PROTOPLASM THROUGH THE WALLS OF VEGETABLE CELLS. 
833 
proceed too far, the protoplasm is attacked and the delicate connexions soon become 
obliterated. Moreover, at the same time the middle lamella becomes swollen and will 
deeply stain, which of all things is to be avoided. A regulated action of the acid 
gives the best results, and if the protoplasm be not sufficiently shrunk to show up 
the processes to the best advantage, the section need only be mounted in strong 
glycerine which will soon bring about the desired effect. Even in one and the same 
preparation, though a successful one, the acid may have acted unequally, due, it may 
be, to varying thickness of the section, and thus the different results produced by the 
acid may be observed at the same time. However, in a well-prepared section, where 
the action of the acid has been properly regulated, plain examples of continuity are 
apparent. Upon longer treatment, the further shrinking of the protoplasm causes a 
greater tension to be exerted upon the processes and rupture ensues. This frequently 
occurs, but the rupture nearly always takes place on one or both sides of the point 
where the thin thread-like process crosses the middle lamella, and seldom at the point 
itself. Thus the threads cannot be said to be merely pulled out of the pits, for 
rupture takes place in such a manner that a longer or shorter length still remains in 
the pit cavity. 
Finally, when the action of the acid has been carried too far, the processes appear 
to have been partially destroyed, and but few can be traced as far as the swollen and 
now deeply-stained middle lamella. Many of the processes appear to be directly and 
uninterruptedly continuous from cell to cell, whilst others are swollen at the point 
where they cross the middle lamella. In other cases between the two ends of the 
strongly-stained processes there is a lighter-stained portion, which connects the two 
This lighter-stained area exhibits a haziness and appears to be somewhat indistinct, 
although well defined from the rest of the swollen cell-wall, and clear enough not to be 
confounded with the middle lamella (Plate 68, fig. 5). Again, when the protoplasm is 
but slightly contracted, and but little tension has been exerted on the threads, the point 
of junction of the two threads is both slightly swollen and also coloured darker than 
the rest. In spite, however, of the fact that in several cases direct continuity appears 
to exist, I am strongly of opinion, both from analogy and from such appearances as I 
last described, that in reality a sieve-plate arrangement occurs, It must be borne 
in mind that the difficulties of examination are great, both on account of the smahness 
of the pits and the thinness of the pit membranes, but in any case we cannot imagine 
that the threads go bodily through the pit, for were it so, the pits would not possess 
a closing membrane, and ordinary staining would soon demonstrate the existence of 
the protoplasm, by which the pit was perforated. 
Although Peeffer’s* results appear to prove that it is the underside of the 
pulvinus which is especially sensitive, I have not been able to establish any difference 
between them as far as histological evidence goes. Nor is this to be greatly wondered 
at, for the method for such a discrimination is essentially rough, and one would hardly 
* See Sachs’ ‘ Text book,’ p. 889. 
