20 DIVISION I. —GENERAL MORPHOLOGY. 
mycelial filament according to the species is either not altered at all at the point of 
origin of a haustorium, or is at most only slightly enlarged there ; or it has a flat nearly 
semicircular protuberance the height of which is at most equal to its own diameter ; or 
it has a protuberance in the form of a bluntly lobed disk scarcely exceeding the breadth 
of the filament, which is pressed closely down on the epidermis, and appears on both 
sides or only on one side beyond the flank of the filament. These lobed attachment- 
disks were first discovered by Zanardini in-Erysiphe Tuckeri. 
The thick mycelial tubes of the Peronosporeae, which are usually without 
transverse walls and spread among the cells inside living plants, often clinging close 
to the outer surface of their walls, send haustoria into the cells, which have very 
different forms in the different species. In Cystopus (Fig. 8 A), Peronospora nivea, 
P. pygmaea, P. densa and others, 
they are like those ofthe Erysipheae 
but much smaller, and they usually 
or perhaps in all cases only make 
a deep indentation in the walls of 
the cells; in P. parasitica they are 
lobately branched tubes, and their 
vesicular club-shaped branches often 
quite fill the cells of the host; in 
most of the pleuroblastic Perono- 
sporeae (Fig. 8 3) they are slender 
filiform lateral branches of the in- 
tercellular filaments with many 
curved and winding ramifications 
in the interior of the cells. In Phy- 
tophthora infestans, which inhabits 
the potato, branches of the myce- 
lium which in this case scarcely 
Fic. 8. Mycelial tubes 7 creeping about in the intercellular spaces with deserve a separate name force their 
their haustoria penetrating into the cells z—z; A of Cystopus candidus, = . . 
from the pith of Lepidium sativum, B of Peronospora catotheca from the way at varlous spots, especially ın 
pith of 4sPerula odorata. Magn. 390 times. the sprouting tubers, into the cells 
of the host. 
The intercellular mycelium of the Uredineae has a variety of haustoria formed like 
those of the Peronosporeae, especially the pleuroblastic species, and with them should be 
mentioned also the winding intracellular mycelial branches of the Ustilagineae. 
The haustoria of Piptocephalis, Syncephalis, and Mortierella are very different 
from those which we have hitherto been considering. Piptocephalis Freseniana is 
parasitic on the larger Mucorini, and its mycelium, like that of its hosts, con- 
sists of tubes without transverse walls. If a growing filament of the mycelium 
of the parasite comes in contact with a mucor-tube, either with its apex or with 
its lateral wall, it spreads out slightly at the point of contact and thus attaches itself 
firmly as with a cupping-glass to the Mucor. A tuft of filiform radiating branching 
processes of such extreme delicacy that nothing is known of their minute structure 
now shoot forth from the middle of the surface of attachment into the cell of the host. 
The length of these suction-filaments is about equal to the diameter of the mucor-tube 
(see section XLIII). Van Tieghem and Le Monnier describe similar arrangements 
in Mortierella and Syncephalis, only in these genera the tubes which enter the cells of 
the host are not so different from those of the rest of the mycelium. 
An allied case, though with important points of difference, is that of the clusters of 
haustoria in Chaetocladium Jonesii, a form which is usually parasitic on species 
of Mucor like the genera just described, and resembles them in structure. The tubes 
of this Fungus, both of its mycelium which spreads in the substratum and of the part’ of 
its thallus which rises above it, become firmly attached at the point of contact to the 

