1898 | EMBRYOLOGY OF ALYSSUM 321 
more regularity than in that of any other dicotyledonous embryo 
yet studied. Capsella is undoubtedly quite symmetrical and 
Alyssum comes very close to it in many respects. I regret that it 
was not possible for me to make comparisons directly with Han- 
stein’s text and figures, as there is so much variation in the illus- 
trations and reprints given in text-books. 
In Vines’s Text-book of Botany, p. 443, the terminal embryo 
cell is figured as dividing first transversely, then longitudinally, 
while the text reverses this. In Goebel’s Outlines, p- 397, the 
figures represent the first division as longitudinal, the second as 
transverse, while the text gives the first and second as longitu- 
dinal and the transverse divisions as a third series. This corre- 
sponds to Alyssum. In Sachs’s Tezt-book, p. 516, 1875, the figure 
is the same as in Goebel’s, but in the text the third series of 
divisions is given as tangential and cutting off the dermatogen 
from the quadrants. Whether this statement is due to faulty 
translation or is so in the original German edition, I cannot say. 
It does not agree, however, with the German edition of 1882, 
for there it is distinctly stated in the text that the first three 
series of divisions are in three directions at right angles to each 
other, although in the explanation of fig. 446, J-IV, he gives 
the second series of divisions as transverse, and does not figure 
Or mention the second longitudinal division. 
Confusion likewise exists with regard to that portion of the 
embryo of Capsella which Hanstein designates as the ‘ hypoph- 
ysis.”’ Its origin is uncertain. It is thought to arise from 
the last division of the suspensor cell; 2. ¢., the cell which gives 
rise to the greater part of the embryo remains dormant after the 
first division in the proembryo, while the suspensor cell continues 
to divide. The last cell arising thus contributes to the embryo. 
Chamberlain® doubts the accuracy of this theory, but thinks it 
Probable that the terminal cell in which the first longitudinal 
division appears is wholly embryonic, while there may be - 
varying number of cells in the suspensors formed before this 
division occurs. This agrees with Alyssum. Hanstein’s 
* Bor. Gaz. 23: 147-1709. 1897. 
