232 RADIATION UIOLOGY 



explanation of how llic nonspccilic li^ilit losses can be estimated and .sul>- 

 lract(Kl from the compound measured curve of intracellular nucleopro- 

 lein. The nonspccilic li<ilit loss in a cell becomes evident as apparent 

 absorption outside the spectral region of specific absorption, in the case of 

 luicleoproteius abo\e 300 m/x. When extinction approaches zero near 

 this point, it is clear that one is perhaps justified in the assumption that 

 scatteriufi; and leHections cause no light losses in the shorter wave length 

 i-egion of strong specific absorption, although the possil)ility of anomalous 

 ilispersion near the absorption peak can by no means be ignored (Scott, 

 1952). Often, however, the light loss at the nonspecific zone is from one- 

 third to one-half that at the maximum of the specific absorption. It is 

 not clear why in one such case "the loss of light was assumed to depend 

 ecjually upon reflection and Kayleigh's light scattering" (Hyden and 

 Ilartelius, 1948), while in another instance it was assumed that "at 3100 

 A light-dispersion conditions the whole absorption and is inversely pro- 

 portional to the fourth power of the wave length" (Caspersson and 

 Thorell, 1942). Where there is obviously considerable nonspecific light 

 loss, it seems logical to expect to see, side by side, the uncorrected and 

 corrected curves, but there has never appeared an intracellular nucleo- 

 protein curve from which nonspecific light loss has been subtracted. In 

 some cases a curve of "light-dispersion" has been published with the 

 measured light. 



In photometric analysis it is a very uncommon procedure to attempt 

 to account for a substantial amount of the nonspecific light loss by com- 

 putation, probably because it is often difficult to determine the optical 

 constants to be used even with simple solutions or suspensions. Caspers- 

 son has chosen to do this with an immensely complex unknown sample, a 

 nucleoprotein mass in a cell, and a full evaluation of the success of this 

 attempt must await more complete details of the computations in specific 

 cases. 



The possibility of a change in light loss or in cell structure as an effect 

 of the radiant energy is not very gi-eat in measurement of absorption in 

 the visible spectrum. On the other hand, such an effect is rather to be 

 expected when working in the middle ultraviolet range, since it is a com- 

 mon experience that many substances rapidly lose their specific absorp- 

 tion upon exposure to this higher energy radiation. Caspersson was 

 aware at the very start of his researches (1936) that this might occur, and 

 presented experimental evidence that, in vitro, the absorption of nucleic 

 acid was less sensitive to ultraviolet radiation than were free guanine 

 and adenine. Caspersson (193G, p. 22) remarked that, "Die Messung 

 dieses Effects ist im mikroscop technisch ausserordentlich schwer, da im 

 mikroskopischen Priiparat Deformationen auftreten." Sections of fixed 

 material mounted in glycerin are extraordinarily stable, and an exposure 

 to intense 254 m/u radiation for many hours causes no measurable reduc- 



