322 Bower . — The comparative examination of the 
definite and regular as in any of the Polypodiaceae (Pig. 22). 
Thus it appears that, notwithstanding the size and internal 
complexity of the mature stem, the character of the apex is 
essentially similar to that of the smaller forms. 
I have but few facts to adduce as to the structure of the 
apex of the stem of the Schizaeaceae. Klein (1. c.) has inves- 
tigated those of Lygodium circinatum and scandens , and 
appears to have found in them a structure similar to that of 
other dorsiventral Ferns, for he gives no special description of 
them 1 . I have examined apices of Lygodmm scandens , and 
of Aneimia hirta , and they both conform to the ordinary type 
of structure with a three-sided apical cell, while the apical 
cone is comparatively flat. 
Passing on now to Osmunda regalis , in a considerable num- 
ber of well-grown plants, from Ireland and elsewhere, the apex 
of the upright stem was found to be terminated by a flat apical 
cone : the meristem is very thin-walled and watery, with 
greatly vacuolated protoplasm : in transverse sections, cleared 
with ‘eau de Javelle,’ in the majority of cases a three-sided 
apical cell is seen, from which segments are cut off in the 
ordinary succession (Fig. 23) ; but the subdivision of the 
segments often shows irregularities. The size of the segments 
relatively to that of the initial cell is variable in different cases, 
and this will be recognised on comparing Figs. 23, 24. In 
some cases the form of the cell is not regular : thus in Fig. 24 
it is four-sided, and the succession of segments is not so readily 
to be traced as in Fig. 23. In some few apices which were 
observed it appeared to be impossible to refer all the tissues 
to any one initial cell (Fig. 25). Now it is to be noted that 
all these figures were drawn from sections which included the 
outer wall of the cells : they represent with accuracy what is 
seen on viewing the apex from above, and no allowance need 
be made for distortion of the cells in preparation. Taking 
then Fig. 25 as representing accurately the actual arrangement 
of the cells, it seems probable that more than one initial is 
present, but the structure is such as cannot readily be referred 
1 Bot. Zeit. 1884, p. 647. 
