A CRITIQUE OF CYTOCHEMICAL METHODS 219 



The sensitivity of the Milloii reaction is nearly doubled if it is measured at 

 365 mn, near the peak in the near ultraviolet (Table 6-2 and Fig. 6-13). 



By contrast with the specific spot tests, the widely used method of con- 

 trast by acid dyes (in which the color is carried in the anion) gives readily 

 detectable color in parts of the cell where protein is concentrated (Table 

 6-2). For this reason, a dye of color roughly complementary to the basic 

 dye for polynucleotide is often used in preparing microscopic slides for 

 histological or pathological examination (e.g., basic methyl green and 

 acid fuchsin or basic hematoxylin and acid eosin). It has long been 

 known that this acid dye staining probably has a sound chemical basis 

 (Mathews, 1898), that it is essentially the binding of the dye anions by 

 the cationic groups (NH^) of the diamino acids of the protein to form a 

 salt (for example, what we may call, for convenience, protein fuchsinate 

 with acid fuchsin). There is evidence for the chemical basis of this stain- 

 ing reaction in demonstrations that in vitro protein binds acid dyes 

 stoichiometrically (Chapman et al., 1927; Fraenkel-Conrat and Cooper, 

 1944). The specificity of acidophilia for the amino groups of protein has 

 been demonstrated by Monne and Slautterback (1951), who showed that 

 deamination removes the acidophilia. As with basophilia the staining 

 must be carried out in acid solution (pH 1.5-2.0) so as to preclude binding 

 of the dye in other than salt formation (Mathews, 1898; Leuchtenberger 

 and Schrader, 1950). The basic amino acids constitute so large a propor- 

 tion of the composition of nearly every protein that acidophilia should be 

 a more sensitive test for proteins than the Feulgen reaction is for DNA 

 (e.g., fast green in Table 6-2). Since protein acidophilia is a measure of 

 diamino acids, it can be employed, in conjunction with either the Millon 

 reaction or ultraviolet absorption, to localize proteins of basic character 

 (Leuchtenberger and Schrader, 1950). 



All the protein methods just discussed determine the presence of pro- 

 tein only indirectly, through using one of its side groups to develop color 

 in a reagent, or to bind a dye. Hence, any semiquantitative conclusions 

 drawn therefrom are subject to all the possible errors which have been 

 discussed earlier (p. 214). Protein can be measured directly by the 

 natural ultraviolet absorption spectrum of its aromatic amino acids 

 (Chap. 5, this volume, and Fig. 6-3). The microscopic techniciue for this 

 has been developed by Caspersson (1940a, 1950), and, as usually carried 

 out, the data obtained are absorption curves of masses of nucleoprotein. 

 The qualitative and semiquantitative ultraviolet methods for these two 

 major cell components, nucleic acid and protein, are therefore discussed 

 together in the next section. 



3-4. ULTRAVIOLET ABSORPTION OF NUCLEIC ACIDS AND PROTEINS 



While chromatin is quite colorless, in the ultraviolet spectrum — at a 

 wave length a little shorter than the region transmitted b}'- glass — it can 



