Bower . — S hi dies in the Phytogeny of the Filicales. 465 
The sorus of Loxsoma is marginal in origin, with a cup-shaped indusium, 
from the centre of which the receptacle springs. Upon it the sporangia 
arise in a basipetal sequence. The position and structure of the sorus 
correspond to what is seen in Thyrsopteris , but the sporangia of Loxsoma 
differ in having an oblique, though imperfectly indurated annulus, and their 
dehiscence in a median plane. I have made new drawings of the sporangia 
from three different aspects. Fig. 26, a , shows a sporangium with its 
attendant hairs presenting its ‘ peripheral ’ face ; that is, the side which is 
away from the receptacle. The annulus is seen with its distal side composed 
of large indurated cells, while the slit of dehiscence lies in the median 
plane. The basal side of the annulus is not indurated, though the sequence 
of its cells is easily traced, and sometimes they show a partial induration. 
The tabular cells which lie within the ring constitute the ‘ peripheral ’ face 
of the sporangium, and they are seen to be numerous. Fig. 26, b , shows the 
same sporangium from the opposite side, i. e. it represents the ‘ central ’ face, 
which is turned towards the receptacle. It is composed entirely of thin- 
walled tabular cells, and the annulus is wholly out of sight. Fig. 26, c, 
shows a similar sporangium seen laterally, and it brings out the conical 
form which the sporangium usually shows. The apex of the cone is at the 
centre of the ‘ peripheral ’ face. The biological meaning of the imperfectly 
indurated annulus was explained in my Studies, IV, Leptosporangiate 
Ferns (‘ Phil. Trans./ vol. cxcii, p. 49). The comparative conclusion there 
arrived at was that Loxsoma appears to be a link connecting the Gleichenia- 
Schizaea affinity with the Dennstaedtiinae. As Professor von Goebel has 
now suggested that Loxsoma is a reduced type from the Cyatheaceae, and 
rejects the comparison with the Gleicheniaceae, I take this opportunity of 
re-stating my position in the matter. 
In my view the underlying type of sporangium is the same in all 
Ferns which show an oblique annulus. Especially is this the case for the 
Gleicheniaceae and Schizaeaceae, as was fully shown in my Studies, IV, 
on Leptosporangiate Ferns ( c Phil. Trans./ vol. cxcii, p. 10 1). In all of them a 
‘peripheral’ is distinguished from a ‘central’ face, and in all the more primitive 
types, where the sorus is not monangial, the position of these faces relatively 
to the receptacle is that which the above terms imply. The differences lie, 
not as von Goebel’s diagram (‘Flora/ vol. cv, p. 41, Fig. 7) would suggest, in 
the position of the ring, but in the greater or less proportions of the two 
faces. I would point out that in von Goebel’s Fig. 7 (1. c., p. 41), which is 
here reproduced as Fig. 27, while in (1) the orientation of the sporangium 
of Loxsoma is quite correctly given relatively to the receptacle of the sorus , 
in his diagram (il) of Gleichenia the receptacle is omitted, and the orientation 
of the sporangium relatively to the leaf-surface is shown, which is quite 
a different thing. I venture to amend his diagram in Fig. 27 bis by putting 
in the receptacle, and turning it into a position corresponding to that of the 
