118 BOTANICAL GAZETTE [FEBRUARY 
develop in the same gametangium (fig. 40), which combines the 
form, structure, and mode of development of the two sex organs, 
certainly shows a close relationship between the organs which 
ordinarily produce egg and sperms separately. We have seen 
how close is the relationship between the members of the axial 
row. The central cell and the primary canal initial are sister 
cells; the division of the former produces two sister cells— 
egg and ventral canal cell—and the division of the latter 
also produces two sister cells—canal cells. We find, further, 
that the second pair may with other canal cells divide longitu- 
dinally and produce two canal rows (figs. 35-37). An organ of 
this nature, with its egg and ventral canal cell, with its double- 
walled venter and single walled neck, must still be looked upon 
as an archegonium, even if it has deviated slightly from the type. 
But when the canal members break up by successive divisions 
until we have an organ such as that shown in fig. go, where the 
canal row is lost in sperm mother-cells; or that in fig. 45, where 
the egg is lost, but other members of the canal series are retained 
as a row, we can no longer regard the organ as an archegonium, 
but must look upon the first (fg. 40) as a bisexual organ and the 
second (jig. 45) as a modified antheridium. It will be seen that 
these three forms—modified archegonium (jig. 37), bisexual 
organ (fig. go), and modified antheridium (jig. 45)—constitute 
a progressive series leading from the normal archegonium to the 
normal antheridium. 
In addition to the abnormal forms already mentioned, two 
important cases of the apparent conversion of archegonia into 
antheridia have been reported from different parts of Europe by 
Lindberg (11) and de Bergevin (20). In these cases, which are 
quite similar, a radical change in the life-history of the organs 
appear. Each organ begins its development with the characters 
of an archegonium, but closes it bearing the evident characters 
of an antheridium., De Bergevin gives no sections of the organs, 
but only a series of external views representing a normal anther- 
idium, an archegonium, and various club-shaped organs leading 
up to a second antheridium more than twice the length of the 
first. He speaks of the process as an interconversion of the s¢* 
