DEVELOPMENT OF SEED IN POLYGONACEAE 
and become quite fleshy. The increase in size crushes the depleted 
starch cells and when the germination has reached the stage shown 
in figure 29, only the crushed remains of the endosperm and aleurone 
layer are present in the seeds. By this time the plumule, P, has 
increased in size, having attained a length of about i mm. and a width 
at its base of about .06 mm. It is conical in shape, the apex being 
slightly rounded. No multicellular hair-like structures are present 
about the plumule as in the preceding examples. Very soon after 
this the broadening cotyledons push off the seed remains, and become 
oriented in a nearly horizontal plane. The differentiation of the coty- 
ledons into stalk and blade is only slightly evident at this time. 
Very little differentiation has occurred in the tissues of the cotyledons 
with the exception of the epidermis which has now become well 
developed. 
Polygonum scandens 
The ungerminated seed is very similar to that of P. Convolvulus 
already described in this paper. In following out the germination, 
conditions are found to be practically the same as described above for 
Rumex crispus, so a detailed discussion of the processes involved 
need not be entered into. A comparison of figures 23 and 24, repre- 
senting certain stages in the germination of Rumex crispus, \\^ih figures 
30 and 31, representing corresponding stages in P. scandens, shows how 
very similar in appearance are the methods of germination in the 
two species. 
Systematic Relation of the Polygonaceae 
Bentham and Hooker (i), include under the Curvembryeae, the 
Nyctaginaceae, Illecebraceae, Amaranthaceae, Chenopodiaceae, Phy- 
tolaccaceae, Batidae, and Polygonaceae. In these families the authors 
speak of the storage tissue outside the embryo as "albumen," without 
stating whether this tissue is endosperm or perisperm. 
The writer sees no reason to doubt that Dammer, writing on the 
Polygonaceae in Engler and Prantl (6), has properly taken out the 
family from the Curvembryeae of Bentham and Hooker and placed 
it in a separate group termed Polygonales. Harz (7) agrees with 
Bentham and Hooker in placing the Polygonaceae in the Curvem- 
bryeae, and looks upon the storage tissue outside the embryo as 
perisperm rather than endosperm. This misinterpretation of the 
structure in the mature seed doubtless led him to place the Poly- 
