THE EVOLUTION OF PLANTS 



205 



bearing pollen-sacs, and the megasporophylls, bearing, 

 not merely megasporangia, but true seeds. The ovule has 

 a pollen-chamber, like the cycads, except that it projects a 

 bit through the micropyle, and, strange as it may seem, 

 fossil pollen-grains have been discovered, well preserved 

 within this chamber. The seeds, about ^<t inch long, 

 have been described as resembling little acorns, enclosed 

 like hazelnuts in smaller glandular cupules (Fig. 89). 

 They are similar to those of the cycads, except that they 

 are not known to have organized an embryo with cotyle- 



FIG. 89. Restoration of a seed of Lyginodendron oldhamium (Lagenos- 

 tema Lomaxi), from a model by H. E. Smedley. (After Scott.) 



dons and caulicle. Instead, the tissues of the female 

 gametophyte only are so far found, retained within 

 the megasporangium, which is enclosed in the integument. 

 In this connection it is of interest to note that the seeds of 

 some modern plants (e.g., orchids) do not possess differ- 

 entiated embryos, but whether this is a primitive or a 

 reduced character is not certain. The pollen was formed 

 in spindle-shaped pollen-sacs, having two chambers, and 

 borne in clusters of four to six on the under side of little 

 oval discs, from 2 to 3 millimeters long. These structures 



