1914J FROMME—AECIDIUM CUP 29 



Endo. Semperoivi by Hoffman. Although I have found no such 

 cells in any of the aecidia included in this study, I have found a 

 single case of a basal cell that seems to have given rise to a branch 

 in the aecidium of Aecidium Dicentrae Trel. Only a few sections 

 of this form have been examined, and hence no estimate can be made 

 as to the prevalence of the condition. In the single case seen, the 

 second basal cell seems to have arisen as a lateral branch of the 

 primary basal cell, and the opening between the two has not been 

 closed by wall-formation. Both basal cells have produced a chain 

 of spores. 



Hoffman compares these branching basal cells to the basal 

 cell of the primary uredosorus as described by Christman. The 

 two cases, however, are not at all comparable. The basal cell of 

 the primary uredosorus gives rise to successive stalked spores that 

 originate as buds on the upper surface of the cell. The comparison 

 is made, therefore, between a bud that produces a stalked spore in 

 one case and a basal cell in the other. As Christman has pointed 

 out, the homology lies between the primary uredospore and stalk 

 cell and the aecidiospore and intercalary cell. The homology is 

 also extended to include the basal cells of the two sori. 



According to Christman, the basal cell is the true morphological 

 unit, and the basal cells of the different sori are to be considered as 

 homologous. This, however, involves the difficulty, which Christ- 

 man recognized, that the basal cells of the uredosorus and teleuto- 

 sorus are not the outgrowth of fusion cells, as are those of the 

 primary uredo and aecidium, but are borne on a mycelium of binu- 

 cleated cells. The basal cell of the micro-forms is also without 

 doubt to be homologized with that of the aecidium, since Olive's 

 work on P. transformans and that of Werth and Ludwigs on P. 

 Malvacearum indicate that the basal cell of these forms is the out- 

 growth of a fusion cell. 



The evident homology between the primary uredospore and the 



t> 



aecidiospore is used by Christman as a strong argument that the 

 eu-type of rust with catenulate aecidiospores cannot be considered 

 as a primitive type, but is rather a highly specialized and derived 

 type. If the homology is to be accepted, and it seems to have been 

 very generally, the catenulate method of aecidiospore-production 



