274 RADIATION HIOLOGY 



since the pollen, when treated, is oriented at random with reference to the 

 radiation source, the sperm nuclei in most of the pollen grains treated are 

 at a greater depth, and the filtration losses and ineciualities of filtration 

 loss are mucli greater. Obviously, under such conditions any corrections 

 for dosage must be at best very rough approximations. 



In the liverwort Sphacrocarpus donncllii, conditions for the comparison 

 of wave-length effectiveness are incomparal)ly better, for the radiation 

 may be applied to the spermatozoid, which is an almost naked nucleus 

 about 0.5 n in thickness. Knapp and Schreiber (1939; see also Knapp 

 et ai, 1939) have compared the effects of monochromatic radiations in 

 this organism. 



For direct cytological comparison of chromosomal effects, the pollen- 

 tube technique used by Swanson (1940) with Trade scantia is also well 

 suited to the study of w^ave-length effectiveness, with minimal interference 

 from internal filtration of the radiation. Studies of this kind have been 

 exploratory only (Swanson, 1942), but they indicate that considerable 

 wave-length differences are to be expected. 



Dosage Effect. The comparison of wave-length effects requires a care- 

 ful consideration of the dosage effect for two reasons: (1) for evaluating 

 the error in approximating equal dosage with the wave lengths compared, 

 because of the varying internal filtration already discussed, and (2) for 

 determining the actual form of the dosage curve, as a basis for interpreting 

 the differences found \vith the wave lengths compared. 



The effect of variation in internal filtration among the individuals 

 treated is to flatten the dosage curve for specific effects. The population 

 treated consists of individuals varying in the proportion of the incident 

 dose that will be received by the gamete nucleus. For example, in a 

 population of maize pollen grains as described earlier, some may be so 

 oriented that the gamete nucleus is reached by radiation that has pene- 

 trated through 1() /i of overlying material, while in others the gamete 

 nucleus can be reached only by radiation that has penetrated through 80 /i 

 of overlying material. The first unit of dosage may produce the effect 

 in any of the pollen grains, and its hits will tend to occur most frequently 

 in the most favorably oriented ones. Added units of dosage may produce 

 additional effects only in the unaffected individuals remaining, which 

 offer a lower probability of hits because of the lowered dose reaching the 

 gamete nucleus in these pollen grains. The flattening of the dosage curve 

 resulting from this factor should occur at all ultraviolet wave lengths, Init 

 should be more pronounced at the shorter wave lengths since these show 

 higher absorption in the pollen grain contents. Using observed values 

 for the position of the miclei within the pollen grain and for the trans- 

 mission losses in jiollen wall and contents, the expected form of the dosage 

 curve at wave lengths 254, 297, and 302 m/x was in fairly good agreement 

 with that observed (Stadler and Uber, 1942). 



