184 Proceedings of the Royal Society of Edinburgh. [Sess. 
tint. This cannot be regarded as having been caused by a differentiation 
in the contents of the threads. In some cases a false impression of a 
transverse septum is given when two adjoining lighter areas occupy the 
whole breadth of the thread, leaving between them a thin transverse line 
of a deeper brown tint. The same false appearance of transverse septa is 
more strikingly shown in threads of Peronosporites antiquarius, and they 
were described as such by Worthington Smith (11). In the case of P. 
antiquarius the absence of real transverse walls has already been noted 
by Williamson (13) and by Murray (6). It is impossible to resist the con- 
clusion that the same mistake was made by Renault (see Meschinelli, p. 9) 
in his description of the hyphse of P. gracilis. In one of his two figures 
(see Meschinelli, tab. v, fig. 14) not a single transverse septum is shown. 
Again, in no single case do the threads show external constrictions — in 
marked contrast to the threads of the fungus examined by Cash and 
Hick, which show abundant transverse septa with marked constrictions 
of the threads at the points of septation. 
Terminal and Intercalary Dilatations. 
I. Intercalary Globular Dilatations (Plate, fig. 3). 
This type is well shown in Plate, fig. 3, taken from the cortex of a stem 
of Lyginodendron Oldhamium (from Upper Foot Mine: Lower Coal 
Measures). A comparison of these swellings with those figured by 
Williamson (13, figs. 28, 36) for P. antiquarius shows that both belong to 
the same type, although naturally there is a marked difference in size — 
those of P. antiquarius measuring about 40 ju, whilst the largest shown in 
Plate, fig. 3, measures only 17/x. A detailed examination of these bodies was 
made on the larger representatives of P. antiquarius, these being very 
well preserved in the British Museum slides of that organism. Each 
globule is connected by an extremely short hypha with one, or sometimes 
two, or even three other globules. Their method of attachment is shown 
diagrammatically in text-fig. 2. The vesicle a is connected by short 
hypliae with three others. As seen in the diagram, this can be explained 
only on the assumption that the vesicle marked a had arisen at the junction 
of a branch with the main thread. A vesicle connected with only two 
others must have arisen along the length of the hypha. Each vesicle is in 
open communication with the attached hyphse. In P. gracilis the same 
facts may be noted with regard to these bodies, but, owing to their packed 
arrangement and smaller size, observation is more difficult. 
These bodies are referred to as oogonia both by Worthington Smith (11) 
