ENDOTHIA PARASITICA AND RELATED SPECIES. 33 



regard as an early stage of their development. They describe them 

 as branching frequently and very crooked, extending around the 

 perithecium as well as upward. The writers have searched in all the 

 species studied for evidence of the presence of paraphyses, but have 

 never seen anything resembling paraphyses as they occur in closely 

 related Pyrenomycetes. If they occur, they would seem to be of an 

 unusual character and difficult to recognize or else are evanescent, 

 disappearing before the asci are mature. 



ASCOSPORES. 



The ascospores furnish one of the most marked characters for the 

 separation of the genus into sections (Plate XVII). In section 1 

 they are more or less cylindric and sometimes curved. In section 

 2 they are more or less elliptic, being broadest in EndotMa para- 

 sitica and narrowest in E. fluens and E. longirostris. The greatest 

 variation in size and shape of ascospores occurs in E. fluens, as in- 

 dicated by the measurements given in Table II. Anderson (1) , Clin- 

 ton (18), and Heald (39) describe and figure the ascospores of E. 

 parasitica as very obtuse and constricted at the septum. The writers 

 have but rarely seen spores of this form. This may perhaps be due 

 in part to different methods of treatment or to the age and condition 

 of the material. Most of the ascospores studied by the writers have 

 been mounted in the fluid medium described on page 30. Fresh 

 specimens have also been studied in water mounts, but with the 

 same general result. The writers are of the opinion, therefore, 

 that the figures of the authors cited above do not represent the most 

 common and characteristic form of ascospores of this species. (Com- 

 pare Plate XVII, figs. 7 to 15.) 

 43737 Bull. 38017 3 



