132 Proceedings of the Royal Society of Edinburgh. [Sess. 
Proofs of the Fungal Nature of the Threads. 
1. In size and form they resemble modern hyphae. 
2. They exhibit a sinuousness characteristic of modern hyphae. 
3. They branch monopod ially. 
4. Attached to them are spherical dilatations similar to those found in 
organic connection with the hyphae of modern fungi. 
5. Their structure is tubular. 
6. They show selective power, some tissues of the host being devoid of 
the threads, whilst others possess them in abundance. 
7. They are found in places where fungi, if fossilised, would be 
expected to occur. 
Habit. 
In the first of the three slides the hyphae and attached vesicles are 
mainly confined to the sclerotic portion of the cortical cells, the leaf-bases 
and the Stigmarian rootlets being free of the fungus. In the second slide, 
showing a young shoot of Lepidodendron Harcourtii in transverse section, 
the hyphal threads are abundant in the cortical cells but absent from the 
stelar tissue. 
In the third slide the hyphae occur abundantly in the cortex of a 
rootlet of Lyginodendron Oldhamium ( Kaloxylon Hookeri), but also they 
are not wanting in the stele, in which both hyphae and vesicles were found. 
Renault states that the fossil was found by him in the cortical cells of 
Lepidodendroid plants. It seems therefore that the habitat lay chiefly in 
the cortical cells of the stems and roots of various plants, but that in the 
case of young plants the stelar cells were also affected. 
The cell-walls of the host were, to all appearance, very little affected by 
the inroads of the fungus, differing in this respect very markedly from the 
cell-walls of the tissues that had been attacked by the fungus described by 
Cash and Hick (1). In the latter case the fungus was indubitably a 
saprophyte, for the number of hyphae in the host-cells is so great that it 
seems a physiological impossibility that a living organism could support 
so many. The presence of hyphae in a host with more or less intact cell- 
walls is presumptive though not conclusive evidence of the parasitic 
nature of the invading organism. The inference, moreover, is strengthened 
by the fact that all the modern fungi to which this fossil fungus is related — 
viz. the Peronosporaceae — are, without exception, parasitic on their hosts. 
Peronosporites gracilis was, further, endophytic in habit, its whole life 
having been passed, so far as can be ascertained, inside the host. The 
