Plagiochasma appendiculatum and P. articulatum 237 
usually empty even in comparatively young androecia, the antheri- 
dia having discharged their contents. In older androecia all the 
chambers are naturally empty. It is clear, therefore, that in these 
cases two growing points are involved. The same sequence of 
development is also seen in longitudinal sections of the lobes. The 
antheridia near the apex may not even have formed the sperm 
mother-cells while those behind may have well developed sperms, 
and still further back they may be quite empty. 
Each antheridium is an ovoid structure with a slender stalk 
and a conspicuous beak. The sperms are long, narrow, coiled bodies. 
Unfortunately no androecia were found in very early stages of 
development so that it was not possible to study the early stages 
of development of the antheridium. Similarly the apical cell, if it 
was ever present, could not be made out. 
Only the bilobed condition of the androecium is met with 
in Plagiochasma articulatum. In P. appendiculatum the androecium 
shows a great variety of form. 
In some plants of P. appendiculatum it was found that the two 
lobes of the androecium showed a second forking. In some robust 
plants growing in a shady and moist place it was seen that the 
androecium went on growing as a simple cushion-like structure in 
the mid-dorsal line for some time, forming a linear-oblong structure 
which divided later into two, and each of the lobes thus formed 
again forked (Fig. 2, e). These ultimate lobes showed an 
acropetal succession in the development of the antheridia. This 
condition is particularly interesting as showing undoubtedly the 
dichotomy of the androecium. In still other cases it was found 
that the androecium was more or less irregulary lobed (Fig. 2). 
Some of these lobes were of a small size and were probably due 
merely to irregularities of growth, but others showed distinct signs of 
growing points, from the arrangement of the scales and the sequence 
of the antheridia. Most of these specimens were quite old 
and the antheridia had discharged their contents, but enough was 
seen to conclude that some of the lobes undoubtedly possessed 
growing points. 
Three different considerations, therefore, show that the 
androecium of P. appendiculatum and P. articulatum is really 
homologous with that of the higher Marchantiales in being a branch- 
system. The dorsal position, like the dorsal position of the female 
receptacle, is quite secondary as has already been shown by the 
writer (New Phytologist, Vol. XI 11, 9). In P. articulatum the 
