COLOR PHOTOGRAPHY. 185 



IX. — Prism Experiment with Seebeck's and Poitevin's Color 

 Beproductions. — First Proof that they are of the Nature 

 of Body Colors. 



The prism experiment with Seebeck plates is beset with difficulties 

 wbich delayed the present investigation very considerably. The silver 

 chloride powder must be retained between two glass plates. It does 

 not suffice to pour benzene between the cover plate and tlie prism in 

 order to see the colors through the latter, for total reflection occurs at 

 the boundary of the air space between the cover plate and the particles 

 of powder. The air must therefore be completely expelled by a liquid 

 of not too small index of refraction. Benzene was chosen for this pur- 

 pose. The introduction of the benzene could not, however, take place 

 after the exposure, for it was found impossible to do this without alter- 

 ing the position of the particles of powder. Hence the benzene was 

 flrst poured between the plates and then the powder was stuffed in 

 between them. 



A square cornered metallic frame served to support the whole. In- 

 stead of glass, a mica plate about 0.08 mm. thick was used on the 

 front side. It was thus possible to avoid a slight displacement of the 

 reference mark with respect to the spectrum in the prism experiment. 



The presence of the liquid did not interfere with the production of 

 the colors by the action of light. These came as before, only more 

 quickly, for by absorbing the free chlorine given off the liquid made the 

 plate more sensitive, but this added considerably to the difficulty of the 

 prism experiment, as the observation required to be quickly completed 

 before the action of daylight obliterated the spectrum reproduction. 



The experiment was finally repeatedly performed with success. The 

 reference mark was drawn with a diamond in the red and blackened 

 with soot. There was no perceptible displacement of colors with respect 

 to it. At the same time it was worthwhile to secure greater certainty 

 by a simple modification. For this purpose pure chloride-of-silver 

 powder was stirred up with collodion and the mixture poured upon a 

 glass plate and dried. There was thus obtained a film in which the 

 chloride of silver was retained by collodion. It was then fixed to a 

 glass plate. The reference mark was made with a lead pencil on the 

 film itself and the prism experiment performed without possibility of 

 error. 



The plates in benzene could not be left long in the daylight. The 

 room was therefore darkened and light was admitted through a hole 

 covered by a double layer of filter paper. The new process had, how- 

 ever, the advantage that the colors appeared more distinctly. Under 

 the prism they Avere, to be sure, darker ; but the result was repeatedly 

 reached with certainty that no displacement of the colors in the prism 

 with respect to those in air occurred for an unbended reference mark. 



No difference in this respect was found between the coarse-grained 



