In their early development and before their antibodyforming
mechanism is fully established the young of mammals
depend entirely on the maternal supply of antibodies.
The passive transfer of immunity acquired by the young
from its mother can occur before birth, after birth, or
both. In man, rabbit and guinea-pig, for example, the
transfer of immunity from mother to young occurs entirely
prenatally. In the ungulate, horse and pig the passive
transfer of immunity occurs postnatally, whilst in dog,
rat and mouse it occurs both pre- and postnatally.
The difference between these groups in the mechanism
of antibody transfer from mother to young was earlier
thought to be due to the difference in the number of
layers intervening between the maternal and the foetal
circulations (Kuttner and Ratner, 1923). In other
words the structure of the placenta was the decisive'
factor in the mechanism of the transfer of passive
impunity to the young.
According to Grosser's (1909; 1927) classification,
there are four types of mammalian placental structure.
(1) the epitheliochorial with six intervening layers,
(2) the syndesmochorial with five, (3) the
endotheliochorial with four and (4) the haemochorial
placenta with three. These were later modified by Mossnan (1926), who added a fifth type, the haemoendothelial with one.
Since animals with epitheliochorial and syndesmochorial placentae transfer antibody entirely postnatally to their young, whereas in those with haemochorial and haemoendothelial placentae transfer is mainly prenatal (with the exception of rats and mice where transfer is mainly postnatal), and in those with endotheliochorial placentae both prenatal and postnatal transfer is considerable, it was perhaps not surprising that a'purely physical hypothesis of antibody transfer should be generally accepted.
But in 1946 Hartley showed that the human placenta
was selective in transferring antibody, so, that refined
horse antibody was not transferred to the foetus at all,
while human. antibody passed freely. Shortly after, in
a series of papers, Brambell and his colleagues (1949,
1950,1951,1952) showed that there was considerable
doubt whether in rabbits antibody is transferred via the
placenta at all. They suggested that antibody
circulating in pregnant rabbits is secreted into the
uterine cavity, whence it passes via the yolk-sac
splanchnopleur into the foetal circulation. Moreover
they showed that the yolk-sac splanchnopleur is a highly
selective membrane, differentiating between -globulins
not by molecular size but by*species of origin. In
particular Brambell, Hemmings and Oakley (1959) showed
that rabbit antibody digested with pepsin was less
readily transferred to the foetus, notwithstanding its
lower molecular weight, than the unmodified antibody.
When rabbit -globulin wap digested with papain (Brambell,
Hemmings, Oakley and Porter, 1960) to yield Porter's
fractions 1, II and III and these fractions were
injected into the rabbit uterine cavity, fraction III
was transferred to the foetus almost as well as the
unmodified -globulin, fraction I and II much less well.
In the Guinea-pig the route of antibody transfer from mother to young was also shown to be via the yolksac
splanchnopleur and the vitelline circulation of the
foetus (Barnes, 1957).
The present work was undertaken to investigate by
the technique of intra-uterine injections of antitoxic
sera used by Brambell and colleagues the selectivity of
the yolk-sac splanchnopleur of the guinea-pig foetus for
homologous and various heterologous antitoxins and for
the antibody fragments obtained by the peptic digestion
of the antitoxin molecule.