35 



by lhe usual precipitation processes. UM- 

 BER used albuminoids of the white of 

 egg, a crystallised albiuuin and a crys- 

 tallised globulin. ASCOLI, however, was 

 able to demonstrate by means of elective 

 absorption that the various immune sera, 

 oblaiiied by treating rabbits with diffe- 

 rent fractions of horse-serum, were not 

 identical wilh one another. The precipi- 

 tabilily of the serimi entirely disappears 

 after treatment by a sufficient amount 

 of the corresponding antigen, whilst by 

 treatment with a different fraction of hor- 

 se-serum a partial reaction is obtained, 

 enabling yet a subsequent precipitation 

 wilh the fraction used in immunising the 

 animal. Many other authors (BERTAREL- 

 LÎ, SACCONAGHl, FRANCESCHELLI) 

 have occupied themselves wilh this as- 

 pii'A of the proMem, but their results 

 do uot lead us to a more secure conclu- 

 sion than those of the first mentioned 

 authors. 



UHLENMUTH is of the oponion that 

 the chemical processes for the pre- 

 paiation and purification of these anti- 

 gens ai'e not sufficiently perfect to allow 

 us to take a stand in favour of one opi- 

 nion, or the other, seeing that the pre- 

 cipilin-reaction is far more sensitive than 

 the ones that can be made with chemical 

 reactives. 



This is also the point of view of the 

 majority of authors who have contribu- 

 ted to these studies. Recently there have 

 appeared authors who propose to explain 

 the simultaneous coexistence of preci- 

 pitin and precipitinogen on the basis of 

 the analogy between this phenomenon 

 and cei'lain states of colloidal equilibrum, 

 maintained by protective colloids (ZINS- 

 SER, YOUNG). Amongst many other ar- 

 guments, one that appears to find the 

 most favour with the authors, is the slow 

 spontaneous precipitation, «autoprecipi- 

 tation>, which takes place in the precipi- 

 tating sera on being kept some time. It 

 is well known, however, that the same 

 predpitation takes place in normal sera, 



kept under the most favourable condi- 

 tions (UHLENHUTH). This point of view 

 has been disputed by WEIL who has go- 

 ne back to the older view of VON DUN- 

 GERN, on the basis of numerous precipi- 

 tations he carried out with crystallised 

 egg alhumen. WEIL claims that, if a 

 pure protein like crystallised egg albu- 

 min is used for immunisation, antigen 

 and antibody are never found simulta- 

 neously in the blood stream. (1) 



This point, however, remains an open 

 question and still awaits a definite- 

 solution. 



The results obtained by our experf- 

 iiients do not seem to be entirely devoid 

 of interest. We thought that, for VOM 

 DUNGERN's opinion to prevail, esperi- 

 mentally, one more road lay open to 

 research, and that this was the one 

 which, taking into account the term of 

 incubation of the precipitins for the dif- 

 ferent fractions of antigenic serum em- 

 ployed, would indicate whether each 

 fraction really determined a different 

 and approximately specific incubation 

 term in the same animal species. Although 

 enormous individual oscillations in one 

 species are known in relation to the ca- 

 pacity of production of the antibodies 

 which interest, us, we began the study 

 of a series of animals, resolved to examino 

 the question from this point of view. As 

 multiple injections shorten the period 

 of incubation, while, at the same time they 

 reduce the period of permanence of pre- 

 cipitinogen in circulation, we only made 

 use, for our experiments, of animals 

 which furnished precipitins after one 

 injection alone. As it is also very likely 

 that the different precipitinogens do not 

 maintain themselves in the same propor- 

 tion in the sera of the animals of one 

 species, we always used horse seinma ob- 

 tained from one blood-letting alone. 



(l) WEIL, cited bi ZINSSER, Infection and Rcsis» 

 tance, 1918, page 271. 



