218 BOTANICAL GAZETTE [MARCH 
obliquely (figs. 75, 78, 79). In cases it is vertical or oblique, — 
the growing point and the cotyledon both arise from terminal 
segments, agreeing with Zannichellia in this particular, but dif- 
- fering from Naias. After having examined a very large amount 
of material, I have come to the conclusion that there is no regu: 
lar order of division in the active nuclei of the young embryo 
at least after the first two walls are formed (figs. 15-27), and 
have thought that where descriptions do not agree it should per- 
haps be ascribed to the variation of development rather than to 
any incorrect statements or observations of the earlier writers. 
In figs. 24 and 25 there are four transverse walls formed; in fig. : 
25 there is no sign of division of the terminal cell; while in fig. 
_ 24 the terminal cell is in process of division. In fig. 16 there 
are three transverse walls, the terminal cell being divided vert 
ae cally, while the cell next to the suspensor is in nhc divi- 
: ‘Sion. 
. In the further development it was im 
ae tider of the divisions, and in the older stages of the em . 
seemed: that the growing point was of lateral in alwa 
(Jigs. 22, 28, 29, 30). The mature embryo is U- shaped ( sees 
and the growing point is in the basal quarter of the emp 
a spntieasg wholly surrounded by the epidermal layer. ae 
__ This work was done at the Ames Botanical La 
orth Easton, Mass., under the direction of Oak 
“Norra Cammnipce, Mass. ae 
a LITERATURE CITED. 
ae Becatar, F.: Butomaceae. Flora — : 504. 
CAMPBELL, D. H.: The morphology of Naias aiid a Zannic 
Acad, Sci. III. 1: 1-62. 1897. a 
Jermey, E. C.: Polyembryony in Erythronium amerit 
2 — 9: 537-541. 1895. | 
fAF - H.: The embryo-sac of A/isma Plantage. 
possible to follow the 
heli. 
H. Mar RSHALL: A enim to our Lcueld “ 
= Linn. ot Bot. 17: i 
