6 Transactions of the Society. 



medium the method is accurate. But in microscopic optics the 

 importance of diffraction is now universally recognized, and as you 

 know full well this is the very case in which light does not travel 

 in straight lines, new wave centres are continually being set up. 

 I remember hearing Lord Kelvin say that he had no interest 

 in any theory, however ingenious it might be, in which he could 

 not visualize the processes as they were assumed to occur. In our 

 case the conception of light in terms of wave-fronts, plane, concave 

 or convex helps us enormously. The fact that the interposition of 

 any lens results in a change in the form of wave-front appears to 

 me to simplify the matter. I might pursue this theme far beyond 

 the accustomed limits of a Presidential Address, but time will not 

 permit. 



Even at the risk of some repetition we may briefly summarize 

 the position. Kesolution is dependent on the effective numerical 

 aperture of our observing system, and the mean wave-length of 

 the illuminant. By the use of a solid cone of illumination it 

 follows that we can resolve structural elements of the order of 3 /i 

 apart, and that by the use of oblique light this interval can be 

 halved. In the latter case, however, we are unable to do much 

 more than determine the periodicity of the structural elements, the 

 microscope behaving as an interferometer. The resulting image 

 cannot be regarded as a representation of actual structure, in fact 

 we can go so far as to say that the image bears only a quantitative 

 relation to structure. The resolution of diatoms, therefore, by 

 means of narrow oblique beams, the only practical application that 

 this method has, may be an interesting experiment in physical 

 optics, but it is not microscopy. Visibility may be secured under 

 the most favourable conditions of a particle of the order of 5 micro- 

 millimetres in diameter, biit the resulting images of objects, rang- 

 ing in size from the limits of resolution to the limits of visibilitv, 

 are not sucli that any idea of form or condition can be established. 

 It is merely a proof that the object exists ; other physical tests must 

 be applied to approximately determine their size and state. 



As I have an ultra-microscope here to-night, perhaps a brief 

 account of it may not be devoid of interest. It is not a novelty to 

 most of those present, but to some it may be unfamiliar. In 

 principle it is founded on the work of Faraday and Tyndall, a point 

 which we may regard with legitimate pride, although it is no 

 exaggeration to say that the great majority of advances in micro- 

 scopy have originated, at lease in principle, in this country. In 

 1903 Siedentopf and Zsigmondy introduced the ultra-microscope in 

 the form we are conversant with at present, in which the Tyndall 

 cone is examined by means of a high-power microscope objective. 

 The effectiveness of the appliance is dependent on the concentration 

 of a great quantity of light on a small area, so that only a few of 

 the particles in the field of view are illuminated. The illumination 



