36 BOTANICAL GAZETTE [JULY 
lus (fig. 23). It is certain that neck cells are formed in the former 
(1), although they are very evanescent and much more difficult to 
demonstrate than in the archegonia of Pinus, for example; but they: 
have not been demonstrated in the deep-seated archegonia (2), 
although it is not unlikely that this is merely due to the difficulty of 
differentiating them from other cells of the prothallus. In a recently 
published account of Juniperus (OTTLEY 9), a case is described and 
figured in which a small deep-seated archegonium occurred outside 
the normal group. The figure leads one to suppose that no neck 
cell could be demonstrated, but no statement on this point is made 
in the text. After fertilization no neck cells can be found in either 
type of archegonium, which accounts for the suggestion made in my 
former paper (10) that neck cells are never formed in Widdringtonia. 
The neck, where present, consists of a single tier of four cells (fig. 23). 
The central nucleus of the archegonium divides (in the first type 
of archegonium) simultaneously in the whole group (fig. 24), a con- 
spicuous spindle being organized with long, slender, and considerably 
twisted chromosomes (fig. 24). The reduced number of chromosomes 
(six) may be counted fairly readily in this division. It is to be noticed 
in fig. 23 that although the central nucleus of every archegonium in 
the basal group is dividing, yet the nuclei of the other archegonia are 
still in the resting condition. The evidence that these do not divide 
is purely negative, but the possibility of the deep-seated archegonia 
being really archegonium initials only, must not be overlooked. 
The case shown in fig. 16 of my former paper (10) indicated that 
rarely a persistent ventral canal nucleus is formed, but out of scores 
of prothalli examined, containing thousands of archegonia, this is 
the only case in which such a state of things was clearly indicated. In 
no other preparation has the ventral canal nucleus been certainly 
identified, although the occurrence of the spindle which initiates its 
formation furnishes additional evidence that it must be formed. 
Doubtless it is normally very evanescent. From STRASBURGER’S 
(12) figures of the same nucleus in Juniperus, it would seem that it 
is by no means a conspicuous object in that genus, and although 
Lawson (5) figures a large ventral canal nucleus in Thuja, it evidently 
disappears quickly, as it is not figured in the mature archegonium, 
and the same was found in OrTLEy’s recent paper on Juniperus (9)- 
