358 The Botanical Gazette. [November, 
EXPLANATION OF PLates XIX anp XX. 
(All figures on Plate XIX are magnified 450 diameters; all on Plate XX 
600 diameters), 
; the Tryo-sac. 
14, a further development of fig. 13. Fig. 15, the mother-cell divided once. 
Fig. 16, the cells divided again. Fig. 17a, the true mother-cell of the embryo- 
n 
» 
2 
: 
> 
o 
3 
oO 
™ 
2 
a 
a 
oO 
oO 
fe} 
a 
hen 
n 
ag 
Q 
° 
8 
: 
B 
gq 
ee 
gn 
° 
mt 
oQ 
2 
N 
Q 
[ory 
8 
s a : ; 
ment of Fig. 18; the nucellus almost gone and the appearance of vacuoles. 
From fig. 20 to fig. 23, inclusive, are shown the division of the mother-cell 
and its further divisions, culminating in the formation of the egg-apparatus, 
the antipodal cells and the endosperm nucleus; the vacuoles and the expat- 
A study of some anatomical characters of North-Americal 
Graminex. IY. 
THEO. HOLM. 
‘The genus Leersia. 
(WITH PLATE XXI.) 
In previously published papers! the anatomical oe 
of Uniola, Distichlis and Pleuropogon have been pacers 
and it is the purpose of this, and a following be tee eee na- 
ow the species of Leersia may be distinguished 4 
tomically. 
It would, of course, have been more proper as done 
from Uniola to the genera allied to it. This was 00° 
when the comparison was drawn between Uniola, ial_has 
and Pleuropogon; but the lack of sufficient ae 
necessitated a change in the order of treatment. idered 
groups, at least, of closely related genera may be yo 
t once, so as togive a broader view of their ana 
divergencies. 
proceed 
‘Botanical Gazette, June, August and October, 1891. 
