400 Farmer. — On the Morphology and 
of a cell-chain, or they may arise by the lateral bulging-out 
of a cell, and the subsequent separation of the protruding 
portion, by a cell- wall. The new ‘apex’ just formed then 
grows independently (see Pl. XXV, Figs. 3, 6, 7) between the 
cell-rows on all sides of it. It may, however, be urged that 
the appearance which I have described is simply due to a tan- 
gential division extending through a number of cells or through 
one cell which has then grown pari passu with its mother-cells 
and as a twin filament, but this suggestion is shown to be 
inadequate by a study of a series of transverse sections in 
which such an occurrence is taking place (PI. XXV, Figs. 4, 5). 
Moreover, I have been able to observe karyokinesis in the 
terminal cell of one of these intruding rows at an early stage 
in its formation, and at a time when the surrounding cells 
exhibited no evidence of continued growth. This observation 
is not, of course, to be regarded as very weighty by itself, but 
it is perhaps deserving of mention taken in connection with 
others, all tending to establish the same conclusion. Another 
argument in favour of the belief that the filaments are actually 
intrusive, is to be found in the fact that immediately in front of 
a cell-row of this character it is very common to find a small 
triangular splitting, just as would be caused by forcing a 
wedge of wood into a piece of timber in the direction of the 
grain (PI. XXV, Fig. 7). 
This intrusion of younger between older cell-rows is by no 
means restricted to the peripheral layers, it occurs also in the 
central tissue (PI. XXV, Fig. 2), which forms the pith in the 
lower end of the flower, and indeed it is a feature by no 
means uncommonly to be met with in organs which are 
growing in any one direction in such a way as to ensure 
a rapid multiplication of cells in a straight line. Roots, for 
instance, and some stems, particularly of such plants as Sela- 
ginellas, show the phenomenon remarkably clearly. It is true 
that I have not succeeded in tracing an intercalated row 
beyond a short distance from its point of origin; this may be 
partly due to the sinuous course which such rows probably in 
all cases take, but in any case it is not likely that they reach 
