INTERCELLULAR RELATIONS OF PROTOPLASTS. 
125 
opening of the sieve pores; later the callus is entirely dis¬ 
solved, leaving the pores open and the sieve plate quite bare. 
But at the same time, however, the protoplasm disappears 
from the sieve tubes, leaving behind only a watery fluid. 51 ' 
On passing from their evolutive stage, the sieve tubes of 
conifers pass directly into a passive state. Another point of 
biological interest is that in conifers each sieve element is 
directly formed from a cambial cell, without undergoing, as in 
Angiospenns, any prior subdivisions. 
A final point of interest is the contents of the tubes in 
the different groups of plants. In Vascular Cryptogams the 
sieve tubes contain neither protoplasm (other than a very 
thin parietal layer), nor nucleus, nor starch ; the parietal 
layer of protoplasm contains, or has adherent to it, a number 
of highly refractive globules of albuminous nature, which 
especially accumulate at the base of the pits. These globules 
are often much more numerous on one side of the sieve plate 
than the other. 
In Grymnosperms, as the sieve tube approaches its complete 
state, the nucleus first disappears, then the bulk of the proto¬ 
plasm, while a thin parietal layer of this persists till the time 
when the pores are opened ; this parietal layer contains a 
number of highly refractive granules, of albuminous nature, 
especially abundant near the sieve plates; starch is absent. 
In Monocotyledons, as the sieve tube approaches its com¬ 
plete state, the nucleus first disappears, then (or, e.g ., in 
Phragmites previously) the protoplasm is reduced to a thin 
parietal layer, with sometimes (Typha, Phragmites f ) a mass 
of very refractive protoplasmic jelly collected on both sides of 
the sieve plate, but more largely on one side than on the 
other. In other cases (according to Bussow generally) this 
jelly mass is not present; numerous refractive granules, 
albuminous in nature, adhere to the parietal protoplasm, 
chiefly near the sieve plates. Starch is rarely present. 
In Dicotyledons, as the sieve tube approaches its complete 
state, the nucleus first disappears, then the protoplasm is 
reduced to a thin parietal layer, a mass of albuminous jelly 
is collected chiefly at one side of the sieve plate (well shown 
in Plate III., fig. 4), and a similar mucilaginous strand often 
traverses the length of the constituent cell. Starch is very 
generally present during the active life of the tubes, and, as I 
have demonstrated in a very large number of plants, mainly 
on one side of the plate. 
* This, however, I have seen show a clear proteid reaction, 
t Janczewski, 1. c. 
