648 Journal of Agricultural Research voi. xv. no. 12 



perhaps the conditions in the rust nucleus approach more nearly than has 

 been supposed the conditions in the Ascomycetes and other fungi. 

 The process of nuclear fusion in the young teliospore is comparable 

 to the same process in other rusts. It seems reasonable to conclude 

 that this fusion is the completion of the process, initiated at the time 

 of cell fusion to form the dikaryon, which takes the place of normal 

 fertilization, and that the actual fusing is a necessary preliminary to 

 the mixing of the chromatin elements and the subsequent reduction 

 division, as suggested by Maire {31) and others. Certainly the changes 

 which take place in the nucleus after fusion suggest a complicated 

 mingling and readjustment of the chromatin which would seem to justify 

 such a view. It is not necessary to regard the fusion as a pseudo fer- 

 tilization {10). There appears to be little reason for doubting that the 

 first division in the promycelium is hetero typical {38) , for it is unique and 

 decidedly dififerent from the second division which immediately follows, 

 Arnaud (i) has compared the mitotic figures in Capnodium meridionale 

 with those in Coleosporium senecionis. Wager's figure 84, Plate 19 (63), 

 of the telophase in the dividing fusion nucleus in the sporangium of 

 Polyphagus euglenae is very like similar stages in the primary division in 

 the promyceUum of Cronartium rihicola. It may be noted that the asso- 

 ciation of the nuclei in P. euglenae and their subsequent fusion in the 

 sporangium are phenomena comparable to the nuclear conditions in the 

 rust dikaryon and fusion in the teliospore. 



ABNORMALITIES 



Before coming to the general discussion and summary it is necessary 

 to mention briefly certain abnormalities commonly met in the different 

 types of sori. .^cia sometimes occur with reversed polarity in part 

 of the sorus, where the spores are produced on aecial chains which grow 

 toward the center of the tree, usually into a resin duct. This change 

 in the direction of growth is probably to be explained by the fact that 

 the developing chains followed the fine of least resistance, in this case 

 into the adjacent resin duct, instead of against the overlying host cells. 

 iEcia also often develop on the roots, under several inches of leaves and 

 loam. Their structure appears to be normal, but their environment 

 is hardly advantageous. Double pycnial layers (PI. 50, C, a, a J are 

 not imcommon. In a short note, Posey, Gravatt, and Colley (45) have 

 reported the finding of uredinia on the stems of Rihes hirtellum Michx. 

 In the cortex of infected stems of this species internal uredinia (PI. 

 5T, C) with normal and reversed polarity were formed in abundance. 

 Internal telia, produced in the pith and cortex of the petioles of Ribes 

 sp.,^ have been described in a previous paper (<?). In such abnormal 



1 Ribes roezli was the name given in the original article, but the species determination was probably 

 incorrect. As the species has not fruited, accurate determination has been impossible. 



