Cuscuta and its Host. 
669 
doubt due to artificial shrinkage of the protoplasm. It was most pro¬ 
nounced in preparations which had at any stage been placed in alcohol 
(Fig. 72, PI. L); it is probable that the protoplasm in this region is easily 
shrunk, having only recently come to occupy its position against the sieve 
area, consequent on the solution of its own wall. In some preparations 
there was hardly any sign of shrinkage, e. g. in Fig. 68. 
There can be little doubt that the mucilaginous change described at the 
tips of the parasite hyphae is preparatory to dissolution. The actual dis¬ 
appearance of the blue mass is rapid and is difficult to detect. Fig. 71 
probably represents one of the last stages in its solution, the blue line being 
a last remnant of the mucilaginous mass. Several other cases, none of 
them very clear, were also met with, in which irregular c blobs ’ of callus 
separated the parasite protoplasm from the junction sieve area. Figs. 62 
and 67 were drawn from two rather curious cases, in which a relatively large 
mass of blue mucilaginous substance probably represented the dissolving 
tip of the wall. Apparently the protoplasm had already crept past the 
jelly-like mass and applied itself to the junction sieve plate. 
C. Functional Efficiency of the Arrangement. 
It is probable that the orientation of the hyphae which make con¬ 
nexion with the host phloem is significant from the point of view of the 
efficiency of the junctions. 
It has already been stated that the haustorium widens out when it 
reaches the phloem, its transverse section becoming much greater than 
before. As the hyphae enter the old phloem of the host they also change 
their direction slightly, and, instead of running in a straight line with the 
long axis of the haustorium, they twist a little obliquely and finally turn 
outwards at an angle with the long axis, thus coming into contact with 
sieve tubes which have not been interrupted by other ingrowing elements. 
In the functional and young phloem the invading hyphae turn still more, 
often twisting repeatedly so that their course is difficult to follow under the 
microscope, and they never remain long at the same focus. Finally, the tips 
of many of them run nearly at right angles to their original direction, that 
is, almost parallel with the host sieve tubes (see Figs. 59, 68, and 70, PI. L, 
Figs. 77 and 78, PI. LI, &c.). The advantage of this change of direction is 
clear when it is remembered that it is always with the lateral sieve fields (or 
occasionally with the lateral sieve plates) that junctions are effected. The 
hypha may be said to lay itself alongside the host sieve tube ; it is thus able 
to bring as large an area as possible into contact with a sieve field. At the 
same time it runs the least risk of injuring the sieve-tube wall, since it avoids 
bringing to bear on it the terminal pressure, which we have already seen to 
be sufficiently great to bulge in even the thick wall of a pericycle cell. 
Y y 
