126 FROMME: SEXUAL FUSIONS IN FLAX RUST 
in somewhat irregular knoblike cells, which are usually somewhat 
larger that the uredospores. The walls are somewhat thicker 
than those of the uredospores and perhaps a little smoother, 
although the older ones are roughened like the uredospores. 
Young paraphyses were found to contain two nuclei at the base 
of the apical knob. See Fic. 30. The stalks were undivided 
and very slender and contained but very little cytoplasm. The 
nuclei evidently degenerate very rapidly, as in the older filaments 
they have disappeared entirely. Of the cytoplasmic contents 
only a few shriveled threads remain. The nature of theée sterile 
filaments is perhaps not entirely clear, but the observation that 
they are binucleated when young suggests that they have a 
common origin with the functional uredospores. 
TELEUTO 
The teleutospores appear later in the season than the uredo- 
spores and are associated with the ripening of the flax. The sori are 
found on both sides of the leaves and more frequently on the stems. 
They are sometimes round and isolated but are more often con- 
fluent in long areas, which are reddish brown when young and 
become quite black when old. The spores are of the elongated 
one-celled type of the Melampsoras and are cemented together 
above in a firm, waxy layer. See FIG. 31. They are sessile on 
the short cubical cells at the base of the sorus, which lose their 
contents with the development of the teleutospores. The young 
teleutospores are binucleated but soon become uninucleated 
through a fusion of the two nuclei into one. Occasionally a 
secondary layer of teleutospores forms an overgrowth above the 
first. This is apparently brought about by the growing up of the 
mycelium around the borders of the primary layer of teleutospores 
to form a secondary layer of shorter teleutospores above the first. 
DIscussION 
The existence of sexuality in the rusts may be considered as 
well established for those forms that have the caeoma type of 
aecidium. The cup type of aecidium, however, needs much 
further study, though Christman’s figure for Uromyces Caladit 
shows that the fusions there are probably essentially similar to 
