Brown . — Studies in the Physiology of Parasitism . VIII. 103 
The plant material consisted of leaves and floral structures, viz. : petals 
of Cereus, Phyllocactus, Gloxinia, Lijium, Tulip, Rose, Begonia, Viola, 
Sweet-pea, Dahlia, Geranium, Cydonia, Pyrus, and leaves of Broad Bean. 
A large amount of the work was carried out with the petals of Cereus 
spectabilis , which offers special advantages. These petals are highly and 
uniformly coloured, and lend themselves readily to plasmolysis studies by 
means of the microscope ; from the same flower a large number of petals 
can be obtained which show similar behaviour, and which, especially when 
taken from the unopened bud, show a perfectly clean surface ; the results 
obtained by both methods of experimentation are well defined ; ancf lastly, 
the petals present a flat surface and are moderately easily wetted. The 
latter is an important practical point which may be illustrated by a few 
examples. In the case of such petals as those of Viola and of some varieties 
of Rose, the drops do not wet the petals at all. They maintain a more or 
less spherical shape, and from their appearance it is clear that a film of air 
is entrapped between them and the surface of the petal. The drop, therefore, 
has only very limited contact with the petal. In such cases it is found that 
the exosmosis figures are always very low. Such petals are further very 
troublesome in use, in that the drops are very liable to shift, coalesce, or 
run off. At the other extreme is the case of the Bean leaf, where the 
drops tend to spread too far, especially along the line of the veins, thus 
making it impossible to ensure that the drops occupy approximately uniform 
areas of the plant surface. In the case of Cereus petals an intermediate 
degree of wetting is shown ; comparatively good contact obtains between 
drop and petal, and the drops present a fairly uniform appearance when 
laid on the petals. 
An attempt was made to get over this difficulty of non-wetting in the 
case of some plants by the addition of traces of substances such as Castile 
soap and saponin to the water, but the preparations tried were found to 
possess too high a conductivity for use in this connexion. 
For the reasons given above petals of Cereus were selected for detailed 
examination. As was anticipated, a considerable degree of variation was 
obtained from drop to drop. Certain well-marked rules were, however, made 
out, of which the following is an account. 
On the same petal the conductivity of the drops varies from place 
to place, being smallest in the centre along the line of the midrib, and 
increasing towards the distal (free) end of the petal. The greatest values 
are obtained along the margins, and especially near the tip. The following 
figures, representing the conductivity of drops which have lain in contact 
with the petals for twenty- four hours in a moist Petri dish at laboratory 
temperature (17-20° C.), illustrate this point. The numbers in the diagram 
(see next page) indicate the position of the drops on the petal. 
