Ltnnaan Society. 439 



spending stigma loses entirely or in part its function, which it re- 

 gains in those cases where this stamen is destitute of antheraj and 

 hence these organs, when perfect, are never placed opposite, but 

 always alternate with each other. 



The tissue of the perfect stigmata in Orchidece is not materially 

 different from that of other plants. It consists of densely approxi- 

 mated utriculi, which enlarge, and are subsequently separated from 

 each other by a viscid secretion. The channel of the style has a si- 

 milar structure, and undergoes similar changes previous to impreg- 

 nation. In the unimpregnated ovarium, the upper portions which 

 correspond to the axes of the placentee, but which do not bear ovula, 

 are neither secreting, nor do they consist of utriculi like those of the 

 cavity of the style : and the same is observable in the six lines mar- 

 ginal to the three placentse ; and these lines, both above and at the 

 margins of the placentae, Mr. Brown calls the conducting surfaces of 

 the ovarium. 



The female organ is now in a proper state to be acted upon by the 

 pollen ; and Mr. Brown has satisfied himself that it acts by being 

 brought into contact with the stigma, as Treviranus's experiments 

 proved. He applied the pollen to the stigmata in several tribes of 

 Orchidece, and found that its grains, either in the entire mass or se- 

 parately, soon produced tubes or boyanx, like those first described by 

 Amici and Brongniart. One tube is emitted from a simple grain of 

 pollen, their number consequently corresponding to that of the cells 

 of the compound grain. These tubes acquire a great length, even 

 while adhering to the grains which produce them, and have a diame- 

 ter less than nrWth part of an inch. They eventually separate from 

 the grain while immersed in the viscid secretion of the stigma. They 

 are cylindrical, neither branched nor jointed, with apparent interrup- 

 tions within, probably from partial coagulations, on the walls, of the 

 contained fluid. With a magnifying power of 150 Mr. Brown has 

 not been able to observe granules in them even in their earliest state. 

 With a power of 300 or 400, an extremely minute transparent gra- 

 nular matter may be detected. 



The tubes thus produced from the pollen mass are generally very 

 numerous, and form a cord, which passes through the channel of the 

 stigma orstyle. On reaching the cavity of the ovarium, this cord divides 

 into three parts, which are applied to ihose upper axes of the valves 

 which are not placentiferous, and at the top of each placenta each of 

 these three cords again divides into two branches, the six ultimatedivi- 

 sions thus produced passing down along the margins of the placentie, 

 along what has before been called the conducting surfaces of the ova- 

 rium. They descend to the base of the placentie, with which they are 

 nearly in contact ; but Mr. Brown has not been able satisfactorily to 

 trace branches from them mixing witli the ovula. These cords are en- 

 tirely con)])oscd of pollen tubes, and are undoubtedly essential to fe- 

 cundation, but in what manner they operate is unknown. Mr. Brown 

 adopts tlie opinion of M. Brongniart, thai the boyrtux arc derived from 

 the inner membrane of tlie grain, and believes the correctness of this 

 opinion to be demonstrated in Jsclepiadece, in which the membranes 



are 



