86 



White. 



than mere description, of the extent and i.ature of tliis variability, the 

 former as regards the variability of the race and the latter as regards 

 the range per individual plant. By inspection of Table 3, the average 

 number of parts per whorl of a flower is seen to show a progressive 

 increase in the expression of the fasciation factor. This may be expressed 

 in tabular form by subtracting the normal number of parts per whorl 

 (5 for the first three and 2 for the gynoecium) from the average for 

 the number of parts per abnormal flower. In order to compare the 

 first three whorls with the gynoecium, their differences should be divided 

 by 2" 5* as there are 2' 5 times as many petals, sepals and stamens 

 to a flower as ovary -locules. The results for families 301 — 1 and 

 303—1 are thus: 



Table C. 



The calyx is thus seen to be the least, and the gynoecium the most 

 affected. The latter is almost three times as abnormal as the former 

 in 301 — 1, and more than three times in the case of 303—1. This 

 progression in the manifestation of abnormal condition is in accordance 

 with the observations on other parts of the plant. The seedlings appear 

 to be normal; the first few leaves are not deranged as to phyllotaxy, 

 and the whole stem remains normal, even in the most fasciated specimens 

 for a foot above ground. The linear expansion increases in extent and 

 the leaves in number as the plant approaches maturity. At maturity, 

 the apex of the stem shows the greatest linear expansion, and sometimes 

 becomes so abnormal that the whole inflorescence is partially inhibited 

 in its development. The greatest alteration in phyllotaxy and the 

 largest increase in number of leaves is characteristic of this terminal 

 portion of the main axis. Owing to the fluctuation in expression of 

 the factor, the stem may not always show the linear expansion through- 

 out its whole length, but in all cases it shows in the inflorescence. 



