464 BIOLOGICAL EFFECTS OF RADIATION 



very resistant alga and yeast cells must be exposed to beams whose 

 intensity is several thousand r per min. in order to reduce the length of 

 exposure to reasonable limits. 



A constant degree of sensitivity is a great advantage for it permits 

 direct comparisons between the results of different experiments. Drosoph- 

 ila eggs have shown a remarkable degree of constancy; 190 r kill half 

 of the eggs obtained from different strains of wild flies which have been 

 worked with for a number of years. Precisely the same dose has been 

 found for Drosophilas tested in Germany (23). 



c. The necessity of using small organisms in these experiments is 

 obvious when very soft radiations are employed, for their penetrating 

 powers are so slight that there is an appreciable difference in the amount 

 of energy received by the upper and lower sides of the object. Even 

 with Drosophila eggs whose greatest diameter is about 0.18 mm., the 

 correction for absorption of rays generated at 12 kv. amounts to 6 per cent. 

 For still lower voltages this correction rapidly becomes much more 

 important. 



d. Genetic purity is considered important, especially in plant mate- 

 rial. But tests with hybrid strains of wild Drosophila indicate that it is 

 not significant. 



It is evident that the choice of test objects which will fulfill these 

 requirements is limited. Many seeds are of little value because of their 

 high powers of resistance. This is true for the protozoa in general. 

 Eggs are usually available for limited periods only, and often in numbers 

 too small to furnish significant data. Drosophila eggs, however, conform 

 to the above requirements. 



The amount of normal variation which occurs in the various test 

 objects which have been used has scarcely been mentioned in the litera- 

 ture, yet an evaluation of the published results of radiation experiments 

 cannot be made without some information on this point. Bolaffio (1), 

 who used the seedlings of Vicia Faba, remarks that "almost never do all 

 seeds in a test act alike. Many times in one group the differences are so 

 great that the entire test which has taken so much time and labor must be 

 discarded. In the presentation of data there is always a strong subjective 

 factor, so that the results represent not a measure but a preference." 

 Lachmann and Stubbe (24), however, believe that the amount of varia- 

 tion can be greatly reduced by appropriate culture methods. Data given 

 by Zuppinger (44) in his work on Ascaris eggs shows that the coefficient 

 of variation varies in different tests between 12 and 28 per cent. 

 But he states that differences of 5 per cent between the means of 

 two sets of experiments are significant. An analysis of data ob- 

 tained from tests with Drosophila eggs indicates a coefficient of 

 variation ranging from 3 to 8 per cent. Table 2 gives the results of a 

 series of 10 tests. 



