, in fact, it is 

 clear that we should see simply this figure such as it is. In 

 this case the black disc might evidently have any velocit}', 

 supposing it, however, sufficient to cause a continued impres- 

 sion, and the number of the slits might also be any whatever. 



Thus for the whole system of velocities, either of opposite 

 directions or of the same direction, there exists a different 

 system, which, with the same transparent discs previously in- 

 verted, produces identically the same regular figures, and 

 which, with these discs not inverted, showsj on the contrary, 

 the regular figures inverted. 



We have all along supposed that, as in the published anor- 

 thoscope, the axes of the two discs were the one a prolonga- 

 tion of the other, and that the slits pierced in the black disc 

 were rectilinear and directed according to the radii of this 

 disc. In these hypotheses we have exhausted all the combi- 

 nations possible ; but we might contrive so that the two axes 

 should be placed at a certain distance from one another, in 

 such a way that the centres of the two movements should no 

 longer be superposed with relation to the eye, and we might 

 moreover give to the slits other directions or other forms. 

 The distorted figures would then probably be much more un- 

 decipherable still ; but their construction would also become 

 much more complicated. 



