Holden and Harper—Nuclear Phenomena* . 73 
of the two nuclei). The upper of the two nuclei in Fig. 23 is 
in a late prophase stage. The polar asters are present but the 
outline of the nucleus between them is still intact. After the 
second division of the nuclei cell division follows immediately 
and we have the four celled promyceliumi. 
If we compare now figures, IS, 19, and 20, with those of 
Sappin-Trouffy and Poirault and Eaciborski, it becomes evi¬ 
dent that the contention of these authors that but two chro¬ 
mosomes are present at this stage in the development of the 
rusts is not supported by the facts as found in Coleosporium 
sonchi-arvensis. The curiously abnormal mitotic figure of 
Coleosporium Euplirasiae as given by Poirault and Eaciborski, 
consisting of only two long irregularly lobed threads, bears 
scarcely the faintest resemblance to the corresponding stages as 
we find them in Coleosporium' sonchi-arvensis. That such ex¬ 
treme differences can exist in such closely related forms is 
hardly to be believed and as the figures in our own preparations 
correspond so much more nearly with those in other fungi and 
algae it seems probable that the preparations of Poirault 
and Eaciborski were distorted in fixation. As noted above it is 
not possible to give the number of chromosomes in Coleospor¬ 
ium sonchi-arvensis with exactness but it is plain that there are 
many more than two, and in view of the unusual appearance of 
the figures given by Poirault and Eaciborski, it is very ques¬ 
tionable whether they have determined the number correctly 
for Coleosporium Euplirasiae. 
The haustoria of C. sonchi-arvensis are normally binucue- 
ated, while the portion of the cell of which the haustorium is 
a part but which is outside of the cell of the host plant, usu¬ 
ally contains no nucleus (Fig. 26). Eosen and later Sappin- 
Trouffy have maintained that the haustoria apparently seek out 
the nuclei of the host plant and apply themselves to its surface. 
There is little evidence of this iu Coleosporium sonchi-arven¬ 
sis. A host cell may contain, several haustoria but they all 
seem indifferent to the host cell nucleus. The haustorium is 
always the end of a hvphal cell, a large portion of which is out¬ 
side the host cell. As a rule the haustoria of Coleosporium 
sonchi-arvensis are oval in form (Figs. 25 and 26). The 
