So 



PLECTOMYCETES 



[CH. 



enter the stomata, extend through the intercellular spaces and send haustoria 

 into the neighbouring cells, and in Erysiphe (or Oidiopsis} taurica the whole 

 mycelium during the conidial stage is located in the^ tissues of the host. 

 We have thus, within the limits of the family, a transition between ecto- and 

 endoparasitism through hemiendophytic forms, and forms which are endo- 

 phytic under abnormal conditions. When perithecia are about to be produced 

 and the mycelium emerges and spreads over the surface of the host leaf, the 

 hyphae both o>i Pliyllactinia and of E. taurica showhaustorial branches (ad- 

 pressoria), though no haustoria are produced. It may be inferred that the 

 ectophytic condition with haustoria penetrating the epidermal cells is primi- 

 tive in the group. 



Indian and Persian specimens of E. taurica have been found under 

 practically desert conditions, others have been collected on plants of the 

 steppes of Turkestan at a height of 6000 feet, and in localities exposed to 

 very dry winds. The suggestion has consequently been made that the 

 endophytic habit in this family is an adaptation to xerophytic conditions, 

 since it both provides shelter for the developing mycelium and obviates the 

 necessity of piercing through the cuticle, which in desert plants is of 

 considerable thickness. 



The Erysiphaceae are propagated during the summer by rather large 

 oval uninucleate conidia (fig. 38). These are ordinarily produced in rows 



on simple conidiophores with one or 

 more basal cells. In the endophytic 

 E. taurica, however, the conidia are 

 borne singly on branched conidio- 

 phores which emerge through the 

 stomata of the host. 



In the case of Phyllactinia Cory- 

 lea, which is met with on a large 

 number of deciduous trees, variations 

 occur in the shape of the conidia 

 borne on different hosts, and indicate 

 the existence of morphological dis- 

 tinctions between the biological 

 forms of the species. 



Before the connection between the 

 conidia and the perithecia of the 

 Erysiphaceae was understood, the 

 nenc name, Oidium, was applied to the former. The name is still used 

 > indicate the characteristic form of the conidial stage and to describe 

 dia when the perithecia are unknown. This was the case with the 

 powdery vine mildew. The conidial form, known as Oidium Tuckeri became 



Fig. 38. Sphaerotheca pannosa Wallr. ; conidio- 

 phores and conidia, x 240. 



