Nuclear Divisions in the Rusts. 
349 
the latter’s conception of fertilization by means of £ nuclear migration ’ has, 
in my opinion, some basis in fact, since, during the earlier stages of cell 
fusion, one of the two nuclei may undoubtedly sometimes be seen to pass 
through a very narrow opening into the adjoining gamete (Figs. 33 and 34). 
But a multitude of other cases in which a broad fusion pore is formed 
(Figs. 26-33, 37-40) proves that the nuclei are not alone involved, and that 
the cytoplasm of the two cells also fuses into one mass. As indicated 
above, the occurrence of both the Blackman and Christman types of fusion 
(Figs. 34 and 36) near each other in the same preparation shows con- 
clusively that appearances which suggest a ‘nuclear migration’ through 
a narrow opening may occur adjacent to a clear case of cell fusion in which 
a broad conjugation pore has been formed. Since the term ‘ nuclear 
migration ’ has been recently used extensively for undoubted pathological 
cases in which nuclei are found passing through minute pores in the 
partition wall ; and, further, since Christman (’07 2 ) has shown that similar 
pathological migrations may sometimes occur between sporophytic cells 
in Puccinia Podophylli , it seems desirable not to employ the expression 
‘ nuclear migration ’ for those cases in which, during the earlier part of the 
sexual cell fusions, the nuclei are seen to pass through narrow openings 
into the adjoining gametes. Such true sexual fusions as are shown in 
Figs. 33 and 34, as well as in Blackman’s Figs. 66 and 68, certainly 
resemble strikingly the pathological nuclear migrations figured by Miehe 
(’01) and Christman (’07 2 ). But the subsequent history of the binucleated 
series of cells thus sometimes initiated in the Rusts undoubtedly excludes 
any interpretation of a pathological nature for such cases. To bring such 
cases into line with Christman’s interpretation of a cell fusion, the ex- 
planation indicated above appears reasonable, viz. that sometimes the 
fusion of the two gametes begins through a narrow opening, so that the 
nucleus in passing through the pore becomes much elongated and con- 
stricted. Thus, in the light of the further facts brought out in this paper, 
the seemingly divergent accounts of Blackman and Christman become, 
in my opinion, easily reconciled, at least for the diffuse caeoma type of 
Rusts. But I wish to emphasize at this point Christman’s word of caution 
as to the probable occurrence of nuclear migrations of an undoubted 
pathological nature in the cup-like aecidia, since I have found a number 
of cases of migrating nuclei between the multinucleated cells of Puccinia 
Cirsii-lanceolata. Much still remains to be explained with respect to the 
aecidium-cup ; and in view of the probably regular occurrence of multi- 
nucleated cells in young stages of the cup, it seems certain that neither 
Blackman’s (’06) ‘ nuclear migrations ’ nor Christman’s (’05) cell fusions 
at the base of every spore- row in Uromyces Caladii will serve to explain 
all of the complications which appear to prevail in this more highly 
differentiated type of Rust. 
B b 2 
