USE OF TABLES 



cording to body weight or body length or age, as the case may 

 be. Thus by the use of the tables the determinations of the 

 deviations shown by the test animals taken individually can be 

 made and these values compared with the corresponding indi- 

 vidual determinations for the control group. 3 



One further use of the tables when these are based on age, 

 may be mentioned. The comparison of the experimental re- 



3 When the experimental conditions produce control and test animals different 

 in size a determination of the relative size of any organ cannot be made directly 

 or by the assumption that its normal size is in proportion to the body lengths or 

 body weights of the contrasted groups but only by comparison of the observed 

 values with previously established normal values. 



The following observed values are taken from Hatai ('15 a), Table 3. D. Nor- 

 mal females 1914 series. They read as follows: 



It is desired to determine in this case whether the relative brain weight of th 

 test animals has been modified by the lipoid-free ration. 



The absolute brain weight of the test animals is 0.160 grams less than that of 

 the controls or 9.2 per cent of the larger number. If we assume that it should 

 be in proportion to the observed body lengths it appears that the expected brain 

 weight in the test animals would be 1.540. Hence the observed value, 1.569, is 

 about 2 per cent high by such a determination. 



If we assume on the other hand that it should be in proportion to the observed 

 body weights it appears that the expected brain weight in the test animals would 

 be 1.262. Hence the observed value is some 20 per cent too high by this determi- 

 nation. No one of these procedures is justifiable though examples of their use 

 can be found in the literature. The only correct method is to compare the ob- 

 served values with the reference table values for the brain weights of animals 

 having the body lengths of the controls and test animals respectively to de- 

 termine in each case the percentage difference between the observed and the 

 table value and finally to compare these percentages. 



Using table 68 and reading the values for the females, we find that in this 

 case the controls are 0.053 grams or 2.97 per cent below the table value while 

 in the test animals the corresponding differences are 0.103 grams or 6.16 per cent. 



The brain in the test animals is therefore smaller than that of the controls by 

 (6.16-2.97) = 3.19 per cent and this value may be taken as expressing the experi- 

 mental modification of the brain in this series. 



The foregoing represents the procedure to be generally used for determining 

 modifications in the relative weight of any organ. 



