NATURAL PHILOSOPHY. 147 



new method, devised by M. Garella, an engineer of the mining school, which 

 completely solves the problem. 



AVlien the operator takes a view on paper, the view is reversed naturally, 

 by the simple transfer to another sheet. Therefore, imagine an apparatus 

 constructed so as to turn on the horizontal plane which supports it, and be 

 directed at pleasure towards any point of the horizon. While this motion is 

 taking place, the exterior objects represented on the ground glass change 

 places in the same direction with the head of the objective, and if the screen 

 is suitably arranged, the operator will be able to move it in such a manner as 

 that these points remain at the same points of the image during the whole 

 time they take to cross the field of vision. As this movement is indefinite, 

 the operator may in this way sweep the whole horizon. To step from this 

 simple observation to the practical realization of a panoramic apparatus, it 

 is necessary to find the mechanical contrivance which coordinates in the re- 

 quired ratio the motion of the apparatus which sustains the glass plate on 

 which the view is to be secured, and the motion of the whole apparatus on 

 its plane. Now, it is easily demonstrated that the motion of the screen must 

 be such that its plane moves without slipping on the ideal cylinder which 

 describes the mean vertical line of the focal plane by its motion round the 

 optical centre of the objective. If this motion of the screen could be effected, 

 each of its points would describe in space an evolvent circle; if, therefore, the 

 evolvent is materially contracted, and secured on the base of the apparatus, 

 and the apparatus which contains the glass plate is fastened to its edges, so 

 as to follow its contour, it would necessarily constantly occupy a position 

 which would assure its relative steadiness to the image. Such is the elegant 

 solution proposed and practised by M. Garella to obtain, on a rigid plate, as 

 extensive panoramic views as the operator may desire. If the operator does 

 not reverse the view obtained directly in the camera obscura, the image is 

 engendered progressively, band by band, on the inner surface of a supposed 

 cylinder, Avhich corresponds exactly with M. Marten's real cylinder. But if 

 he desires to reverse the objects at once, the image must be traced on the ex- 

 terior surface of a cylinder of the same radius. He Avill then see that the 

 centre of rotation of the apparatus must be moved back of the screen, and 

 to a distance equal to the principal focal length of the objective. This 

 change (which is the geometrical translation of a change of sign) enables 

 the evolvent of circle to continue to guide the apparatus which supports the 

 plate and secures the clearness of the images. M. Garella' s plates exhibit by 

 their clearness the perfection of this mechanical process. They cover a sur- 

 face of great length, and when they are seen lying flat it is easy to detect that 

 they represent impossible scenes. Each part maybe examined by itself; but 

 when the spectator attempts to take in the whole, he at once perceives very 

 great deformities in the horizontal lines, which are caused by the difference 

 existing between the position in which it is seen and that in which it was 

 formed, part by part. As soon, however, as the curve of the ideal cylinder 

 is reestablished where it was formed, and the observer places his eye at the 

 point occupied by the objective of the camera obscura, all the objects appear 

 at their real angles. 



THE "ELLIOTYFE" PROCESS. 



A new application of photograph}' is now a matter of such frequent an- 

 nouncement, that the reader's patience is sorely tried by an invitation to 



