THE NECTAR GLAND IN CACTI 147 



wall of the epidermis especially, at the angles of the contiguous 

 cells and that this may extend from said angles into the inner 

 periclinal walls. There results from this condition a more or less 

 complete network of irregularly cutinized strands (fig 5) marking 

 the outlines of the inner periclinal walls of the epidermal cells. 

 The non-cutinized cell walls are, so to speak, supported within 

 this curious framework of cuticle, which, moreover, may be laid 

 down throughout the whole of the cell wall, especially in those 

 radial walls which lie longitudinally with respect to the axis of the 

 gland as a whole. But, on the other hand, it is frequently dis- 

 continuous, so that a short portion of the radial wall, even at the 

 angles of contiguous cells, remains uncutinized (figs. 3, 5). A 

 brief study of our figures will serve to make these relations clear. 

 It will be seen, also, that of the total thickness of the outer wall 

 of the epidermis, the cutinized portion is far in excess of the cellu- 

 losic part. 



The beginning of secretion is evidenced in the eruption of the 

 whole outer cell wall, the radial, and sometimes even the inner walls 

 being torn away, leaving their cutinized portions attached to the 

 top of the gland (fig. 3). Preparations treated with alkanna bring 

 out these facts unequivocally. 



Meanwhile, there is a concomitant disintegration of the remain- 

 ing cellulose walls and of the protoplasm, in which a small amount 

 of oil appears as minute droplets (fig. 3). On the surface of a 

 gland of which the whole apical area is freed from the capping 

 epidermis there can eventually be found little or no trace of either 

 the cellulose of the cell walls or of the protoplasm, more, however, 

 toward the periphery of the exposed glandular area than toward 

 the center. Some glandular material, strands of inner cuticle 

 (fig. 5) and masses of calcium oxalate intermixed adhere, however, 

 in an irregular, amorphous layer. As the gland ages, there is a 

 further accumulation of this salt in the upper part. In the base of 

 the gland, spheroids of calcium oxalate are in evidence, sometimes 

 associated in the same cell with rhomboids. In fact, the whole 

 appearance of the upper portion of the gland indicates that there 

 has occurred an autolysis of the cellulose walls and of the proto- 

 plasts of the cells involved. That sugars occur is shown by a 



