Hill . — The Histology of the Sieve- Tubes of A ngio sperms. 269 
nodal swelling appears to belong definitely to the thread itself. 1 The 
darkly-staining median nodes of the sieve-field slime-strings appear, there- 
fore, to be similar to those described for Finns , 2 and may be due to the 
action of the ferment on the small portion of the middle lamella between 
the two opposed callus-rods. It has been noticed that these nodes can be 
stained with Safranin without previous mordanting, and that the threads 
themselves remain unstained by this treatment ; also after the action of 
Russow’s Iodine and 75 per cent. Sulphuric acid the median nodes stand 
out as highly refractive granules, arranged along the middle of the middle 
lamella, when all other structures have vanished (Fig. 43, PI. XVIII). After 
two days in 10 per cent. Potash, and subsequent washing in water and weak 
Hydrochloric acid, the middle lamella nodes had apparently vanished 
from the sieve-threads in sections of Vitis, since they could not be seen 
after treatment with Russow’s Iodine and 75 per cent. Sulphuric acid. 
With the ‘ Acid Violet ’ method of staining, however, the nodes reappear, 
and it is probable that in this latter case we have to deal simply 
with a precipitation effect in the empty spaces once occupied by the 
substance of the nodes themselves. The callus also disappears with the 
Potash treatment, and the split middle lamella with the thread-tubes in 
the two separated portions of the wall can be clearly seen (Fig. 42, PI. XVIII). 
Owing to the very small size of the median node, it is by no means easy 
to obtain a clear idea of its nature, but from the experiments which have 
been made it appears that this conspicuous structure is a kind of connecting- 
link between the two callus-rods of a single string containing or enclosing 
the median portion of the string itself (Fig. 41, PI. XVIII). 
The Sieve-Plates. 
The study of the development of the sieve-plates is a matter of greater 
difficulty than that of the sieve-fields, owing to the extremely delicate 
character of the membrane which is to become the sieve-plate. 
The most common stage of development met with in surface views of 
properly preserved material shows the young sieve-plate furnished with more 
or less polygonal areas, which give the callus reaction, separated from each 
other by narrow bands of cellulose. These cellulose bands, which will con- 
stitute the mesh of the adult sieve-plate, appear in section as ridges on either 
side of the membrane, and it is owing to their elevation above its general sur- 
face that the young sieve-plate membrane assumes its deeply pitted character. 
The pits are lined by callus in the form of little basins 3 (Figs. 1 and 3, 
PL XVII), but the actual pit-closing membranes, together with the ridges or 
mesh of the plate, are formed of cellulose (Figs. 1 and 2, PI. XVII). Younger 
stages have been observed in which, owing to the shallow pits, the cellulose 
1 Cf. Gardiner on the mode of formation of the initial cell-wall ; Proc. Camb. Phil. Soc. xiv, 209. 
2 Hill, 1. c., p. 587, PI. XXXI, Fig. 11. 
3 Cf. Russow, Sitz. d. Nat. Ges. Dorpat, 1882, p. 304. 
