554 CLARENCE E. McCLUNG 



only this, but the grouping is maintained and the derivatives 

 of each of the original series remains associated with its fellows 

 and, with them, forms a common vesicle. So far as prophase 

 and telophase conditions are involved it would not be observed 

 that anything unusual in nmnerical relations obtains. Even 

 in the matter of this multiplication the chromosomes exhibit 

 individual differences, and one may have advanced in the proc- 

 ess beyond the others and in this way produce a total number 

 not strictly a multiple of three. But no matter how many 

 or how few there may be, each one of the individual groups repeats 

 the characters of the original member. How far removed from 

 mere chance crystallization out of an indifferent matrix is the 

 ordered and definite reproduction of distinct morphological 

 entities which we see here. In the very face of what seems at 

 first glance to be the grossest of variabihty there appears only 

 another marked instance of exact organization. It is of interest 

 to observe also that division of the chromosome may occur 

 independently of mitosis, which suggests, in connection with the 

 observation that the actual splitting of the chromosome is a 

 prophase change, that mitosis may be more concerned with cell 

 division than with chromosome separation. 



3. Chromosome sizes 



If the total amount of chromatin is constant for each cell and 

 if the number of chromosomes is unchanged, it might reasonably 

 be expected that the same series of sizes would always appear. 

 These conditions are certainly realized in the individual, where 

 the closest correspondence exists between the series and between 

 individually recognizable elements. That like correspondence 

 exists between individuals seems probable when the seriation in 

 the complex is noted. An exact determination of chromosome 

 sizes is however a very difficult undertaking, owing to the variety 

 of forms they assume. One is forced to estimate relative vol- 

 umes, and an exact determination between two nearly equal 

 sizes of different form can not be made with certainty. Fortu- 

 nately the difficulty of comparing sizes is less in the case of the 



