Bower. — Studies in the Phytogeny of the Filicales. 449 
each three, but in Fig. 9, III, only two. These facts are consistent with the 
stalk being composed of four rows of cells, as will be seen to be the case 
from transverse sections. 
These drawings give a comprehensive idea of the external characters of 
the sporangium of Metaxya. It should be compared with that of Lophosoria , 
since there is so obvious a similarity between these types of Ferns. By 
comparison with PL XXXV, ‘Annals of Botany,’ 1912, which illustrates 
Losophoria , it will be seen that not only are the sporangia of Metaxya much 
smaller, but also more elongated, and composed of fewer cells. For instance, 
the cells of the complete annulus in Fig. 18 of Lophosoria total 39, those in 
Fig. 9, I, of Metaxya are only 27, and in Fig. 9, ill, 25. Or again, the cells 
on the peripheral face of the sporangium of Lophosoria shown in Fig. 18 of 
PL XXXV (1912) number 6 o, but those in Fig. 9, I, of Metaxya are 17, or 
in Fig. 9, III, they are 25. The number of cell-rows composing the stalk 
is also smaller. Lastly, there is the very essential difference in the position 
and interruption of the annulus. This is in Lophosoria markedly oblique, 
and continuous ; in Metaxya it is as a rule interrupted at the stalk, and 
shows only traces of the oblique position, being almost vertical. Moreover, 
the stomium is ill-defined in Lophosoria , but in Metaxya it appears of the 
well-defined type characteristic of so many of the advanced Lepto- 
sporangiate Ferns. The spore-output per sporangium appears to be the 
usual number of 64. 
In writing some time ago on the sorus of Gleichenia , it was pointed 
out (‘Ann. of Bot.,’ xxvi, p. 275) that in G . pectinata the Gleicheniaceous 
type of sorus has reached the point of ineffectiveness in its increase in 
number of sporangia, since they are so closely packed that, being all 
of like age and dehiscing by a median slit, they cannot shed their spores 
when ripe. It was shown that there are four possible ways out of the 
difficulty, which may be adopted singly or in combination: (1) by increas- 
ing the length of the sporangial stalk, (2) by adopting a lateral dehiscence, 
(3) by extending the area of the sorus, or (4) by elongating the receptacle. 
Gleichenia adopted none of these, but other Ferns did, and have progressed 
as a consequence. Metaxya is one of them. Its receptacle has remained of 
about the same height as in Gleichenia , but its area is greatly enlarged, 
while the sporangia have lateral dehiscence, and are rather longer stalked. 
Thus a combination of three of the modifications above noted is to be found 
in the sori of Metaxya. The transverse dehiscence is shared by all the 
more advanced Leptosporangiate Ferns. The elongation of the stalk 
is here only relatively slight. It is the increase of the area of the receptacle 
which is the most marked feature, which in the absence of any elongation of 
the receptacle makes the large number of sporangia in the simple sorus of 
Metaxya a practical possibility. It is naturally to certain sections of the 
comprehensive genus Polypodium that we shall look for conditions which 
