288 University of California Puhlications in Botany [Vol. 5 



as due to the modification of general physiological conditions, in many 

 instances probably to nothing more than variations in the hydration 

 of cell colloids. 



The results of the experiments above described appeared to indi- 

 cate that some phy^siological balance of the plant, a disturbance of 

 which causes flower-fall, was sufficiently delicate and responsive to 

 variations in nutrition to make of value the experiments noted below 

 in which the parent was grafted on the hybrid and vice versa. A very 

 considerable number of efforts were made to secure graft unions 

 between FiH38, N. sylvestris, and N. Tab. var. macrophylla. All 

 attempts to graft F^HSS and N. sylvestris using the former both as 

 scion and stock have been to date unsuccessful. A number of grafts 

 of F^HSS and N. Tab. var. macrophylla have been successfully pro- 

 duced, though the operation is much more certain to succeed when the 

 N. Tahacum parent is used as the scion. It was held as possible that 

 the hybrid growing on the parent might produce a larger proportion 

 of normally matured pollen and ovules or that the parent growing 

 on the hybrid would show some lessening of reproductive vigor. No 

 such result was found to occur in either case. The hybrid comes into 

 flower when growing on the parent with as small a proportion of 

 pollen of normal appearance as it does when growing on its own roots, 

 while the parent growing on the hybrid produces an abundance of 

 normal pollen and the usual proportion of viable seeds is formed. 

 However, the hybrid flowers and young f rviits were retained for a much 

 longer period than was the case when the same plant grew on its own 

 roots under field conditions. 



IV. Summary of Results 



1. The Fi hybrids between Nicotiana Tahacum varieties and N. 

 sylvestris produce a very small quantity of pollen of normal appear- 

 ance, while practically the entire contents of the anther cells consists 

 of shriveled, functionless grains. 



2. Fj pollen of normal appearance could not be caused to germinate 

 in its own stigmatic secretion or in that of its parents or in any one 

 of a great variety of artificial germinating fluids. 



3. The pollen of the parents germinates readily in the stigmatic 

 secretion of the F^ flowers. The germination and growth reaction of 

 the pollen of N. Tab. var. macrophylla and N. sylvestris in tap water 

 and in 10 per cent solutions of dextrose, levulose and maltose were 



