( 835 ) 
other pair, with a like result, and we shall find that other details 
such as the shape of the fingers, are also shown much more clearly. 
In my opinion such series of four skiagrams, which may be viewed 
two by two in various ways, has a decided advantage over the 
usual pair of stereographic pictures. I have given the name “Tetra- 
phany” to this method. 
The result is almost as good if we make three such negatives 
as in Figure 3 with the centres of projection at the angles of 
an equilateral triangle (Triphany). In practice a threefold exposure 
will probably be sufficient. Having regard to the usual oblong shape 
of the photographic plate, it is advisable to place the foot-points of 
the principal axes in the position shown in Figure 4. 
It may in special cases be advantageous to place the three anti- 
cathodes in a straight line instead of a triangle. We may thus 
obtain a stereoscopic view of any two of the negatives. The 
outer pair give a stereo-image of half size and at half the original 
distance. 
With the ordinary mirror-stereoscope, the breadth of the plate 
should not be more than the distance of exposure, whereas the length 
of the plate is not limited. In Tetraphany (in the form of a square) 
the length as well as the breadth of the plate is limited by the 
exposure-distance. 
In Triphany (equilateral triangle) the mirror-stereoscope may easily 
be adopted for viewing the negatives. A third set of mirrors is 
added, so that each pair of images may be examined without having 
to interchange the plates. If the exposure has been made according 
to figure 4, the length of the plate must not exceed the exposing- 
distance, whereas the breadth must not exceed } 3 of the exposing- 
distance. (This number is the ratio of the vertical to the sides of 
an equilateral triangle, i. e 56: 65). 
For the examination of the images it is simplest to have a lens- 
stereoscope with three lenses, the centres of which make an equilateral 
triangle having sides of 65 millimeters. 
The principle of Polyphany which I have thus described for 
Röntgenography, may also be made use of for any other variety of 
stereoscopy. 
Il. Symphany. 
In drawing the mathematical reconstruction of a mirror-stereoscope, 
the virtual images must coincide witb the original position of the 
negatives. The stereo-image appears on the exact spot where the 
