IN THE DIVIDING ROOT-TIP CELLS OF THE ONION. 13 



MITOTIC SYNCHRONIZATION IN HOMOLOGOUS TISSUE-SAMPLES. 



In connecting procession indices of highest values, one observes that 

 such connecting lines run in the direction expected — that is, they trace, 

 as on the crest of an actual mitotic wave, progressively through suc- 

 cessive tinie-intervals and mitotic stages. This is one of the cardinal 

 proofs of the adequacy of the scheme of attack here followed, because 

 it demonstrates conclusively that the greatest theoretical handicap of 

 the plan (namely, the possibility that the mitotic processes are not 

 running approximately parallel in the homologous tissues sampled) does 

 not exist. For if such parallelism did not exist, no such orderly pro- 

 cession, as is here traced, would be possible. 



Additional evidence that, in homologous samples, mitotic processes 

 do run parallel is found in the work of Ward,^ Kellicott,^ and Kar- 

 sten.^ They show that in mitotically active tissues there is rhythm and, 

 moreover, that the high points of such pulsations occur, more or less 

 specifically, for the same tissues grown under the same conditions, at 

 definite periods of the day. Further evidence consists in the fact that 

 in the growing root-tips of the onion the different tips from the same 

 individual onion, grown under the same condition and having attained 

 the same length and appearance in the same period of time, must have 

 passed through processes of cellular growth and mitosis practically in 

 a parallel fashion. A discussion of an index of mitotic homogeneity 

 is presented later in this paper (p. 30). 



CAUTIONS IN METHOD. 



There are two other features of the mitotic cycle which should be 

 considered in their bearing upon the relation between stage-count and 

 average relative duration: 



(1) The cycle begins in a single cell, while at the end of stage 10 we 

 find, in place of one mitotically active cell, two resting cells. Must 

 rectifications be made looking toward a correction in the determination 

 of the average relative duration of the resting or other stages on this 

 account? The origin of the cells observed makes no difference in the 

 fact that the longer they, on the average, remain in the resting or in any 

 other stage, the more apt they are to be found in that stage at a subse- 

 quent random observation-instant. If (a) the number of cells in the 

 tissue sampled be small, and (b) all must be counted, and (c) all mitotic 

 sequences in all cells synchronized exactly, the law of averages would 

 not take care of this doubling factor in its bearing upon average relative 

 duration; but in the tissues studied only a small fraction of the cells 

 were used, and the mitotic indices of these tissues had been fluctuating 



^ Ward, H. M. "On the biologj' of Bacillus ramosus (Fraenkel), a schizomycete of the River 

 Thames." Pro. Roy. Soc. 58: 265-468, 1895. 



^ Kellicott, W. E. "The daily periodicity of cell-division and elongation in the root of Allium." 

 Bui. Torr. Club, 31: 529-550, 1904. 



^ Karsten, G. "tJber embryonales Wachstum und seine Tagesperiode." Zeit. Bot. 7: 1-34, 

 1915. 



