When working with bacteria, bacteriophages and viruses, the limits, first of resolution and then of visibility, were approached. Examples showed how resolution was increased by increasing the numerical aperture (N.A.) and the refractive index. Internal structures, involution granules and spores in dried not-fixed bacteria, bacteriophages from B. coli and staphylococci can be resolved in this way. A method of azimuthal illumination was shown to give similar results with active cells. Further, a method was described in which resolution was obtained by reduction of intensity; the reduction was effected by crossing polaroid discs or by interposing neutral grey filters.

This method of increased resolution by decreased intensity was illustrated by a number of photographs showing the separation of otherwise inseparable holes in a screen, the resolution of the shape of otherwise unresolved holes and the application of this method to viruses. In this way the slit between diplococci could be resolved after highest N.A. and n had not rendered it resolvable. The elementary bodies of certain viruses were separated and their shapes resolved. This method has been applied successfully to the measurement of the sizes of virus bodies. The measurements were checked and confirmed with a microphotometer built on the artificial star principle.

Most of the results were recorded photographically as well as observed visually, showing that the effect of increased resolution by reduced intensity can be produced if the eye is the perceptive instrument or if the photographic emulsion is employed for the purpose. The medium of reception, whether retina or emulsion grain, imposes its modifications on the effect, which was realized with a number of different optical systems by various makers.

To sum up, it appeared in the course of the experiments described that all ways of increasing resolution must lead to the same effect, that is the reduction of the diameter of the Airy disc or diffraction fringe. Whether this is done by increasing the N.A. or n, or decreasing the wave-length or the intensity, the result must be in the same direction. The factors N.A., n and λ are represented in the formula for resolution, and it was suggested that intensity should be included.