[Illustration: FIG. 6.--Pattern produced in gelatine by the diffusion of drops of concentrated solutions of nitrate of silver and bromide of ammonium.] Instead of the water or salt solution, we may spread on the plate a 5 or 10 per cent. solution of gelatine, containing various salts in solution. If now we sow on this gelatine drops of various solutions which give colorations with the salts in the gelatine, we may obtain forms of perfect regularity, presenting most beautiful colours and contrasts. The drops, of course, must be placed in a symmetrical pattern. In this way we may obtain an endless number of ornamental figures. In order to cover a lantern slide 8-1/2 cm. x 10 cm., about 5 c.c. of gelatine is required. To this amount of gelatine we add a single drop of a saturated solution of salicylate of sodium, and spread the liquid gelatine evenly over the plate. When the gelatine has set, we put the plate over a diagram, a hexagon for instance, and place a drop of ferrous sulphate solution at each of the six angles. The drops immediately diffuse {61} through the gelatine, and the result after a time is the production of a beautiful purple rosette. The gelatine must be carefully covered to prevent its drying until the diffusion is complete. The preparation may then be dried and mounted as a lantern slide, and will give the most brilliant effect on projection. If the gelatine has been treated with a drop of potassium ferrocyanide solution instead of salicylate of sodium, a few drops of FeSO_4 will give a blue pattern. Or we may treat the gelatine with ferrocyanide of potassium and salicylate of sodium mixed, and thus obtain an intermediary colour on the addition of FeSO_4. We may, indeed, vary indefinitely the nature and concentration of the solution, as well as the number and position of the drops. The results have all the charm of the unexpected, which adds greatly to the interest of the experiment. [Illustration: FIG. 7.--Pattern produced in gelatine by the diffusion of drops of silver nitrate and sodium carbonate.] These experiments are not merely a scientific toy. They show us the possibility, hitherto unsuspected, that a vast number of the forms and colours of nature may be the result of diffusion. Thus many of the phenomena of life, hitherto so mysterious, present themselves to us as merely the consequences of the diffusion of one liquid into another. One cannot help hoping that the study of diffusion will throw still further light on the subject.