GROWTH 



<J3 



readily understood if we preface its description by a 

 simple illustration. Suppose that we have a small 

 box full of marbles of different kinds and sizes, 

 some glass, some agate, some porcelain, etc. By 

 pushing down a dividing partition exactly in the 

 middle we could divide the box into two halves, the 

 cubic content of each of which would be the same. 

 It will be seen that only by the rarest accident would 

 the various marbles be so distributed through the 

 box before we divided it that the actual contents of 

 each half space would be identical, and then only if 



FIG. 32. Four stages in the direct (amitotic) division of the follicle cells 

 of the cricket's egg. (From Dahlgren and Kepner.) 



there were no odd marbles in the original box. Sup- 

 pose, however, that to begin with we should exactly 

 divide each and every marble in two and put one 

 half on one side of the partition and the other half 

 on the other side. The result would be that the two 

 halves of the divided box would be not only equal 

 in size, which they were before, but also identical 

 in contents. If we could endow box and contents 

 with the power of growth, we see that when both 

 should have doubled in size, we would have two 

 boxes of marbles each similar to the original box. 



