426 BOTANICAL GAZETTE [JUNE 
= 
formation was frequently inhibited, in consequence of which the 
chromosomes failed to separate normally. With the arrest of mitosis 
further development apparently ceased. 
The experiments with strychnin were unsatisfactory. Solutions 
ranging from o.o1 to 1 per cent inhibited mitosis and disorganized 
the cytoplasm, causing the death of the cells. The nuclei were not 
deformed and the chromatic figures were normal. Strychnin seems 
to arrest cytoplasmic activity swiftly, without producing visible 
changes in the mitotic figure. 
As a result of their investigations, certain authors state that nuclei 
can be made to divide amitotically through the influence of toxic 
solutions. Others, who used the same technic and methods, deny 
that such solutions produce amitosis, and find that in every case when 
division occurred the resulting nuclei were formed only by mitosis. 
BLazeEK (3) found that benzol caused the vacuoles in the cytoplasm 
to increase greatly in size; Nemec (18) observed that chloroform 
and potassium nitrate. produced granulation of the spindle fibers; 
WASIELEWSKI (30) ascribed doubling of the nucleoli to the action 
of chloral hydrate; Wovycicxt (34) states that ether prevented the 
formation of division walls in dividing cells; and W1ssELINGH (31) 
found that under the influence of phenol the cell structures were 
poorly differentiated. 
The authors just cited attributed the above-mentioned abnormali- 
ties solely to the action of the toxic substances used. In the experi- 
ments described in this paper all these abnormalities were observed 
in the toxicated material, and also in the controls grown in distilled 
water. These results appear to indicate that the action of distilled 
water is a factor which has been overlooked in interpreting the effect 
of toxic solutions on mitosis, and that numerous ab lities ascribed 
to the action of toxic substances are not necessarily so produced. 
CONCLUSIONS 
1. The practice of growing controls in distilled water, common in 
certain physiological experiments, is open to serious objections, since 
these controls are themselves under abnormal conditions, and are 
subject to the same progressive decline of cell function as occurs in 
dilute toxic solutions, though at a slower rate. 
