* 
Scutellum, yet there are many grasses in which it reaches a considerable 
16 E 3 
the Graminew. Unlike the scutellum, the epiblast has no vaseular 4 
system, and on this account much dispute has arisen over its morpho- 
logical nature. This, however, is no proof against its leaf nature, as 
there oeeur many instances of reduced organs without fibrovaseular _ 
systems. E 
Although the epiblast is usually very small in comparison with the a 
. 
size. Zizania aquatica (fig. 22, Pl III) has a well-marked epiblast 
extending from the base of the lengthened axis to about the middle of 
the plumule, or even further. E 
In the genus Stipa there are marked differences in the size of the — 
epiblast in the different species. S. tenacissima (fig. 30, Pl. IV) has a 
very small one; in S. pennata it is long and attenuated; while in 8. ie 
richardsonii (fig. 36, Pl. V) and S. viridula (figs. 32 and 33, Pl. IV) it is 
very large and broad. Sometimes the margin of the epiblast is curved e | 
or deeply keeled, although in the large majority it is Straight. Homa  . 
locenchrus oryzoides ( fig. 28, Pl. IV) has a very large epiblast which — 
greatly resembles the form of the scutellum. 1t is large and broad at 
the base and has a shield.like appearance. A small protuberance 
occurs at the base similar to that in the scutellum. E 3 
A fact which seems to indicate that the epiblast has the same mor- 
phological value as the seutellum is, that it is inserted on the axis 
almost at the same height, Even in Zizania (fig. 22, Pl. III), where the 
axis elongates to a great length and the insertion of the scutellum is 
almost at the base of the embryo, the epiblast is inserted directly 
opposite. : 
The absence or presence of the epiblast appears to be fairly constant 
in the different tribes, and the presence of an epiblast in a tribe where 
the majority of the genera have none may indieate that it is not in its 
natural position. The tribes Andropogonew and Maydee, to conclu 
from the genera examined, might be regarded as being without a 
epiblast. The scutellum almost entirely surrounds the embryo, as may 
observed in any of the cross sections figured, more especially amon, 
the Maydew. In the tribe Zoysiee there occur genera without an 
epiblast, as Nazia, Trachys, and Anthephora, while Zoysia has a dis- 
tinct one. The Tristeginew, to conclude from Beckera and Arundine 
have no epiblasts. The Panicee appear to be without an epiblast, 
only exception being Olyra, which has a very large epiblast coverin; 
the plumule. From its great resemblance to the embryos of Leersia 
and Oryza it seems probable that it should be placed with them, 0 
Should form a Separate intermediate tribe. The Oryzec appear to have 
a very large epiblast, The Phalaridew, so far as examined, vary. 
Anthozanthum and Phalaris having small epiblasts, while Ehrhart 
has none. 
a 
The Agrostidew examined, although varying greatly in their external 
form, all possess an epiblast, In the large majority it is small, but in 
