320 ERWIN CHARGAFF 



range of variabilities that may prevail when two very large polyampholytes 

 combine with each otherJ^ 



When a polyacid, such as a deoxy pentose nucleic acid, combines with a 

 polybase, such as histone or protamine, many different salts or complexes 

 may form owing to the multiple possibilities of cross-linking taking place; 

 these may range from a close-knit fabric, the two partners being warp and 

 woof, to a treelike arrangement, in which protein molecules are attached 

 as branches to the nucleic acid trunk (or vice versa), or even to a stoichio- 

 metric sandwich, in which the minimum number of protein molecules that 

 can be accommodated is aligned lengthwise on a nucleic acid molecule. In 

 most cases, hybrids between all these forms probably will occur. This is a 

 field in which poor reproducibility is almost guaranteed, though, statis- 

 tically, there may be little difference between the results. The first event 

 that occurs when the two components are brought together may condition 

 all subsequent reactions; and the composition of the solvent, the relative 

 and absolute proportions of the partners, and even more the order and 

 rate of their mixing, will influence the quahty of the reaction products. 

 Moreover, the direction that electrostatic attraction takes first in a given 

 case will not be without influence on the type of secondary valence bonds 

 established subsequently. These considerations are of importance for a 

 decision whether a nucleoprotein isolated from a tissue may be considered 

 as intact or whether it has been converted to an artifact in the course of 

 its preparation owing to a random reassociation of its separated compo- 

 nents. (Compare also Section II. 1.) 



The effect of a series of electrolyte concentrations on several properties 

 of calf thymus nucleohistone, prepared as described in Section II.3.a.(l), 

 has been studied by Crampton et aL^" (See also Section II.4.a.) "Native" 

 nucleohistone preparations, which had not been exposed to higher than 

 0.14 M salt concentrations, were much less soluble in 0.6 M NaCl than in 

 0.7 M; those that had previously been in contact with M NaCl showed no 

 such differences. The latter preparations also had a higher viscosity. Other 

 observations that are relevant to this problem will be mentioned later in 

 connection with the discussion of the fractionation of nucleic acids (Sec- 

 tion VIII). 



The first observations on the formation of a precipitate when nucleic 

 acid and protamine are mixed are due to.Miescher.^ Since that time numer- 

 ous studies of the interaction of deoxypentose nucleic acids and proteins 

 have been published, of which only a few can be mentioned here. Much of 

 the early work is marred by the employment of degraded preparations. 

 Reactions with protamine" '^^ and histone,^''"'*^ egg albumin^"""^^ and 



^8 A. Katchalsky, Progr. Biophys. and Biophys. Chem. 4, 1 (1954). 

 " P. Alexander, Nature 169, 226 (1952). 



