A CRITIQUE OF CYTOCHEMICAL METHODS 233 



tioii of extinction or noticeable change of appearance. When living cells 

 are studied, however, the possibility of both types of alteration must 

 always be kept in mind. The physical changes induced in nuclei by 

 radiation, presumably of the sort Caspersson noted in 1936, were later 

 pointed out by Brumberg and Larionow (1946), and considered in much 

 more detail by Ris and Mirsky (1949), who believed that nearly all pub- 

 lished ultraviolet photographs of living cells show signs of radiation 

 injury, in that the nuclear details are too conspicuous. 



Although there is some room for dispute about the exact nature of the 

 effects of ultraviolet radiation on the structure and absorption of living 

 cells, there is little question but that "the living cell is, as a rule, an 

 unsuitable object for microspectrophotometric studies," because of its 

 great motility and because great structural changes occur during irradia- 

 tion (Caspersson, 1950, p. 57). This point of view is different from that of 

 earlier years, when great emphasis was placed on the advantages of 

 applying these methods to living cells (e.g., Caspersson, 1947, p. 127: "It 

 must be possible to apply the results directly or indirectly to the living 

 cell itself"). Mellors et al. (1950) have shown that these particular 

 obstacles to ultraviolet studies of living cells are by no means insurmount- 

 able. Using a microscope with achromatic reflecting optics, they dis- 

 persed the light after it had passed through the microscope, and photo- 

 graphed the whole ultraviolet spectrum of a mercury vapor lamp on one 

 negative. With very sensitive emulsions, it was found that up to 85 

 such photographs could be taken before the cell became injured to the 

 extent that mitotic division could not proceed. There may be many 

 physiological problems which can be profitably attacked directly by this 

 method. For studying the question of the nucleoprotein composition of 

 cells, however, the wide range of specific absorption in living cells seems 

 an unnecessary complication, and "for most problems suitably extracted 

 objects give the cleanest data" (Caspersson, 1950, p. 57). 



A major task in making significant absorption studies of cells is that of 

 selecting the region of the cell to be measured. All of the cytologist's 

 experience and special craft are sometimes called upon in overcoming this 

 difficulty. How can the absorption of particular cellular structures be 

 isolated from that of adjacent structures and from a surrounding visibly 

 structureless sort of background? This is accomplished easily enough if 

 a highly specific absorption is localized in a particular structure, as is 

 chlorophyll in chloroplasts or the Feulgen reaction in the nucleus. The 

 successful quantitative application of the latter (Pollister, Swift, and 

 Alfert, 1951) is in large part due to its high specificity and sharp localiza- 

 tion (Fig. 6-2F) . When, on 1 he other hand, a reaction is widespread, often 

 to nearly all parts of the cell, it is difficult to determine the absorption 

 of one particular component. Examples are protein tests (Fig. 6-25), 

 basophilia, and ultraviolet absorption, or cases where the cytoplasm 



