438 J. MCA. KATER. 



nucleus) in trypanosomes and shows that without a favorable 

 modification of technique the problem is almost invincible. 



Some of the early contributors to our knowledge of the cytology 

 of yeasts unquestionably saw and illustrated, with a fair degree 

 of accuracy, not only the resting nucleus but also stages in its 

 division. Janssens (1902) and Janssens and Leblanc (1898) 

 considered the division of the nucleus to be an intermediate form 

 of mitosis. Swellengrebel (1905) and Fuhrmann (1906) as a 

 true mitotic process. Although these last two articles are in 

 the main correct they have not been generally approved, and 

 the ideas of Guilliermond (1904, '12, '17 and '19), which gain 

 weight by the mere bulk of his work on yeast, seem to meet with 

 more favor. This author, who is responsible for a large part of 

 our knowledge of the well developed sexuality of yeasts and for 

 an excellent account of the typical metazoan mitosis found in 

 spore formation in Schizosaccharomyces octosporus, maintains that 

 in bud formation the nucleus divides by a process identical with 

 amitosis in the tissue cells of higher organisms, where, as Conklin 

 asserts, it is not a reproductive phenomenon at all. To accept 

 this would be to admit that mitosis and amitosis are funda- 

 mentally alike and interchangeable. This would undermine a 

 large part of our knowledge of cytology and genetics. The 

 problem most assuredly warrants critical study. 



METHODS. 



Pure cultures of Saccharomyces cervicix were used for this 

 work. They were cultivated on both liquid and solid media. 

 French proof broth was used for the liquid medium, French 

 proof agar for the solid. 



The organisms were transferred to slides which had been 

 previously smeared with albumen fixative and the moist films 

 were fixed either in corrosive-acetic-alcohol (95 per cent, alcohol 

 saturated with mercuric chloride 95 parts, glacial acetic acid 5 

 parts) or Bouin's solution (saturated picric acid solution 75 

 parts, formalin 20 parts, glacial acetic acid 5 parts). Iron-alum- 

 haematoxylin counterstained with light green or not counter- 

 stained at all was found to be the best means of staining. 

 Delafield's haematoxylin and carbol fuchsin were tried without 



