238 



MEMOIRS OF THE NEW YORK BOTANICAL GARDEN 



These differences are very slight indeed. Such results indicate 

 very clearly that the increase in the number of cotyledons over 

 the normal which characterizes this race is not fundamentally due 

 to a broadening of the axis. 



The accompanying table shows that about fifty per cent of the 

 plants are recorded as having produced shoots from the axils of 

 the cotyledons. 



It is interesting to note that a higher proportion of the plants 

 are recorded as producing axillary shoots in the less mature series 

 than in the more mature series. The difference is probably due 

 merely to the errors of random sampling, but the fact that the 

 axillary shoots are not more abundant in the more mature series 

 indicates that they are not structures of late development. 



Primordial leaves 



The following conditions present real obstacles to the deter- 

 mination of the number of primordial leaves with a high degree of 

 precision. 



{a) When the seedlings are very young all the primordial leaves 

 are not fully expanded, and there is consequently some technical 

 difficulty in determining the true number, {b) When the plants 

 have grown to a considerable size shoots are apt to have developed 

 from the axils of the cotyledons which by this time may have 

 fallen. It is exceedingly difficult, as pointed out above, to dis- 

 tinguish between these axillary shoots and the longitudinal or lat- 

 eral division of the stem. This is especially true if the cotyledons 

 have fallen. Since the earlier leaves on the axillary shoots are 

 apt to be simple, they would, if included, increase the numl)er of 

 primordial leaves, (c) It seems quite probable that some of 

 the axillary shoots have started development during the matura- 

 tion of the seed. Since it is impossible to distinguish between 

 these axillary shoots and branches of the stem originating other- 

 wise, i)rinionlial leaves found on them sJiould be counted in. 



