190 
H. W. RICKETT 
separate pots, and sketched each day under a binocular microscope. Young 
plants with only two or three lobes were selected in most cases, and watched 
in this way until they had attained the size of normal mature plants. The 
lobes differ sufficiently in shape and size so that one can feel certain that 
one is following the growth of each individual lobe from its beginning until 
maturity. The plants were kept in the Wardian case and showed the 
peculiarities of growth already described. 
This study shows that the lobes of the thallus are formed at the tip 
and are pushed back into a lateral position, as they increase in size, by the 
elongation of the median portion of the thallus, or midrib. It is also seen 
that they are not "middle lobes" in the ordinary sense, that is, lobes pro- 
duced simply by the outgrowth of the thallus between two divisions of the 
originally single growing region; for very commonly two lobes are formed 
at the same time, without any consequent branching of the thallus. If 
all the notches between the lobes represent divisions of the apical region, 
it must be that some of these branch regions are arrested in further develop- 
ment, and that their cells merely mature without dividing further. When 
one lobe is formed singly, it often resembles the ordinary ''middle lobe" 
of such a form as Ricciocarpus natans, and the growth of the thallus con- 
tinues on either side of the lobe. This, however, is not necessarily the case, 
since in many cases only one of the regions to either side of the lobe pro- 
duces new growth, the middle lobe being pushed aside and coming to lie on 
one side of the central axis, without a corresponding lobe on the other side. 
This is of common occurrence, and is often responsible for the curving 
growth of the thallus as a whole. 
In the light of the foregoing studies, therefore, it is reasonable to suppose 
that a lobe is formed by the occurrence of vertical divisions in a lateral 
segment of an initial cell, and this, as both the sections and the study of 
living plants show, may, at least in greenhouse cultures, occur anywhere 
in the apical region, irrespective of divisions of the latter. 
In nature, the growth of the thallus, to judge from the accounts of 
previous authors, is apparently much like that of Riccia. According to 
Douin's (5) description of Sphaerocarpos, there are usually two notches; 
separated by a large middle lobe, present in a mature plant. In the 
notches are usually two other small lobes, with a growing point on either 
side — thus four growing points- in all. At this point, he says, growth usually 
stops, though it is not uncommon to find plants having five large lobes with 
small ones in each notch. The growth, of course, is limited by the end of 
the growing season. The only difference between this history and that of 
Riccia, according to the same author, is the fact that in Sphaerocarpos the 
middle lobe remains undivided, whereas in Riccia it becomes cleft as the 
midrib elongates. Hence in the latter case the growth of the midrib is 
more rapid than the intercalary growth at the base of the lobes, while in 
the former case the reverse is true. In nature, evidently, the lateral seg- 
