94 
Humphrey . — The Development of 
description of F. pusilla , states that in this species the development of the 
antheridium does not in any way differ from that of the normal type for 
the Jungermanniaceae. Thus it seems that Fossombronia longiseta forms 
a connecting link between such forms as Sphaerocarpns and Anenra . The 
discharge of spermatozoids was observed under the microscope, and was 
found to resemble closely that occurring in Porella L The cells composing 
the antheridium-wall are quite hygroscopic, and within a few seconds 
following the addition of water, those at the top of the antheridium, which 
appear to be somewhat larger than the others, separate and roll outward, 
followed in rather rapid succession by the cells below (Fig. 13 a , PL V). 
Many of the cells are several times longer than broad, and when distended 
become markedly U-shaped. The form assumed by the cells on absorbing 
water seems to aid materially in pulling them apart, just as in Porella , 
though in a less marked degree. As the antheridium opens, the mature 
spermatozoids stream out, and within twenty to thirty seconds the anther- 
idium is empty. The ripe spermatozoids soon become free and swim about 
actively. 
Spermatogenesis. 
Up to the present time a number of investigators have studied the 
development of spermatozoids in the Hepaticae and have given more or 
less complete accounts 1 2 . Among the most comprehensive are the studies 
of Schottlander 3 , Guignard 4 , Strasburger 5 , and the recent and exhaustive 
work of Ikeno 6 on the spermatogenesis of Marchantia polymorpha. The 
development of spermatozoids in Fossombronia , while in many respects 
agreeing with Ikeno’s account 7 , presents, on the other hand, some interesting 
differences. Mitosis in the spermatogenous tissue of half- or two-thirds- 
grown antheridia was studied in a number of stages and does not differ 
in any way from the mitosis of the spermatid mother-cell. The nucleus, 
at first relatively small and almost hyaline, enlarges rapidly as it approaches 
the spirem stage and a considerable amount of chromatin appears. At 
the same time the nuclear membrane becomes apparently much thicker, 
but this seems due to the presence of chromatic material and to the 
cytoplasm, which at this stage is quite granular. The nucleolus is large 
and conspicuous, but as the prophase advances this structure appears to 
become more or less vacuolate, as described by Farmer 8 for F. Dmnortieri 
in the mitosis of the sporogenous cells. The most careful study of the 
1 Campbell (’95), p. 101. 
2 Campbell (’87), Bd. v, p. 120. 
2 Buchtien (’87), Heft 8. 
2 Le Clerc du Sablon (’88), p. 876. 
3 Schottlander (’92). 
4 Guignard (’89), p. 1. 
5 Strasburger (’92), Heft 4. 
6 Ikeno (’03), Bd. xv, pp. 65-68. 
7 id. (’03), Bd. xv, PI. Ill, Figs. 1, 9, 10. 
8 Farmer (’95), vol. ix, PI. XVI, Fig. 4. 
