Appendix.―Filtration experiments with virus of infectious ectromelia

The primary demand upon any filtration method, as an adjunct to microscopical studies, is that it should provide a means of fractionating from a polydisperse system particles which within certain limits are of a definite order of size. To accomplish this it is essential to have a series of graded filters of known porosities, capable of being reproduced with constancy. Then, by adopting a carefully regulated filtration technique, graded filtrates can be obtained, the particle size of dispersed material in any one of which is determined by the grade of filter employed. The use of such a method in conjunction with microscopical examination of filtrates provides at once a direct method for the analysis of disperse systems. This, of course, assumes that the disperse phase is resolvable by the optical methods available. Mr. Barnard has already indicated briefly the part played by filtration in the present study on the virus of infectious ectromelia. I propose to describe in more detail the lines along which the filtration experiments have been conducted, and particularly the steps in the development of the general experimental method.