23,1 Trelease: Elongation of Banana Leaves 91 



Data of rainfall and temperature for the days during which 

 measurements were made are given in Table 3 ; but these data 

 are not sufficient to allow any close comparison between them 

 and the growth data, no clear relations being apparent. These 

 climatic values indicate that the five days considered were very 

 much alike with respect to the two main weather conditions 

 usually employed in climatology. 



As far as the available information goes, it is necessary to 

 say that some portion of the variation shown in Table 1 was 

 probably related to differences in the environmental conditions 

 of the several days, but it is impossible to suggest just what 

 conditions were thus effective, or to what degree. Certainly a 

 large part of the variation was due to internal differences 

 among the plants, differences within the plant body and not 

 related to simultaneous differences in the surroundings. 



Individual variation among apparently similar plants con- 

 stitutes one of the greatest difficulties in the interpretation of 

 the results of physiological studies. This is perhaps the most 

 important question in the whole of plant physiology to-day, 

 and it will probably remain so until we are able to obtain a 

 group of plants that will agree within a small range of varia- 

 tion. Since the rates of elongation of these banana plants 

 form a good example of individual variation, it appears im- 

 portant to call attention to the degree of variability that they 

 exhibited. 



It is seen at once from the data of Table 1 that the ten 

 plants, selected as apparently alike, differed very greatly 

 among themselves with reference to the total increment of leaf 

 elongation for the same twenty-four-hour period, with reference 

 to their day and night increments for the same day and night 

 periods, and also with reference to the ratio of night increment 

 to preceding day increment for the same day. These data 

 illustrate the very high degree of variation that may be expected 

 in a group of such plants, even when they appear by visual 

 observation to be alike, and when all seem to have the same 

 exposure. Averages based upon variables having such wide 

 deviation are plainly not reliable unless properly understood; 

 and the conclusions drawn from these data must of course be 

 based upon a consideration of the variability exhibited. 



The data of Table 1, just discussed, illustrate the great vari- 

 ability among the different plants during the same periods— 

 that is, during each of the five days when measurements were 

 made. It seems important to determine, also, how the plant 



