It is best to begin with antibodies which conform to their original definition as new serological properties attributable to an antigenic stimulus.

Their origin is still disputed. They are often regarded as cellular secretions due to adsorption or ingestion of antigen. For my part, I consider that they are formed in the plasma by filtration through capillary endothelium which has adsorbed antigen.

The precise specificity which characterises many antibodies is difficult to explain. I have suggested that it may be due to an interplay of reactions between the dextro- and laevo-rotatory forms of particular combining affinities.

But antibodies are not always the exact counterparts of their antigens. I have discussed some of the possible reasons why animal idiosyncrasies may be responsible for irregular results. A defective yield may also be due to the antigen, which may be present in a masked condition.

It is often taken for granted that the characters of the antibodies found in the serum of an actively immunised animal are identical with those actually present in the animal's circulation. I have given reasons for thinking that this assumption is frequently erroneous.

Coming now to the more elastic conceptions of “antibodies,” which relate to properties not attributable to an antigenic stimulus, the part played by alexin in the circulation of the normal animal has first to be considered. It is not a distinctive chemical entity, nor is it simply a chemico-physical condition of the plasma; it is a complex of chemical and physical factors which defy analysis. The dual mechanism represented in vitro between alexin and “immune body” does not afford a true picture of alexin's activities in vivo.

The humoral factors in natural resistance are too complex to be regarded as a dual mechanism. The difference between susceptibility and resistance often depends on that precise sequence of events in the plasma's activities which is peculiar to the species or even to the individual.

Acquired immunity is a reconstruction of the normal mechanism, with the emergence of a definitely new property. It is not simply the introduction of a new antibody but involves also utilisation, with readjustment, of the complex biological conditions normally present. As a result of the antigenic stimulus: (1) antibodies may be formed which are the precise mirrors of their antigens; (2) by a similar but less precise mechanism, antibodies may arise which are not the exact counterpart of any particular antigen; (3) the antigenic stimulus may lead to complex humoral changes which are not identifiable as antibodies.

Current opinions on immunological principles, as expressed by well known authorities, exhibit divergencies which cannot be satisfactorily reconciled with each other. There is, however, general recognition that the present status of knowledge is highly deficient. Though the time is not yet ripe for anything like a finally satisfactory reconstruction of these principles, efforts in this direction are needed, with readjustment of the old ideas by the aid of hypotheses about what is going on in the living body.