FISHER: SEED DEVELOPMENT IN THE GENUS PEPEROMIA 149 
In Coffea arabica, Hanausek (28) found actual polyembryony, 
as he called it, in which two to three embryos were formed in 
branches of the nucellus, one embryo developing in each branch. 
The abnormal branching of the nucellus in Coffea resembles that 
in P. verticillata, although the storage material in the former is 
endosperm, while in the latter it is chiefly perisperm. 
This case of Coffea, that of Taraxacum, mentioned above, 
and other similar cases have been classed under pseudopoly- 
embryony by later writers, but since they are cases of the 
occurrence of more than one embryo in a single seed, no good 
treason is apparent for calling the phenomenon false polyembryony. 
It is true that these cases, in which more than one embryo arises 
in one nucellus—each in a separate embryo sac—are quite dif- 
ferent from the ordinary cases of polyembryony, but they do not 
differ from them in the essential character of polyembryony, 
that is, the initiation of more than one embryo in one ovule. 
Two ovules, each surrounded by a separate inner integu- 
ment, each containing an embryo sac, and both inside a common 
outer integument, were described and figured by Schacht (60, 
bl. 3, f. 18) in Orchis Morio. Braun (3) was of the opinion that 
these twin ovules, as he called them, arose by division of the 
nucellus, rather than by a growing together of two nucelli. He was 
led to this view by the appearance inside of the common outer 
integument of two entirely separate inner integuments. 
In Monotropa Hypopitys, Schacht (60, p. 40) found a quite 
similar case, in which two embryo sacs filled with endosperm were 
Surrounded by a single integument. 
The case of Orchis latifolia, described and figured by Schleiden 
(62, p. 599, pl. 6, f. 2), is, as suggested by Strasburger (72), 
doubtless to be interpreted in the same way as Schacht’s twin 
ovules in Orchis Morio (60). 
Strasburger (72) described a case, similar to that found in 
Orchis Morio, in a closely related orchid, Gymnadenia conopsea, in 
which each of the two embryo sacs already contained an embryo. 
Luerssen (43, p. 296) mentioned a similar case, which he found 
in Iris sibirica, 
While the abnormally branched or lobed ovules in Peperomia 
verticillata, Monotropa, Morus, Coffea, Iris, Gymnadenia, and 
