DEPARTMENT OF BOTANICAL RESEARCH. 67 



have been chosen from which such vacuoles are absent. The proto- 

 plasts of many (though not all) pollen-grains appear to satisfy these 

 conditions. These offer ready-to-hand material, with certain disad- 

 vantages indeed, but such as are not too serious. These are chiefly 

 due to the presence of the more or less rigid envelopes (extine and 

 intine). In the future it may be found quite possible to eliminate 

 them by using j^ounger material. For the present, the observations 

 which follow have been derived from the study of mature pollen, for 

 the most part of Eschscholtzia calif arnica. 



Man}" pollens burst on being placed in tap-water. When the proto- 

 plasm is thus set free, an internal secretion of water is begun, in conse- 

 quence of which the whole mass becomes a froth, suggesting that the 

 formation of water vacuoles in the growing cell is the result of growth 

 and not a method of increasing pressure. Inasmuch as no vacuoles 

 are present, the bursting depends entirely on the hydratation capacity 

 of emulsion colloid,s and not upon the presence of solutions of high con- 

 centrations, such as would be required to overcome the resistance of 

 the envelopes. According to expectation, the hydrophile colloids con- 

 stituting the protoplasm behave in certain respects as, e. g., do the 

 hydroceliulose content of certain tannin cells and the analogous muci- 

 lage of the mucilage cells of mallows and cacti, which, when allowed 

 to, may absorb water with sufficient avidity to burst the cell wall. 

 Thus, the protoplasm of pollen will swell and may burst the envelopes 

 even in solutions of high concentration, either after initial shrinkage 

 or so quickly that an initial shrinkage may be assumed to occur, if 

 indeed it does; e. g., pollen of Gossypium bursts in 0.45/ N KNO3 in 

 50 per cent cane sugar and 25 per cent glycerol. Similar behavior is 

 shown by other kinds. Even in concentrated salts {e. g., KNO3), 

 glycerine, etc., the protoplasm swells sufficiently to distend the enve- 

 lopes in the course of 15 to 30 minutes, though it is preceded by 

 partial or complete shiinking. Weak acids and alkali cause swelling 

 and (when the concentrations are proper) bursting and the formation 

 of coagula characteristic of each reagent. 



Pollen which will not burst at once in water may do so (Lupinus) 

 after about 2 hours' germination, when the pressure becomes sufficient 

 to overcome the resistance of the thin membrane just at the apex of 

 the pollen-tube, and it is important to note that minute water vacuoles 

 which are present in the tube protoplasm at the time have no measur- 

 able influence. The bursting of the envelope, therefore, serves as one 

 index of the hydratation capacit}^ of the protoplasm, and, without 

 disregarding others, serves to bring out many phases of behavior. The 

 escaping protoplasm is variously distorted as it escapes, as a jeUy 

 would be under the same conditions, by the envelopes, and the observer 

 is enabled to note the change or absence of change in the form of the 

 suspensoids, and thereby to judge of their physical characters. 



