450 Messrs Gardiner and. Hill , The Histology 
the accompanying group of threads presents the appearance of the 
so-called “ pit- threads — “aggregirte Plasmaverbindungen” — on 
the one side, and of the “ wall threads ” — “ solitare Plasmaverbind- 
ungen ” of Kohl 1 — on the other side of the middle lamella, and it 
is therefore clear (and this is also true for the endosperm of 
Phoenix) that the distinction which has been made between these 
two forms of threads in thick-walled endosperms is not a valid one, 
since all gradations from the “ pit ” type to the “ wall ” type may 
occur in the endosperm tissue of one and the same seed. 
The composition of the reserve cellulose in Galium as in Tamus 
is rather a matter of conjecture. Yet sections of the fresh endo- 
sperm give fairly definite results, which tend to prove that the 
walls are composed of a substance of the nature of pecto-cellulose. 
With weak watery methylene blue, all the walls take up the stain, 
but those just under the testa stain more deeply. With Congo 
red, on the other hand, all the walls are stained pale pink, with the 
exception of those just beneath the testa, which are scarcely 
coloured by the stain, and a similar effect was also noticed with 
a solution of iodine in potassium iodide. On treating sections 
with iodine and sulphuric acid the cell walls turned a bright 
greenish blue. 
If sections of a seed which has been allowed to germinate are 
examined, it is seen that the space occupied by the cotyledons has 
become considerably enlarged, and the endosperm cells lying 
around this cavity have become somewhat crushed and depleted of 
their contents. The walls of the cells also in this region appear 
thinner and more mucilaginous than those situated at a greater 
distance from the developing embryo. In sections treated with 
methylene blue or Congo red these differences are very sharply 
marked, for with the blue the cells around the cotyledonary cavity 
remain uncoloured, whereas the walls further away are fairly 
deeply stained, but with Congo red, on the other hand, the walls 
of the cells which surround the cavity, and which appear mucila- 
ginous, stain a deep rose-pink, whilst those which are as yet 
unaltered only take on a faint pink coloration (Fig. 12). Similar 
effects to these were noticed in sections stained in safranin, and 
in this case the mucilaginous walls are unaffected by the stain. 
From such results it seems highly probable that the composite 
cell walls are being disorganised by an enzyme in such a way that 
the pectic compounds, which were seen to stain with methylene 
blue or safranin in the ungerminated seed, are being removed, 
whilst a matrix, consisting of an hydrated form of pure cellulose, 
which gives the characteristic reactions with Congo red and iodine 
and sulphuric acid, is left behind. 
1 Kohl, “ Dimorphism us der Plasmaverbindungen,” Ber. d. Deut. Bot. Ges., 
Bd. xviii., 1900, p. 364, Taf. xn. 
