IN ORCHIDE^ AND ASCLEPIADEiE. 507 



Believing, therefore^ this is to be the only mode in [705 

 which impregnation is effected, I proceeded to examine the 

 immediate changes produced by the application of the 

 pollen masses to the stigma. 



From numerous observations and experiments made 

 with this view, chiefly in Satyrinse or Ophrydese, and Are- 

 thusese, not however confined to these tribes, it was ascer- 

 tained that the grains of pollen, soon after being applied to 

 the stigma, either in the entire mass or separately, produce 

 tubes or hoyaux analogous to those first observed in one 

 case by Professor Amici,^ and afterwards in numerous 

 others, and in many families, by M. Adolphe Brongniart." 



In Orchideee one tube only is emitted from the abso- 

 lutely simple grain, while the number of tubes generally 

 corresponds with that of the divisions or cells of the com- 

 pound grain. These tubes are of extreme tenuity, their 

 diameter being generally less than l-2000th of an inch, 

 and they acquire a great length, even while adhering to 

 the grains producing them. From these, however, they 

 separate generally while still involved in the secretion and 

 mixed with the utriculi of the stigma ; and I have never 

 observed an instance of a tube with its grain attached to 

 it lower than the tissue of the stigma. In form they are 

 perfectly cylindrical, or of equal diameter, neither dilated 

 at the apex nor sensibly contracted in any part of their 

 course. I have never found them either branched or 

 jointed ; but have frequently observed apparent interrup- 

 tions in the tube, probably caused by partial coagulations 

 of the contained fluid. Even in their earliest stage, while 

 in length hardly equal to the diameter of the grain, I have 

 not been able to observe them to contain distinct granules 

 in employing a magnifying power of 150. With a Poo 

 power of 300 or 400 indeed, extremely minute and very 

 transparent granular matter may be detected; but such 

 granules are very different from those which have been 

 supposed to belong to the grahis of pollen. 



As an entire pollen mass is usually applied to the surface 



AM delict Soc. Hal. xix, par. 3, p. 254. dnnal. des Sc. Nat. ii, p. 6G. 

 Annul, des Sc. Nal. xii, p. 34. 



