of Terrestrial Algae . II. 725 
ment with silver nitrate in the absence of light. These granules are 
apparently the ones that are particularly prevalent in the healthy cells sur¬ 
viving drought. 
The general conclusion to be arrived at is that, in fresh material of the 
terrestrial forms investigated, the cells are in diverse states as regards 
drought resistance, resistance to hypertonic solutions, and permeability to 
stains. There would, however, probably be no means of grouping these cells 
in sharply delimited categories, since in all these respects there appears to 
be a practically complete grading. The numbers of cells consequently that 
are killed or become unhealthy during a period of drought or exposure to 
hypertonic solutions will depend on the condition of the original material, 
and in this respect Pleurococcus and Cystococcus differ from all the others in 
possessing a far larger number of resistant cells. 
The experiments with material placed in open vessels in hypertonic 
solutions show that a certain, and no doubt varying, percentage of the cells, 
which are characterized by rapid recovery from plasmolysis, are little resis¬ 
tant to these conditions and die in a short space of time, whilst a relatively 
large percentage, though in great part remaining plasmolysed, continue in 
a healthy condition for many weeks or even months. Under conditions of 
drought or in a sealed slide (cf. p. 689), on the other hand, practically all the 
cells in the final stage exhibit no plasmolysis, though a very large number 
of them show the permanent slight contraction of the protoplast which does 
not disappear even in tap-water. These cells, which are readily permeable 
to such stains as eosin and erythrosin, are apparently in an unhealthy state 
though not dead, and their abundant occurrence both in drought material 
and in a sealed slide indicates that, in both these cases, the conditions are 
more deleterious than in open hypertonic solutions. In all probability these 
cells correspond to those which remain plasmolysed under the latter treat¬ 
ment, since the numbers of perfectly healthy unaffected cells, always rather 
small, are approximately the same in all three cases. The cells with slightly 
contracted protoplasts are probably in a moribund condition, from which 
recovery is either very slow or altogether impossible ; recovery may depend 
on the intensity and duration of the previous treatment. There are some 
data that indicate that a relatively slight drought produces a condition that 
is reversible, a more pronounced one an irreversible condition. In any case 
it may be well to point out that the drought to which these forms have been 
subjected in our experiments was either more prolonged or more intense 
than anything that is likely to occur in nature. 
With all kinds of treatment, however, there usually remain a certain 
number of healthy green cells, not plasmolysed by even high concentrations 
(25 per cent.), generally impermeable to the various stains used, and in the 
filamentous algae usually characterized by the possession of numerous fine 
peripheral granules. Even supposing that all the cells above described as 
