Nuclear Divisions in the Rusts. 
357 
thready structure at one side of the chromatic mass. Mantle fibres now 
extend out from the diverging polar centrosomes to the chromatin and 
accomplish the division into the two daughter-halves. During the later 
phases, a few polar radiations generally appear. 
While the chromosomes in the case of Triphragmium are not sufficiently 
differentiated in the equatorial-plate stages to be definitely counted, during 
the anaphases, however, the chromatic radiations which converge to the 
poles become quite distinct and are then seen to number eight. These 
chromatic strands are regarded as corresponding to chromosomes. Further, 
they are often seen to be segregated into two distinct groups of four each, 
and each group is seen to be attached to a distinct centrosome. This 
double character of the daughter-nucleus is interpreted as expressing in 
somewhat clearer terms the doubling noted by several earlier writers. 
Rather, however, than denoting the longitudinal splitting of a single centro- 
some, as held by Poirault and Raciborski ; or the existence side by side in 
the one nucleus of two distinct chromosomes (Sappin-Trouffy and Maire) ; 
or the presence in the same reduced nucleus of the chromatin derived from 
the two parents (Blackman), it is suggested that the double character may 
result from a precocious division of the centre, in preparation for the next 
mitosis. 
3. The apparently normal and regular occurrence at the base of certain 
young aecidia of one to many multinucleated cells, points to the necessity 
of a broader conception as to the mode of development of the aecidium-cup 
than that held by either Blackman or Christman. While the part which 
these multinucleated cells take in the development of the aecidium is 
as yet somewhat obscure, the evidence appears to point to the conclusion 
that they are sporophytic structures and that they result from the stimulated 
growth which follows the sexual cell fusions. Should this prove true, it is 
obvious that the ‘ fusion cell ’ does not at once function as a ‘ basal cell at 
the bottom of each spore-row, as maintained by Christman for this type of 
Rust. Further, the occurrence of occasional instances suggesting ‘ nuclear 
migrations undoubtedly of a pathological nature, between the multi- 
nucleated cells of Puccinia Cirsii-lanceolati , throw doubt on the idea as 
to the normal origin of the binucleated condition in the aecidium-cup 
by this means. 1 * * 
State College of Agriculture and Mechanic Arts, 
Brookings, South Dakota. 
November , 1907. 
1 Since the above went to press, two papers on the subject (The Relation of 4 Conjugation ’ and 
‘ Nuclear Migration ’ in the Rusts ; and The Relationships of the Aecidium-cup Type of Rust) 
were read by the writer before the Chicago meeting (1908) of the American Association for the 
Advancement of Science, abstracts of which are to be found in Science XXVII : 213-15, Feb. 7, 1908. 
