84 BOTANICAL GAZETTE [august 



and a simultaneous division occurs. A cell wall is then formed which 

 separates thfe top or spore cell from the pedicel, Sappin-Trouffy 

 (25) later describes a similar budding in the urcdospore pustule of 

 Uromyces hetae Pers. The process is the same except that a conjugate 

 division of the two nuclei in the basal cell provides two nuclei for the 

 young bud, instead of the basal cell sending one of its two nuclei into 

 the bud. In my opinion it is altogether likely that the nuclear behavior 

 in Gymnosporangium sahinae is the same as that described for Uromy- 

 ces hetae. In connection with his account of Uromyces hetae^ Sappin- 

 Trouffy points out the morphological likeness between the inter- 

 calary cell of the aecidium and the stalk cells of the uredospores, 

 Sappin-Trouffy (25) also shows that the cell at the base of the 

 row of aecidiospores remains binucleated during spore-formation. 

 The nuclear phenomena found in Uromyces erythronii DC- 

 are different from those described by Rosen. The two nuclei 

 which normally lie irregularly placed in the end cell of the "spor- 

 iferous filament" come to lie side by side just before division. 

 Simultaneous division of the two nuclei provides two nuclei for the 

 cell which is to be separated off and leaves two to remain in the end 

 cell of the hypha. Sappin-Trouffy finds that in a great many rusts 

 the binucleated phase has its beginning in these end cells. In his 

 account there is no mention made of a large central organ such as w^as 

 described by Massee. His observ^ations, like Rosen's, would indi- 

 cate that the rows of spores are perfectly independent structures. 



PoiRAULT and Raciborski (21) gave a similar account of the 

 behavior of the cells at the base of the aecidium. They believe that 

 a single spindle is formed in the process of conjugate division, and 

 with the work of Sappin-Trouffy mentioned above have established 

 quite satisfactorily that the two nuclei of the cells of the uredo and 

 teleuto stages have each a distinct line of ancestors dating back to 

 the basal cell of the aecidium at least. 



In 1896 Richards (22), working on Uromyces caladii FarL, found 

 a large hypha, which he called the fertile hypha, at the base of the 

 young aecidium. This hypha gives rise to several short branches, on 

 the ends of which are borne the rows of spores. The "basidium,'' 

 as he calls the end cell of one of these short branches, contains two 

 or more nuclei; one migrates to the apex and there divides. The 



