356 MORPHOLOGY. 



A. lophantha). In the Orchidacea, the entire product of the solution 

 of the parent-cells and special parent-cells is viscid, and glues all 

 the pollen granules in a mass, and is easily recognised between 

 them, as a tough substance, capable of being drawn out in threads. In 

 the Asclepiadacea it appears that only the parent-cells are absorbed, 

 and that at a very early period ; the special parent-cells are not ab- 

 sorbed at all, but adhere closely to one another, and so form, out of 

 the whole mass of pollen a loculus, or small cellular body, which 

 appears clothed with a special membrane, since in the outermost 

 layer of the special parent-cells no pollen granules are developed. 

 Probably there exists in all the cells, from the parent-cells to the 

 pollen granules, a circulation of sap (in the latter most certainly, 

 in the young condition), the currents of which form reticulations on 

 the cell-wall. 



I have, in the preceding account, given the history of the develop- 

 ment of the pollen, essentially in accordance with the excellent investi- 

 gations of Nageli*, as I have nothing more complete of my own to 

 oppose to it. Let Plate IV. and the explanation be compared with the 

 preceding. I by no means suppose that what has been said will be found 

 the complete settlement of the entire matter, and I cannot withold some 

 thoughts which strike me with regard to the formation of the cells. 

 There are two difficulties which certainly nowhere oppose so great a bar- 

 rier to the completion of investigation as here ; namely, the attainment of 

 perfect stages of development, and the correct arrangement of them in 

 their proper series. I will offer the following objections to Nageli. I 

 will first mention that I have convinced myself, both in the parent-cells 

 and the special parent-cells (in the cases of Pepo, Passiflora princeps, 

 and Arum maculatum), and also in the young pollen-ceH'(in the cases of 

 Lupinus, Larix, Pinus alba, Juniperus, Richardia cethiopica, Arum ma- 

 culatum, and Fritillaria imperialist that the cytoblast is clearly to be 

 recognised as parietal (sometimes even in the pollen-cell which has 

 sent a tube into the nuclear papilla of the seed-bud). In Fritillaria, two 

 kinds of cytoblasts are easily to be distinguished, those which give the 

 origin to the pollen-cells, and lie embedded in the walls, and those 

 which form later in the pollen-cells and here, as is not rare in the 

 pollen-grain generally, give origin to a transitory process of cell-forma- 

 tion.f A second point, wich appears to me important, is that I have 

 frequently observed, between the period of the existence of the empty 

 parent-cells and their regular division into two or four special parent- 

 cells, free cytoblasts floating among the granular contents of the parent- 

 cells (Passiflora princeps), and I have seen also this accompanied by very 

 delicate young cells with cytoblasts on the walls (in Passiflora princeps, 

 Pepo, and Rhipsalis salicornioides). In the last-named plant I ob- 

 served all the transitional stages between free cytoblasts and the perfect 

 formation of the special parent-cells (or the pollen-cells ?) ; in other 

 plants I have not yet succeeded in such complete observation. Finally, 

 I must observe that the entire assumption of special parent-cells appears 

 to me questionable. It is no matter of doubt that in the stage near ma- 

 turation, each pollen-cell appears surrounded by a tolerably thick gela- 



* Contributions to the History of Development of the Pollen of the Phanerogamia 

 (Zur Entwick. des Pollens, &c.). Zurich, 1842. 



f Nageli, loc. cit. pp. 20, 21. j Meyen, Physiologic, vol. iii. p. 186. 



