238 RADIATION BIOLOGY 



ill isolated thymus nuclei, very much as Caspersson had done earlier for 

 spermatocytes. They reported MX 10~^ mg per nucleus, and, for the 

 first time in cytochemistry, compared a figure obtained by cytological 

 photometry with one obtained by chemical analysis of a mass of 

 known luimbor of isolated nuclei. The two values agreed within 10 per 

 cent, which appeared to be excellent validation of the microspectrcjphoto- 

 metric procedure. It later appeared that this was merely a fortuitous 

 coml)ination of errors, since all other chemical studies indicated the 

 amount per nucleus to be nearer 6.0 X 10~^ mg (see table in Davidson and 

 Leslie, 1950b) ; hence the value computed by Pollister and Ris from ultra- 

 violet absorption is probably considerably less than one-fifth the real 

 amount. 



This type of study was later carried out under much more favorable 

 conditions by LeiK'htenberger and coworkers (1951). The citrate-iso- 

 lated nuclei were swollen in glycerin to minimize nonspecific light loss and 

 to increase the homogeneity. The ultraviolet absorption was demon- 

 strated by cross-scannings of the spherical nuclei, according to Caspers- 

 son's earlier procedure for spermatocytes (p. 237). The biochemists then 

 determined the DNA, RNA, and protein content, and showed that the 

 last two could have only slight effect upon the absorption. The amount 

 of DNA per nucleus computed from these cytological data (5.4 X 10~^ 

 mg) is much closer to that found independently by the chemical analysis 

 of the remainder of the same sample of isolated nuclei (6.1 X 10~^ mg). 



It must be pointed out that in such nuclei the DNA is determined under 

 very special conditions of minimal RNA and protein, brought about by 

 the citrate isolation. Such a condition is rarely, if ever, encountered in 

 intact fixed tissue (see Pollister and Leuchtenberger, 1949a, Pollister, 

 1952b). 



Di Stefano (1949) also computed DNA in frog cartilage nuclei from 

 ultraviolet absorption data. By comparison with chemical analysis 

 (Davidson and Leslie, 1950b), it appears that these values were deficient 

 by about as much as those given for thymus nuclei by Pollister and 

 Ris. 



Di Stefano likewise compared the value from ultraviolet absorption 

 with that computed from the Feulgen reaction, on the assumptions that 

 at maximum reaction all purines had been removed and that the fuchsin 

 regeneration from Schiff's reagent proceeded according to the scheme pro- 

 posed by Wieland and Scheuing (1921). The DNA amounts per nucleus 

 computed by these two cytological methods were in satisfactory agree- 

 ment, but, as just indicated, these are far below the value obtained by 

 chemical analysis. Protein-DNA ratios for tissue nuclei were computed 

 from Feulgen and Millon reactions, before the invalidation of the Pollister 

 and Ris and the Di Stefano attemj)ts at absolute (juantitation became 

 known (Pollister and Leuchtenberger, 1949). The Feulgen data of the 



