20 Campbell. — Studies on the Araceae. 
In Pistia, which so far is the only Aroid whose embryology 
has been studied at all completely, formation of quadrants in 
the young embryo is associated with a lateral formation of 
the root like that in Lysicliiton , and is very suggestive of the 
formation of the root in Isoetes ; indeed the early segmenta¬ 
tion of the embryo, as well as the later structure, is very 
much alike in Pistia and Isoetes. 
Summary and Conclusion. 
1. In Dieffenbachia and Aglaonema there is no question as 
to the axial origin of the ovule ; this is probably the case 
also in Lysichiton , but not so certainly in Anthurium. 
2 . In all of the forms examined there is a large develop¬ 
ment of the base of the ovule, and in Aglaonema and Lysi¬ 
chiton of the integuments also, so that the embryo-sac does 
not occupy as much of the ovule as is usually the case. 
3. Up to the time of fertilization, the embryo-sac in the 
forms examined (except possibly Aglaonema) shows the 
ordinary type of development. 
4. The endosperm in all the forms examined very early 
forms a continuous tissue, completely filling the embryo-sac. 
5. The antipodal-cells in Lysichiton become very large, 
and undergo secondary divisions, so that a large group of 
cells results; in Aglaonema a similar group of cells was seen, 
but its development could not be traced; the antipodal-cells 
could not be certainly distinguished in the older embryo-sacs 
of Anthurium. 
6. There are probably two types of segmentation of the 
young embryo; in the first there are formed two transverse 
divisions before any longitudinal walls appear; in the second 
there is a quadrant-formation, suggestive of the early 
divisions in the Fern-embryo; the differentiation of the 
organs takes place at a late period, and it is not possible 
to trace them back with certainty to the primary divisions 
of the embryo. 
7. The cotyledon is very large, the stem and root much 
