CHEMOTROPISM AND OSMOTROPISM 



181 



that the germ-tubes were drawn in at the stomata when positively chemotactic 

 substances were present, but passed over the stomata along the surface of the 

 epidermis when they were absent. The same result is obtained when the clean 

 epidermis of an onion scale, or a thin plate of mica is bored with fine holes and laid 

 on a mass of gelatine containing the substance to be tested. If gelatine containing 

 a chemotropic substance is placed in a capillary tube which is brought near to 

 a filament growing in water, the filament, if irritable, will show a chemotropic 

 divergence towards the open end of the tube. 



Miyoshi found that phosphates and ammonium-salts, and hence also meat- 

 extract, exert a strong attraction upon Penicillium glaucum, Aspergillus ntger, Mucor 

 mucedo, and Saprolegm'aferox, which is already perceptible in solutions of o-oi per cent, 

 strength. Cane-sugar, grape-sugar, and dextrin are less effective, especially in the 

 case of Saprolegnia, while such nutritive substances as glycerine and quinic acid exert 

 little or no chemotropic action. 

 In the case of pollen-tubes, how- 

 ever, Miyoshi found that cane- 

 sugar, grape-sugar, and dextrin 

 exerted an especially strong 

 chemotropic attraction, whereas 

 phosphate of ammonium, pep- 

 tone, and meat-extract excited 

 no positive chemotropism. It 

 is possible that this is not always 

 the case under all circumstances, 

 for Lidforss found that proteids 

 and diastase both produced 

 strong attraction \ 



The above-named sub- 

 stances act in general as 

 stimuli to bacteria, which are 

 also attracted by potassium 

 nitrate and sodium chloride, 



although these salts exercise no chemotropic action on fungal hyphae or pollen- 

 tubes. In all cases, however, hydrochloric and other acids exercise a repellent 

 action even in considerable dilution, and the same action is exercised by all 

 substances when sufficiently concentrated. 



The penetration of the hyphae of fungi through the cell-walls of a host-plant is 

 in part the result of chemotropic stimulation, but the whole problem of the relations 

 and interactions of parasites and their host is one of extreme intricacy 2 . This also 

 applies to the conduction of pollen-tubes to the ovules, which, according to Miyoshi 3 , 



FIG. 38. A portion of the epidermis from the under side of 

 the leaf of 7*radescantia discolor which had been injected with 

 a solution of cane-sugar. The germ-tubes from the spores of 

 Penicillium glaucum are seen growing towards and partly 

 into the stomata. 



1 Lidforss, Ber. d. bot. Ges., 1899, p. 236. 



3 Cf also Nordhausen, Jahrb. f. wiss. Bot., 1898, Bd. xxxili, p. i ; Behrens, Centralbl. f. Bad., 

 2. Abth., 1898, Bd. iv, p. 514. On the penetration of cell-walls by bacteria cf. Buller, Die Wirknng 

 von Bacterien auf todte Zellen, Leipzig. Dissert,, 1899. 



3 Miyoshi, Flora, 1894, p. 76, and the literature there given. On the path of the pollen-tcbe 



