Johnson. — On A rceuthobium Oxycefri. 147 
two division-stages, gives a mature embryo-sac. If we think 
of the papilla in A. Oxycedri as an ovule, we have an arche- 
sporium which is apical, hypodermal and multicellular, from 
which two embryo-sac mother-cells develope into ripe embryo- 
sacs, the cells between them being undeveloped sporogenous 
tissue. If this view be correct, we have in this parasite an 
ovule which is more highly developed than in any other 
Phanerogam. It would too be quite opposed to the basis of 
the distinction of the Loranthaceae from the Santalaceae. The 
absence of a distinct ovule in the Loranthaceae is the essential 
ordinal character of separation from the Santalaceae. To 
regard the papilla in A. Oxycedri as an ovule would be to 
give to the most modified genus of the Loranthaceae a struc- 
ture on the absence of which the separation of the Lorantha- 
ceae from the Santalaceae is based. It appears to me to 
be impossible to regard it as an ovule, and yet, as in the 
case of Myzodendron , to exclude it from Santalaceae. 
I hoped that a consideration of the relative time at which 
the papilla in the different Loranthaceae appears would throw 
some light on its morphological value. Hofmeister in Loran- 
thus enropaeus and in Viscnm album, and Treub in Loranthus 
sphaerocarpus , found that the ovarian papilla appeared after 
the carpels as an upgrowth from the floor of the ovary ; and of 
L. sphaerocarpus it is recorded that the segments appear 
later, followed by the formation in their interior of their 
respective embryo-sacs. Whether the placenta in L. sphaero- 
carpus should be regarded as formed by the fusion of ‘ claws ’ 
of the carpels, as in Santalum album and Primulaceae, or 
as a derivative of the apical part of the floral axis, does 
not appear. 
The time of appearance of the papilla is different in A. 
Oxycedri . Baillon found that the apical part of the floral 
axis persisted throughout the development of the female 
flower, changing from a hemispherical to a conical swelling in 
becoming the ovule. It should however be mentioned that 
the embryo-sacs were overlooked. While in the case of this 
plant a knowledge of the time and mode of development of 
