o74 



'IHK AMETITC {\ \ATUT{\U^T [Vor LV 



jiiilhors are teased out iii oi'.liiKiry arcto-caniiiiic, avoiding the 

 us(> of steel iiistrnments. 



The ehromosomes are usually most .accurately counted in the 

 nietaphase of the second division, in dicotyledons. When the 

 preparation is a day or two old, the cytoplasm has swollen ; and 

 a sli{?ht tap on the thin coverglass above any particular cell will 

 usually free the cytoplasm from tlic cell wall, and another tap 

 flatten it out with its contained clu'oiiiosonics. 



Satisfactory results have been obtained by these mdliods dur- 

 incf the past year with Datura, Cannn, AnI trrh niton, Linaria, 

 Brassica, Dahlia, Sccale, Asparagus, Matthiola, Pkascolus, Stizo- 

 lohium, Tmdescantia, Hemerocallis, Iris, Gladiolus, Zea and 

 Portulaca. The methods failed with (Enothera and Rfiodo- 

 dendron. 



The second of the above methods will probably be of the 

 widest applicability. The preparations will keep for a week or 

 more, if an excess of stain and of iron are avoided. The method 

 is quicker for counting ehromosomes than staining sections with 

 iron htematoxylin, and in favorable eases the results may be 

 more certain. Thus in good preparations of Datura over a 

 thousand pollen-mother cells are scattered singly on one slide, 

 many of tlit'ui shownig the metaphase of the second division, and 

 som.' liavmu' hotli plates m one plane, with the chromosomes well 



