A CRITIQUE OF CYTOCHEMICAL METHODS 



239 



last-named authors lead, by the Di Stefaiio type of computation, to DNA 

 vahies about half those obtained by chemical analysis of isolated luiclei 

 (see Pollister, Swift, and Alfert, 1951). 



As this brief summary shows, in practice the computation of absolute 

 amounts from microscopic absorption measurements has been far from 

 uniformly successful, a result not unexpected in view of the many poten- 

 tial errors. It is apparent, however, that for many problems in cell chem- 

 istry this dubious extrapolation is not necessary; an adeciuate answer is 



300 



600 



700 



400 500 



WAVE LENGTH, mu 



Fig. 6-13. Absorption curves, measured in a Beckman spectrophotometer, of various 

 substances which are analyzed photometrically in cytological preparations. Unless 

 otherwise indicated, the length of the absorption path was 1 cm. The light vertical 

 line cutting each curve indicates the spectral region which is measured in the micro- 

 scopical preparations. Curve I, natural absorption of deoxyribonucleic acid, concen- 

 tration about 0.05 mg/cc (redrawn from Pollister and Mirsky, 1946); curve II, tyrosine- 

 mercurial formed by the Millon reaction, tyrosine concentration 0.045 mg/cc; curve 



III, fuchsin regenerated from standard Schiff's reagent by addition of formalin; curve 



IV, azure A in distilled water, concentration 0.5 mg/cc, thin cuvette {data of M. 

 Flax); curve V, methyl green (Natl. Anil., cert. No. NG35) in carbol-glycerin-alcohol 

 mixture, concentration 0.01 mg/mm' {data of C. Leuchtenberger). 



obtainable if the question of relative amount or of relative change can 

 be answered. Numerous quantitative studies with this aim have been 

 published, some of which are reviewed in Pollister (1952a). The specific 

 methods are those which have already been discussed in Sect. 3. The 

 in vitro absorption curves and the wave lengths measured in microscopic 

 material are shown in Fig. 6-13. Since these methods are colorimetric 

 (that is, measurement at a single wave length) and the apparatus is simple 

 (Fig. 6-7), large numbers of absorption measurements on individual 

 nuclei, nucleoli, or other cell structures are readily obtained. On anj^ one 

 type of structure the values are found to vary considerably, the 

 highest extinctions being often as much as twice the lowest. When 



