ZOOLOGY AS A SCIENCE 



15 



all times. Temperature, humidity, air cur- 

 rents, and all other environmental factors 

 must be controlled in the room where the 

 rats are kept. Observations of any abnor- 

 malities among the rats must be recorded. 

 These abnormalities would include infec- 

 tion or any other factor that might influ- 

 ence the results. 



4. Co7npiling and interpreting the data. 

 The data must then be assembled in the 

 form of charts, tables, and graphs (Fig. 

 1-6). The rats which received no vitamin 

 Bi failed to grow normally, whereas those 



rate of growth within each group; also, the 

 poorest control rat was only slightly better 

 than the best of the experimental group. 

 However, when averaged out and plotted 

 as in Fig. 1-7, the results are clear-cut. If 

 one studied only two rats and by chance 

 got rats No. 5 ( control ) and No. 2 ( experi- 

 mental), the results might be inconclusive; 

 hence the need for an adequate number 

 of animals. This variation among animals 

 can be measured mathematically and the 

 significance of the variation computed very 

 accurately. This is the field of biometry ( ap- 



250- 



zzo 



^ 190 

 • 160-1 



2 130 

 ui 



100 



10 



Control Rat w?5 .^txperimental^at N^2 



E)cperimental Rat N^^ 



B, added 



30 35 40 4B 50 55 60 



TIME IH 0AY6 



65 



70 



75 



80 85 



Fig. 1-7. Similar to Fig. 1-6 except the growth is recorded for specific control (#3 and #5) 



and experimental (#2 and #4) rats only. 



receiving this substance were healthy nor- 

 mal rats. When the vitamin was added to 

 the diet of the experimental animals, an 

 almost immediate positive response was 

 noted in their growth. It may be concluded, 

 therefore, that vitamin Bi was necessary for 

 the normal growth of the experimental 

 animals. 



More can be learned about this experi- 

 ment if more data are studied. For exam- 

 ple, how much variation was there in the 

 growth rate of the rats? The two extremes 

 in both the control and the experimental 

 animals are plotted in Fig. 1-7. It is obvious 

 that there is considerable variation in the 



plication of statistics to biological prob- 

 lems ) and is veiy important in determining 

 the answers to many experimental problems. 

 5. Conclusions. It can be concluded that, 

 within the limits of this experiment, vita- 

 min Bi is necessary for normal growth in 

 rats. Such a statement leaves room for fur- 

 ther research which might prove that some 

 substance other than vitamin Bi is the 

 controlling factor. For example, one species 

 of bacteria, growing in the intestine of the 

 rat, might inhibit the use of vitamin Bi, 

 whereas another species might actually 

 produce the vitamin. Such a thing is pos- 

 sible, but until it is proved by further ex- 



