32 BULLETIN 380_, U. S. DEPARTMENT OF AGRICULTURE. 



PYCNOSPORES. 



The pycnospores of all the species are oblong elliptic to cylindric 

 in shape and so small as to make accurate measurement very difficult. 

 Slight but apparently constant differences in their size in certain 

 groups of species may, however, be traced. These differences are 

 clearly shown in Table I. 



EndotJiia gyrosa, E. singularis^ and E. longirostris have smaller 

 pycnospores than the other species, the most frequent lengths being 

 3 and 3.5 \x. The pycnospores of E. singularis are slightly broader 

 than those of E. gyrosa and E. longirostris^ being 1.5 to 2 [a, as 

 against 1 to 1.5 ]k in the last two species. 



Endothia fiuens, E. fluens mlssissippiensis, and E. parasitica are 

 even more closely similar in the size of their pycnospores than in that 

 of their ascospores, the most frequent size being 4 by 2 [jl. The 

 pycnospores of E. tropicalis are much larger and more variable in 

 size and shape than those of other species. They range from 3.5 

 to 7 [Ji. in length and from 1.5 to 2.5 [/. in width. 



ASCI. 



The writers have not attempted a study of the origin and early 

 development of perithecia or asci in any of the species of Endothia. 

 Work on this subject has been published by Anderson and Rankin 

 (6), for Endothia parasitica, but the nuclear phenomena and origin 

 and development of the ascogenous hyphge are not yet entirely clear. 

 The part termed a trichogyne by these authors seems more likely to 

 be the initial stage in the development of the neck of the perithecium 

 than the relic of an organ of fertilization. 



The asci appear almost or quite sessile in most species, and 

 though varying considerably in size and shape, as indicated in Table 

 I, are usually oblong elliptic or subclavate, having a sort of inner 

 membrane inclosing the ascospores and some thin granular matter 

 extending to the apex of the ascus, where a slight thickening appears, 

 as described and illustrated by Anderson for Endothia parasitica. 

 A similar condition is found in various species of Pyrenomycetes and 

 probably functions in some way in connection with the discharge of 

 the ascospores. The asci are generally wider and slightly longer in 

 E. parasitica than in E. f.uens and other members of section 2. The 

 asci of E. gyrosa are shorter than those of any other species. E. 

 tropicalis has the longest asci. The asci of none of the species show 

 a very wide range of variation, as Table I also indicates. . 



PARAPHYSES. 



Most students of Endothia have reported paraphyses wanting in 

 this genus. Anderson (1, p. 33, fig. 32) and Anderson and Rankin 

 (6, p. 579, fig. 83) report paraphyses present and figure what they 



