8 THE PRESIDENTIAL ADDRESS 



compromise ; and it has been so in this instance. According to 

 our present views the cathode rays in one aspect consist of a stream 

 of electrified particles ; in another, they consist of wave trains, the 

 length being variable in inverse relation to the momentum of the 

 particles. 



Now cathode rays have the property of being bent by electric 

 or magnetic forces, and far-reaching analogies have been traced 

 between this bending and the refraction of light by solids ; indeed, 

 a system of ' electron optics ' has been elaborated which shows how 

 a beam of cathode rays issuing from a point can be reassembled into 

 an image by passing through a localised electrostatic or magnetic 

 field having axial symmetry. This constitutes what has been called 

 an electrostatic or magnetic lens. It is then possible to form a 

 magnified image of the source of electrons on a fluorescent screen, 

 and that is the simplest application. But we can go further and form 

 an image of an obstructing object such as a fine wire by means of 

 one magnetic lens, acting as objective, and amplify it by means of a 

 second magnetic lens, which is spoken of as the eyepiece, though of 

 course it is only such by analogy, for the eye cannot deal directly 

 with cathode rays. The eyepiece projects the image on to a fluor- 

 escent screen, or photographic plate. So far we have been think- 

 ing of the electron stream in its corpuscular aspect. But we must 

 turn to the wave aspect when it comes to consideration of theoretical 

 resolving power. The wave-length associated with an electron 

 stream of moderate velocity is so small that if the electron microscope 

 could be brought to the perfection of the optical microscope, it 

 should be able to resolve the actual atomic structure of crystals. 

 This is very far indeed from being attained, the present electron 

 microscope being much further from its own ideal than were the 

 earliest optical microscopes. Nevertheless experimental instru- 

 ments have been constructed which have a resolving power several 

 times better than the modern optical microscope. The difficulty 

 is to apply them to practical biological problems. 



It is not to be supposed that the histological technique so skilfully 

 elaborated for ordinary microscopy can at once be transferred to the 

 electron microscope. For example, the relatively thick glass sup- 

 ports and covers ordinarily used are out of the question. Staining 

 with aniline dyes is probably of little use, and the fierce bombard- 

 ment to which the delicate specimen is necessarily exposed will be 

 no small obstacle. Certain standard methods, however, such as 

 impregnation with osmium, seem to be applicable : and there is 

 some possibility that eventually the obscure region between the 

 smallest organisms and the largest crystalline structure may be 

 explored by electron microscopy. 



In referring to the limitations on the use of lenses I mentioned 



