SVEDELIUS: LIFE-HISTORY OF 
observation at the Nicobar Islands. The currents near 
Paumben and Jaffna just where this plant lives are in fact 
very strong. According to Walther (34) the current in the 
Paumben channel often reaches a speed of 5-6 knots in an 
hour. 
The ovules are anatropous with two integuments (fig. 7). 
These are both fully developed, whereas, according to 
Muller (21), in Vallisneria the outer integument after some 
time ceases to grow, so that only half the length of the 
nucellus is provided with two integuments, and thus the 
micropyle is formed by one integument only. 
I have not been able to investigate the development of the 
embryo sac and the fertilization in my material, owing to 
the difficulty of making the fixing solution penetrate the 
copious mucilage in the young ovaries. 
In the embryo sac the antipodals are rather large. (PI. 
., A, fig. 13). Sometimes the embryo sac has a pouch- 
like deepening in its antipodal end, though not so large as 
m E odea (cf. Wylie, 35). The cells of the tissue of the 
nuce us are here but little different from the rest. They 
smaller, but their walls are thicker and stain better. 
From these smaller cells some larger ones radiate in a semi¬ 
circle, probably because they have some nutritive function in 
connecfion with the antipodal cells. The fertilization 
i sell have not observed, but have often seen pollen tubes 
embry ° snepensor cell (H. 
■’ A ’ flg - ,2 ) jttfrt as in Elodea (Wylie, 35). The 
embryo seems otherwise to be of the usual mo ocotyledonous 
type. Very few endosperm cells are fon^dTn "the'embryo 
embry ° 18 S ' l8pended as « in a 
dnr ta« the farther development of the 
simnlT.tr * ‘ ntegnmente do not develop any more, but 
Vo lionifi 80 nm<dl that *bey eqnal the growing embryo. 
° hgmfia ‘ kl0n place in their tisane, and when the fruit 
