164 ‘BOTANICAL GAZETTE [FEBRUARY 
very interesting endophytes or ‘‘Raumparasiten,”’ which have been 
made known principally by the researches of KLEBs (13). The 
climax of this series is found in Phyllobium dimorphum, which 
penetrates dying leaves of Lysimachia nummularia. Its adult 
body is strikingly similar to that of Rhodochytrium. There is a 
long empty neck, with an external cellulose button connecting the 
internal cyst with the wall of the zoospore from which it developed, 
just as in Rhodochytrium. In its most typical development this 
plant is confined to the vascular bundles of its host, into which it 
penetrates very much as does Rhodochytrium. It sends out, more- 
over, numerous interlacing rhizoids, which follow along the bundles 
for considerable distances, and even extend up their branches. On 
germination the resting cysts give rise to biciliate zoospores which 
conjugate as in Rhodochytrium, except that there is a slight sexual 
differentiation, microzoospores and megazoospores being formed in 
different cysts. The cysts and the zoospores have abundant 
chlorophyll, but haematochrome is also present in considerable 
amounts in some stages of the life cycle. Little is known of the 
finer structure or cytology® of this plant, but, so far as one can 
judge from the evidence available, it is remarkably close to Rhodo- 
chytrium. The most important difference between them would 
seem to be the presence of chlorophyll in the one and its absence 
in the other. OLTMANNs (21, pp. 322 ff.) believes that these forms 
belong to a natural series. He agrees with LAGERHEIM that 
Rhodochytrium is an alga, saying ‘‘while the first named genus 
[Phyllobium| cannot be considered more than an endophyte, as we 
have already clearly demonstrated, Rhodochytrium is one of the rare 
examples of an alga which has lost its chlorophyll on account of 
parasitism.” 
It will be seen, therefore, that the decision as to whether Rhodo- 
chytrium is an alga or a fungus depends upon the criteria by which 
the line between them is to be drawn. If the question is to be 
settled by definition, we should follow VurLLemIN (32) and call it 
a fungus, for it would be very difficult to frame a definition of the 
fungi which would exclude Rhodochytrium. This position is also 
6 OLTMANNS states, on the basis of unpublished observations by GRUEBER, that 
the cyst is uninucleate. 
