Cuscuta and its Host . 675 
such evidence research must go back to the developing cell plate, of 
which our knowledge is at present so meagre. 
The presence or absence of connecting threads between non-genetically 
connected tissues is, however, at any rate a strong indirect argument for or 
against the view that their formation is intimately connected with the 
genesis of the cell. Strasburger apparently expects to find such threads 1 
at least between cells of the same or of sufficiently closely allied plants. 
When working on the histology of Macrocystis pyrifera and Laminaria 
saccharina 2 1 tried to find out whether connecting threads were developed in 
the septa formed between the tips of the invading cortical hyphae and the 
longitudinal walls of the primary pith filaments. It was, however, not only 
very difficult to distinguish these septa, but I had no young material 
adequate for their study. 
We are also still ignorant as to what happens in the case of sliding 
growth and grafts. 3 
But the present investigation has, I believe, made it clear that in two 
cases in which cells not genetically connected are brought into contact 
later, no connecting threads are formed. In the one case, that of the 
approximation of the separate invading hyphae to form a tissue, the 
adjacent longitudinal walls fuse together, but no connecting threads are 
ever developed in them. 
Now this is a case of fusion between the young thin cell-walls of cells 
which belong, not to different plants, however closely allied, but to the 
same plant. The fusing cells are, moreover, in an identical stage of 
development, so that if ever the formation of connecting threads were 
to be expected between cells not genetically connected it would be here. 
The other case, that of fusion of part of the tip of a hypha with a host 
sieve area, has formerly been described as resulting in a fusion sieve plate, 
the slime strings of which would, one supposes, be derived from both host 
and parasite. But I have been unable to find any evidence that this is so. 
The junction sieve area is merely the intact sieve area of the host sieve 
tube ; the parasite wall immediately overlying it first becomes swollen and 
mucilaginous and then is completely dissolved away. 
It would seem that the application of naked protoplasm to a bored-out 
wall would be especially favourable to the formation of £ plasmodesmic 
pseudopodia ’ which should project into the wall and fuse with the host 
protoplasm. Yet not only is there no sign of the development of such pro¬ 
cesses, but the parasite protoplasm is probably not even very closely applied 
to the junction sieve area. The junction between the sieve tubes of Cuscuta 
and its host therefore afford no argument in favour of Strasburger’s hypo¬ 
thesis. Indeed, the two cases, taken together, may be regarded as a strong 
argument against such a view, 
1 Strasburger, 1901. 2 Sykes, M. G., 1908. 3 Gardiner, 1898 a, p. 310. 
