Campbell ’ — The Embryo-Sac of Pep eroniia. 113 
Theoretical Considerations. 
The writer has already expressed his opinion that in 
Peperomia we have to do with the most primitive type of 
Angiosperm yet described. Johnson is inclined to differ 
from this conclusion, believing that the peculiarities are to 
be explained as secondary modifications. His reasoning is 
not, however, quite convincing. He says 1 , ‘I am inclined 
to believe that the peculiarities of the embryo-sac of Peperomia 
have been secondarily acquired, and are analogous to those 
found in Angiosperms of peculiar habit, e. g. many aquatic, 
parasitic and saprophytic forms.’ Now the habit of Peperomia , 
so far as the writer is aware, conforms to none of the above 
categories, and is that of any normal green plant. Moreover 
the peculiarities of the embryo-sac are not reductions — unless 
we consider the embryo, which is not the question here ; but 
consist rather in an increase in the number of parts, for 
which there is no parallel, so far as the writer is aware, as 
a result of the aforesaid peculiar habits. 
Piper and Heckeria are undoubtedly related to Peperomia ; 
but Saururns certainly cannot be considered «• a very near 
relative, as Engler thinks it sufficiently distinct from the 
Piperaceae to warrant the establishment of a separate family. 
Moreover the flowers, with their syncarpous gynoecium, are 
certainly much more specialized than the exceedingly simple 
unicarpellate flower of Peperomia . Just why the flower in 
Peperomia should be considered as a reduced form is not 
clear. From a study of many low types among the Mono- 
cotyledons, which cannot be readily derived from higher 
types, e. g. Naias , many Araceae, it seems more reasonable 
to consider the single carpel, with a single axial ovule, as 
the primitive type for the Angiosperms, and with this 
Peperomia agrees perfectly. 
While Johnson states that Piper and Heckeria have 
‘ essentially typical ’ Angiospermous embryo-sacs, he gives 
no details or figures, so that it is impossible to judge whether 
1 loc. cit., p. 9. 
I 
