306 Sykes. — Anatomy and Histology of Macro cystis pyrifera and 
seen (Figs. 8 and 9 d , PL XIX) situated in a blue area which represents the 
original position of a group of threads, and these, together with the inter- 
vening callus and wall, when any unaltered wall is still present, appear 
to fuse and disorganize, giving rise to the single slime-string. 
The sieve-plate is now fully developed and is a distinct sieve with open 
pores, through which, in optical section, slime-strings are often seen to run, 
though in many cases, owing to the thinness of the sections, they have 
fallen out of the pores during the processes involved in staining the pre- 
paration. Three to five sieve-tubes with sieve-plates in this condition may 
be found in any radial row in the oldest material of Macrocystis examined. 
The obliteration of the innermost active sieve-tubes appears to be 
a very rapid process. Callus spreads over the surface of the sieve- 
plate in between the threads and then begins to accumulate at the centre 
of the plate. The mucilaginous and protoplasmic contents of the tube 
are at this stage collected into a densely staining mass over the central pad 
of callus (Fig. 17 , PL XIX) and continue to deposit more callus, gradually 
increasing the size of the pad. A thick layer is thus spread over the whole 
plate and a callus-mass results (Fig. 19, Pl. XIX) on both sides of the plate, 
but is usually larger on one side than on the other. Meanwhile callus is 
often formed down the whole length of the tube (Figs. 20, 23, PL XIX), in a 
manner similar to that described above 1 for the primary pith filaments, and 
accumulates in small masses which are laid down by the dense protoplasm 
lying against the wall of the tube, till finally in many places the lumen is 
almost obliterated (Fig. 21 , PL XIX). 
ii. Longitudinal sections of the terminal sieve-plates. In longitudinal 
sections the course of events is not always easily followed. In the youngest 
sieve-plates examined, the threads are very delicate and often much broken, 
but they can be seen to be arranged singly (Fig. 11, PL XIX). More 
difficulty was experienced in following the formation of the groups of 
threads, as the sieve-plate at this stage often becomes very swollen. In 
several cases single threads were seen in the central portion of the plate, 
while groups occurred on the periphery. Fig. 13, PL XIX, shows the 
occasional appearance of groups of threads in a plate in which the separate 
arrangement is still prevalent. In numerous sieve-plates which as yet had 
no callus, the threads could be clearly seen to be arranged in groups 
throughout. In several cases a median node, such as that described in 
Pinus 2 , was seen on each thread, and in one such instance callus had been 
formed and the threads were still visible traversing the callus-rod. 
With the appearance of the callus the threads stain much more deeply 
with Safranin. These alterations are first noticed at the periphery of the 
1 p. 302. 
2 Hill, The histology of the sieve-tubes in Pinus. Ann. of Botany, xv, 1901, p. 589, and 
Figs. 8, 10, 8 c c., PI. XXXII ; and p. 590 and Fig. 20, PI. XXXIII. 
