8 KANSAS UNIVERSITY SCIENCE BULLETIN. 



had formed. A culture coming from an isolated mycelium 

 soon reverted to the ordinary branched and unbranched forms, 

 though the mycelium type persisted long enough to show 

 a tendency to heredity. The appearance of these mycelia, 

 apparently, does not depend on temperature or the nature of 

 the substratum. Higher temperatures seem to favor the ap- 

 pearance of branched forms. The author holds that these 

 variations represent new characteristics and are not to be re- 

 ferred to atavism. 



R. Massim ( 1906) , working with a pure culture of Bacillus 

 coli mutabilis, found that colonies remained white on Endo 

 agar, indicating lack of power to ferment lactose. Trans- 

 plantations of young colonies continually gave white colonies 

 on this medium, but transfers from older colonies sometimes 

 gave a proportion of distinctly red colonies, which remained 

 red on further transplantation. These red colonies he sup- 

 poses to arise by mutation in the sense of de Vries. 



I. EXPERIMENTS ON YEAST. 



In my own work on Saccharomyces anomalus, I have made 

 use of a culture kindly furnished me by Professor Freeman, 

 of the University of Minnesota, a culture which originally 

 came from Doctor Barker, of England. My researches were 

 conducted in two directions : First, the selection of cells 

 varying from the normal in size; second, the selection of 

 cells varying in form. 



In the first series I attempted to obtain a race exhibiting 

 cells permanently larger than the normal by repeated selec- 

 tion of cells of unusual size. As in all experiments made in 

 the course of this work, the series was begun with a pure cul- 

 ture proceeding from a single isolated normal cell. Cultures 

 were made for the most part in glucose bouillon in hanging 

 drops, and the isolated cells were grownin the same medium. 

 In conducting these experiments, a cell, considerably larger 

 than the normal, was isolated, and, after a considerable num- 

 ber, often hundreds, of offspring had been formed, a second 

 large cell was isolated from these, and so on. A check con- 

 sisting of unselected cells was frequently compared under 

 similar conditions. 



In one series this repeated selection was practiced twelve 



