160 Plant Tissue Culture 



of the increase, so long as the culture does not 

 branch. If the branches formed are ignored, this 

 measurement will then be somewhat beloiv the true 

 value as measured by dry-weight or ammo-nitro- 

 gen methods, but will never, on the average, exceed 

 the true value. It will be in error in proportion to 

 the number and importance of the branches pres- 

 ent. Since normally growing roots seldom pro- 

 duce branches in great number or of great length, 

 the error introduced in this way will seldom be 

 great. Thus, for example, one series of 20 cultures 

 studied in June, 1939, in the control nutrient, had 

 a total linear increment of the main axes of 7449 

 mm. in a week's time. There were 492 branches 

 (an unusually large number) with a total length 

 of 2804 mm. Since the cross sectional area of the 

 branches would average about one-fifth that of the 

 main axes, the ratio volume of branches : volume 



2804 560 



of main axes would be about — ^— : 7450 = „.^q = 



1:13 = 7.5%. Ignoring the branches thus intro- 

 duces an error of -7.5 per cent into the record. 

 Similarly, an experimental series in which potas- 

 sium was omitted from the nutrient (April, 1939) 

 gave a linear axial increment of 1654 mm., an 

 increment in branches of 214 mm., and an error of 



9.14. 4.2 



^F : 1654 = tSt = 1:38 = 2.6%. An error of 

 5 1654 



- 2.6 per cent is less than that likely to be intro- 



