ENDOTHIA PAEASITICA AND RELATED SPECIES. 29 



Development of the stromata, The writers have not followed the 

 development of the stromata in culture, but an examination of nu- 

 merous sections of Endothia singularis, E.gyrosa, E.fluens, and E. 

 parasitica and a study of the three latter species under field condi- 

 tions on various hosts shows that their development is by no means 

 as uniform as indicated in Anderson's description (1). 



According to Anderson, the pycnidium develops first, and about 

 the young pycnidium the stroma is quickly formed, while the 

 perithecia arise later, usually in the lower portion of the stroma. 

 This may perhaps be considered the typical course of development, 

 and pycnidia are often found above the perithecia, but all variations 

 occur. A large stroma may be developed without a sign of a pyc- 

 nidium (PL XV, fig. 2). In some cases there is a considerable por- 

 tion of the stroma above the pycnidial cavity (PL XIV, fig. 2), or 

 the pycnidial cavities may be surrounded by a thick stroma (PL 

 XIV, fig. 4, and PL XV, fig. 1). Sometimes, on the other hand, 

 they are large and irregular, with little stroma (PL XV, fig. 3). 



The perithecia by no means uniformly arise below the pycnidia, 

 but the two often occur side by side in the same stroma (PL IX, 

 fig. 2; PL XIV, fig. 3; and PL XII). Sometimes, even, the 

 perithecia are above the pycnidia (PL XIV, fig. 2). There seems 

 to be no constant relation either as to the relative number of pyc- 

 nidia or of perithecia in a single stroma. Sometimes the pycnidial 

 portion is much larger (PL IX, fig. 1) ; sometimes the perithecia 

 predominate (PL X, fig. 2) ; and sometimes the two portions are 

 practically equal (PL XII). 



A like variability apparently occurs in the sequence of the fruit- 

 ing bodies. As the figures show, the pycnidia sometimes develop 

 after the perithecia; the reverse order is frequent; while in several 

 sections (PL XII, and PL XIV, fig. 3) the two types of fruiting 

 bodies were side by side and were producing mature spores abun- 

 dantly at the same time. Just what factors determine the produc- 

 tion of each type of spore or prevent or delay spore production is 

 unknown. It seems probable, however, that climatic influences may 

 prevent the development of ascospores in many cases. The action 

 of climate may be very indirect, however, for no ascospores of any 

 species have yet been obtained in artificial cultures, though En- 

 dothia fiuens, E. fluens mississippiensis, E. tropicalis, and E. para- 

 sitica produce pycnospores abundantly on a variety of media. Cer- 

 tainly, climatic factors would not account satisfactorily for the fact 

 that pycnidia and perithecia are produced at the same time in ad- 

 jacent stromata, or even in different parts of the same stroma. 



The size of the perithecia is rather uniform in the various species 

 (PL X, fig. 3, and Pis. XI and XVI), being about 0.35 mm. in diame- 

 ter. They are typically globose to pyriform, but are usually more or 



