1903] EMBRYO SAC OF CASUARINA 103 
origin, near the micropylar end, is itself the egg, or develops the 
sex apparatus of one to four cells, usually three, one of which is 
the egg. The others (synergids), it is thought, are homologous 
with the neck cells of gymnosperms and pteridophytes. Treub 
draws this conclusion from their origin from a single micropylar 
cell with cellulose wall, and from their appearance. There are 
no antipodals. 
This was the first case of chalazogamy reported. The passage 
through the chalaza is facilitated by the presence of the pro- 
longations of the sterile megaspores, into which the tube enters, 
leaving them to reach the egg apparatus that is to be fertilized. 
It is stated, however, that the tube never enters the fertilized 
sac, but becomes attached to it at some point between the egg 
apparatus and the antipodal region. Sometimes a nucleus was 
observed in the tube, but never more than one. 
The egg remained unchanged for some time before its fertili- 
zation. The exact moment of fertilization was not determined, 
but before it occurred the formation of a large number of endo- 
sperm nuclei had taken place, even as many as fifty. This con- 
clusion was based on the fact that sacs were found which formed 
endosperm before the pollen tubes reached them. 
Casuarina thus shows a megasporic sporogenous tissue remind- 
ing one of the pteridophytes, a number of embryo sacs suggesting 
gymnosperms, and “synergids’’ suggesting the canal cells of 
gymnosperms ; and it was unique in the route of the pollen tube. 
These considerations led Treub to believe that the Casuarinaceae 
are intermediate between angiosperms and gymnosperms. 
The species that I studied was Casuarina stricta Ait. The 
material was collected in California from introduced trees by 
Mr. Albert C. Herre and identified by him. It was killed in 
chrom-acetic acid; but unfortunately the pieces were large, and 
hence the killing was not rapid enough to catch cells in mitosis. 
Only the pistillate cones were secured. The ovaries were care- 
fully teased out with needles and sectioned in paraffin. Longi- 
tudinal sections 5 # thick and stained with safranin and gentian- 
violet were found to be the most satisfactory. The work was 
almost all done at night. Light was obtained from a gas-lamp 
