500 H. S. Jennings and George T. Hargitt. 



a ''wild" culture. These races, as we shall see later, have bred 

 true to their diverse sizes for periods up to three years. The mi- 

 cro-nuclear relations were determined quite independently of any 

 knowledge of the relative sizes or other distinguishing features of 

 the races, in the following way : The samples given me from which 

 to start my cultures were designated by letters only. I received 

 seven of these samples, labeled respectively Ci, Li, D, ko, khaiaz, 

 c, i. I was quite unfamiliar with the history and characteris- 

 tics of these various lines, and only after the number of micro- 

 nuclei was determined for each was reference had to the literature 

 on the subject or comparisons instituted with regard to size and 

 number of nuclei and the like. Thus the facts were first deter- 

 mined quite independently of any possible interpretations. 



The animals are too thick to allow of the easy observation of 

 micro-nuclei upon whole specimens even when stained and cleared. 

 Therefore it was decided that sections should be used. At the 

 first several killing fluids were tried, picric acid mixtures, mer- 

 curic chloride mixtures, formalin, osmic acid mixtures, etc.; 

 also various staining agents were tried with each killing fluid. 

 No essential difference was observed in the various methods and a 

 single plan was therefore followed throughout. For killing, Wor- 

 cester's fluid (10 per cent formalin saturated with mercuric 

 chloride) was used hot for a few minutes, and the sections stained 

 with Heidenhain's iron-hematoxylin, or with Ehrlich's hematoxy- 

 lin and eosin. The procedure was identical in all cases. By kill- 

 ing the animals just before they divided, the micro-nuclei were 

 usually found to have moved away from the meganucleus and to 

 have begun their preparation for the ensuing division, or perhaps 

 to be in the act of dividing. This made it possible to determine 

 whether division was progressing normally, and if two micro- 

 nuclei were present to determine whether and how both were 

 dividing. Also it gave a possible chance of finding how the assump- 

 tion of two micro-nuclei in a race with only one as a rule came about, 

 if indeed such a thing did occur. For each race, therefore, speci- 

 mens dividing or about to divide were used; for comparison 

 specimens from each race not in the dividing condition were 

 sectioned. 



