158 
AMERICAN JOURNAL OF BOTANY 
[Vol. io, 
was nine. It is conceivable that the last division of the neck-canal cells was 
not shown by his material. 
Janczewski reported also that archegonial development in Fossombronia 
pusilla differs from that in Pellia only in the smaller number of neck-canal 
cells, the central cell dividing when there are four neck-canal cells which 
later increase to eight (probably by a division of each of the four). This 
author reported also that “ Jungermannia [. Lophozia] excisa and Radula com- 
planata agree exactly with Fossombronia.” 
However, he found conditions somewhat different in the Marchantiales. 
In Riccia Bischoffii the axial cell, after the cap cell has been cut off, divides 
to form the central cell and a neck-canal mother cell. The latter by two 
successive divisions gives rise to four neck-canal cells, and the former divides, 
just before fertilization, into a large egg and a small ventral canal cell. In 
Preissia commutata, Marchantia polymorpha, Reboulia hemisphaerica, 
Lunularia vulgaris, and Plagiochasma Rousselianum the process is reported 
to be similar except that in Marchantia the division of the central cell occurs 
somewhat earlier, sometimes even before the division of the penultimate 
cells of the neck-canal row. 
Garber (1904) has followed in detail the development of the archegonium 
of Riccia natans and confirms the observations of Janczewski, adding that 
the central cell grows rapidly during the period before its division. 
The detailed account of archegonial development in Marchantia poly¬ 
morpha given by Durand (1908) confirms all points reported by Janczewski 
and adds that the ultimate neck-canal cells sometimes show a belated 
nuclear division which is not followed by a division of the cytoplast. Haupt 
(1921) confirms Janczewski as to the manner of the division of the central 
cell in Reboulia and as to the number of neck-canal cells present at the time 
of this division. He finds, however, that these cells (four in number) are 
later increased to eighteen or twenty. 
The more modern studies on members of the Jungermanniales reveal 
probably a greater variety of conditions than Janczewski suspected, though 
none of them give the details of the development of the axial row. Hum¬ 
phrey’s (1906) work on Fossombronia longiseta, although only fragmentary 
as concerns the archegonium, seems not to accord with the former author’s 
observations on Fossombronia pusilla, or with the conditions found in 
Riccardia pinguis. The work of Haupt (1920) on Fossombronia cristula 
does not include the development of the axial row in sufficient detail to 
admit of a comparison. His earlier work (1918) on Pallavicinia Lyellii 
is more satisfactory in respect to the archegonium. In this species the 
axial cell is said to divide into a primary neck-canal cell and a primary 
ventral cell. 
The development of the axial row [neck-canal row] usually precedes the division of 
the primary ventral cell, although frequently mitoses can be seen in the neck cells after the 
formation of the ventral canal cell and egg. In most cases about ten neck-canal cells were 
