PRECIPITATION OF ANTIBODY FROM ANTI-RINDERPEST SERUM     151

be observed from this general review of the literature that a large amount of work
has been done on the fractionation of immune serum and more or less definite
knowledge has been gained as to the nature of the protein fractions with which the
antibodies in different types of anti-sera are associated. Since, however, the
nature of the different precipitated fractions of the protein depend on the nature
of the precipitant used, it is considered advisable to give a short summary of the
precipitation methods hitherto used before the object of the present research is
detailed.

     One of the earliest workers on this subject was Dieudonné [1897], who found
that globulins precipitated by means of carbon dioxide and acetic acid contained
no antitoxin, but all the antitoxin was precipitated on saturation with magnesium
sulphate. This work was done on anti-diphtheria serum. Belfanti and Carbone
[1898] and Seng [1899] used the same method of precipitation. Brieger and
Boer [1893], and Astros and Rietsch [1900] precipitated the globulins and antitoxin
by means of the chlorides of sodium and potassium. Hiss and Atkinson [1901] used
magnesium sulphate. Ammonium sulphate is, however, probably the most widely
used salt for precipitating euglobulin and total globulin, and some of the early
users of this substance were Brieger and Ehrlich [1893], Wasserman [1894], Brodie
[1897], Pick [1901], Gibson [1906], Ledingham [1907], Gibson and Benzhaf [1910],
and Hartley [1914]. Among recent workers with this method may be mentioned
the names of Findlay, and Ledingham, Morgan and Petrie, reference to whose work
has already been made. Most of these authors used a solution of ammonium
sulphate, but Barr, Glenny, and Pope [1931] have recently pointed out the supe-
riority of using solid ammonium sulphate in many of these fractionations. Felton
(loc. cit.) has introduced a modification of the older method of acid precipitation,
and both he and Goodner (loc. cit.) found that more than 80 per cent. of the immune
bodies are precipitated by this method. The principle of the method is that the
anti-serum is brought down to a certain hydrogen ion concentration at a dilution of
1: 10, and the resulting precipitate containing most of the immune bodies is
filtered off.

                  OBJECT OF THE PRESENT INVESTIGATION.

     Although, as already mentioned, Felton's experiments were carried out with
anti-pneumococcus serum, it seemed to us that his methods might be applicable to
anti-rinderpest serum. According to Hartley [1914] the antibodies in anti-rinder-
pest serum are mainly associated with the water insoluble protein, and it was
thought probable that most of the antibodies would be precipitated with the so-
called Felton fraction. In large scale work on the separation of antibodies, this