ANGIOSPERMS. 363 



form a band of several cells, concave towards the inside, in each of the four pro- 

 tuberances (Fig. 286 I?), but in some cases the transverse section shows only one 

 archesporial cell in each protuberance, and there is therefore only a longitudinal row 

 of these cells in each ; this is the case in the Compositae and Malvaceae (Fig. 286 C) ; 

 still more rarely only single archesporial cells are formed, as in the Mimoseae. There 

 is usually but little division of the cells of the archesporium before the formation of 

 the tetrads ; consequently the number of the pollen-mother-cells is usually not much 

 greater than that of the cells of the archesporium ; but it does sometimes happen 

 that the original simple layer or row of archesporial cells divides in all directions and 

 produces several layers or a cylindrical mass of mother-cells (Fig. 286 E). The 

 layer of cells (primary tapetal layer 1 } formed, as was stated above, by the longitudinal 

 division of the layer from which also the layer of primary mother-cells is derived, and 

 lying between the latter and the epidermis, divides according to Warming usually 

 into three layers, in which radial, horizontal and tangential walls are formed alter- 

 nately. The innermost of these three layers (Fig. 288 A ep, Fig. 287 B n), supple- 

 mented by a corresponding layer on the side of the group of mother-cells next the 

 axis of the anther, developes into the iapetum, which becomes disorganised as the 

 sporangium matures, as in many of the Vascular Cryptogams. The same fate befalls 

 the cells of the next outer layer, which do not however assume the glandular appear- 

 ance of the tapetal cells. The outermost of the three layers, which lies therefore 

 immediately beneath the epidermis of the loculament, forms the layer of fibrous cells 

 which causes the dehiscence of the anther (Fig. 299 G, /3), and to which we shall 

 return again a little later. The large pollen-mother-cells (Fig. 288 A, sm) are at 

 first thin-walled ; but their walls become considerably and in most cases unequally 

 thickened (Fig. 288 ), and the thickening material is usually distinctly stratified. In 

 many Monocotyledons the mother-cells now become completely isolated, the loculament 

 enlarges, and the cells float singly or in groups in a granular fluid which fills the cavity 

 (Fig. 288-^), a circumstance which at once recalls the formation of spores in the Vas- 

 cular Cryptogams. But in many Dicotyledons, as Tropaeolum, Althaea , and other genera, 

 the very thick-walled mother-cells do not become isolated but completely fill the locu- 

 lament, and usually separate after the disruption of the wall of the anther in water. 



As the cell-wall of the pollen-mother-cell becomes thickened, the protoplasmic 

 body rounds itself off, and the large central nucleus divides when the preparation for 

 the formation of pollen-grains (microspores) commences. The latter process 2 presents 

 two modifications. In one, which is the most frequent in the Monocotyledons, the 

 division of the nucleus of the pollen-mother-cell is followed by the formation of a wall 

 between the two daughter-cells, so that the mother-cell is now divided into two cells 



1 [' Primary tapetal cell,' ' primary tapetal layer,' are adopted as renderings of ' Schichtzelle ' 

 and ' Schicht zwischen Archespor und Epidermis,' that is, the cell or layer cut off from the periblem- 

 cell or layer, the other portion of which forms the archesporium. A distinctive term for this cell or 

 layer is wanted, and the one here adopted appears suitable, implying as it does that the cell or layer 

 is that from which the tapetum is derived; the other cells or layers formed from it may be 

 distinguished as ' parietal cells,' ' parietal layers.' In the ovule the primary tapetal cell is Stras- 

 burger's ' tapetal cell ' (see page 386). The occurrence of this cell or layer has been already referred 

 to in Lycopodineae and Gymnosperms (see pp. 294, 332).] 



2 Strasburger, Zellbildung u. Zelltheilung, III. Ed., Jena, 1880, p. 130. 



