176 SECTIONAL ADDRESSES. 
evidence of identity cannot be obtained until a virus has been artificially 
cultivated in an optically homogeneous medium. Meanwhile it is a 
question of the strength of a presumption, on which opinions may 
legitimately differ. Let us recognise that the evidence is not perfect, but 
beware of a merely sterilising scepticism. I suspect that the attitude of 
some critics is coloured by past history of the search for viruses, and 
especially by that part of it concerned with the curious objects known as 
“inclusion bodies,” which are readily demonstrated with relatively low 
powers of the microscope, in the cells of animals and plants infected with 
certain viruses. The nature of these bodies will probably form the 
subject of part of our discussion. From the earlier and admittedly hasty 
tendency to identify them as infective protozoa, opinion seems to have 
swung too quickly to the opposite extreme, of dismissing them as mere 
products of the infected cell. It is so comparatively simple, in some 
cases, to separate these bodies, that it is surprising that so few efforts have 
been made to test their infectivity. However, the power of such a body — 
to convey at least one virus infection has been demonstrated ; and since 
they have further been shown, in several cases, to consist of a structureless 
matrix packed with bodies looking like minute organisms, the burden of 
proof in other cases seems to me, for the moment, to rest on those who 
suggest that they consist wholly of material precipitated by the altered 
metabolism due to the infection. We shall probably hear evidence for — 
both points of view. 
The physical evidence, obtained by filtration through porous fabrics 
and colloidal membranes, and by measuring rates of diffusion, is, of course, 
ae concerned with the size of the units of infective material, and must 
e taken in conjunction with the evidence provided by the microscope. 
The crude, qualitative distinction between the filterable and non-filterable — 
agents of infection has long since ceased to have any real meaning. There 
is no natural limit of filterability. A filter can be made to stop or to pass 
particles of any required size. It is now realised that the only proper use — 
of a filter in this connection is to give a quantitative measure of the 
maximum size of the particles which pass it. Evidence from failure to — 
pass must always be subject to correction for the effects of electrostatic — 
attraction and fixation by adsorption on the fabric of the filter. A large 
amount of filtration evidence has, further, been vitiated by reliance on 
determinations of the average pore size of the filter. In dealing with an _ 
infective agent, the test for the presence of which depends on its pro- 
pagation under suitable conditions, it is obviously the maximal pore size © 
which is chiefly significant. For these reasons a good deal of the evidence 
showing that certain viruses can be detected in the filtrates, obtained 
with filters which will not allow hemoglobin to pass in perceptible quantities 
must be regarded at least with suspicion. Dr. Elford will tell us of his 
recent success in preparing filter-membranes of much greater uniformity, 
with a small range of pore-diameters. His measurements with these of 
the sizes of the particles of different viruses, show a range approaching the 
dimensions of the smallest recognised bacteria, on the one hand, and 
falling as low, in the case of the virus of foot-and-mouth disease, as about 
three or four times the size of the hemoglobin molecule ; the latter being 
