Physiology of the Spermatozoa of Ferns . 565 
The strongly alkaline potassium carbonate does not attract. 
This is not surprising on account of the fact that this substance 
(owing to hydrolysis 1 and consequent formation of O H - ions) 
exercises a strongly toxic effect upon spermatozoa at a con¬ 
centration below that probably necessary for the ion K + to 
give an appreciable attraction. 
The following 2 substances give attractions of approximately 
equal strength : malic acid, ammonium hydrogen malate and 
the malates of sodium, potassium, ammonium, calcium, barium 
and magnesium. All these substances are dissociated and 
each contains the negative radicle of malic acid, 
They attract, as already mentioned, much more strongly than 
any other substances which do not contain the radicle in 
question. Some of the bases, namely sodium, ammonium 
and calcium, when occurring as ions in solution, according 
to my previous argument do not appear to appreciably attract 
at all. We may therefore conclude that the C4H4O5" ions 
attract. 
Although we have good reason to suppose that the ion K + 
attracts, solutions other than malates containing K + ions cease 
to attract at the concentration approx. ¥ <nr m °l- Potas¬ 
sium malate, however, attracts at Toroo It appears, 
therefore, that at this concentration potassium malate owes 
its attraction entirely to the C 4 H 4 Or ions, while theK + ions 
are practically chemotactically inactive. It is doubtless due 
to the indifference of all the kations at low concentrations 
that all malates attract with about equal strength. 
If it is the ion C 4 H 4 0 r of malic acid which is responsible 
for the attraction of this substance it is no matter for surprise 
that when this group of atoms is present in an undissociated 
compound, attraction no longer takes place. This happens in 
the case of the di-ester of malic acid, as was pointed out by 
Ostwald. Both Pfeffer 3 and Voegler 4 found that this substance 
1 Ostwald, Grundlinien der anorganischen Chemie, 1900, pp. 255, 256. 
2 Compare Tables I, II, and III; Pfeffer, loc. cit., Bd. i, p. 381; and Voegler, 
loc. cit., p. 659. 
3 Bd. ii, p. 655. (1 & / 0 and 0*1 °/ 0 solutions were tried.) 
4 Voegler, loc. cit., p. 659. 
