1915] BURLINGAME—ARAUCARIA BRASILIENSIS 21 
which do not closely invest the nucellus. The female gametophyte 
is covered by a very thin layer of nucellar tissue above. The pollen 
chamber occupies almost all of the exposed portion of the nucellus 
and probably laid bare the gametophyte at its maturity so that 
the free-swimming (probably) sperms had direct access to the 
archegonia. 
One may suppose that when pollination first began the nucellus 
was freely exposed, and that the integument was either wanting 
or less developed than in Physostoma. Since these seeds were 
freely exposed on leaflike organs, there must have been developed, 
as the first necessary step to pollination, a sticky secretion on the 
nucellus to catch the microspores or pollen grains. It must be 
further supposed that the pollen grain was able to secure sufficient 
food from this secretion to maintain itself for such a length of time 
as was necessary for its further development, and until the game- 
tophyte had broken through the nucellus and exposed the matured 
archegonia. It is supposable that the processes that produced the 
sticky secretion might in the course of time develop the habit of 
further destroying the cells of the tip of the nucellus to produce 
a rudimentary pollen chamber. The further step of eroding this 
chamber deep enough to allow access to the archegonia without 
waiting for the growing female gametophyte to rupture the nucellus 
would appear to be easy and logical. 
During the development of the pollen chamber the integuments 
would be developing in the direction of greater efficiency in securing 
the deposition of the pollen on the tip of the nucellus. As they 
closed up the sticky secretion would be exuded as a pollination 
drop to catch the pollen. If the ovules stood upright gravity 
would effect the delivery of the pollen to the pollen chamber. In 
any case, the pollen would probably be retracted along with the 
pollen drop when it began to dry up. It is in this stage of develop- 
ment that the seeds of Cycadofilicales are found fossil. The 
reason (it appears to me) is that seeds that had been fertilized (or 
were far enough along to be fertilized soon) fell to the ground and 
continued their growth. If this were the case one would expect 
to find fossil only those seeds that had fallen too soon to be able to 
continue growth. 
