INTERCELLULAR RELATIONS OF PROTOPLASTS. 
123 
Tims Russow himself mentions the case of Yucca aloifolia , 
15 years old, where the sieve tubes in all the fibro-vasal 
bundles were still active and had callus coverings to their 
sieve plates, the callus, nevertheless, being no thicker in the 
oldest than in the youngest tubes ; while, on the contrary, in 
many herbaceous Dicotyledons of exceedingly rapid growth 
(Bryonia), the callus formation is often very considerable. 
To resume our account of the sieve development. The 
callus projections now swell, fuse together, and completely 
cover all the parts of the young sieve plate; and in those 
places where previously the cellulose wall was thinned (called 
above the “ mesh work ”) this wall now entirely disappears, 
having been either dissolved or transformed into callus, so 
that at these points the callus on the two sides of the plate 
is fused. The young sieve is now ready for perforation. In the 
meantime great changes have taken place in the contents of the 
tube, which it is not a part of our present purpose to describe, 
and masses of protoplasmic jelly have accumulated on either 
side of the still closed sieve plate, separated from it by the 
parietal layer of protoplasm. Opposite to some or all of the 
meshes of the callus this jelly puts out a minute prolongation, 
covered by the parietal layer, which gradually penetrates the 
callus substance of the sieve, perforates it, and fuses with the 
similar mucous substance of the cell on the other side of the 
plate, forming thus a continuous filament. Up to this time 
the sieve tube has been in what Janczewski calls an “ evolu¬ 
tive ” state ; it now passes into what he calls the “active ” 
state, m which it will remain for a varying period (see Plate 
III., figs. 2 to 5). A sieve plate thus formed consists there¬ 
fore of a very delicate network of cellulose, shown still 
faintly in Plate III., fig. 2, taken from Bryonia dioica, covered 
by a callous envelope, which lines also the perforating canals. 
If in this state the cell-wall, &c., be entirely removed by 
sulphuric acid the connecting threads will perhaps be left 
intact, communicating from one cell to the other through the 
thickness of the sieve, as is shown in Plate III., fig. 6 (Cytisus 
Laburnum). 
Such is the typical method of development of the sieve 
perforations and plate in Angiospermous Phanerogams. In 
the references to the literature above, I have noted that a 
certain amount of relation appears to exist between the 
viability of the sieve plate and the seasons. My own 
investigations upon the roots of hardy dicotyledonous trees, 
both deciduous and evergreen (e.g., Cerasus Laurocerasus, 
Rhododendron ponticum, Cytisus Laburnum , various species of 
Acer, Quercus, JEsculus, and those of Cheiranthus Cheiri), lead 
