HARRIS: TETR.\COTYLEDONOUS R.\CE OF PHASEOLUS VULGARIS 233 



Concerning the structure of the seedling of the 1907 plant, or 

 of that of its six Fi (1908) and six Fo (1909) descendants, nothing 

 is actually known. The seeds were planted directly into the 

 field. Seedling characters were not studied until 1912. 



It seems highly probable that these seven or nine seeds were 

 produced by a single plant of the 191 1 culture. It does not seem 

 profitable to consider the various points at which the fundamental 

 v^ariation may have occurred. That the race originated some time 

 between 1907 and 191 1 inclusive is clear from the fact that a 

 relatively normal strain of plants from the same line of plants is 

 at present under cultivation at the Station for Experimental 

 Evolution. 



In the 1 912 germinations, line 139 gave a total of 4,375 seed- 

 lings of which 4,239 were normal and 136 (including the 9 plants 

 the offspring of which are here considered) were abnormal. Of 

 the 4,375 seedlings studied in 1912 a series of 71 abnormal (exclud- 

 ing the 9 here especially under consideration) and 79 normal 

 individuals were grown to maturity and produced a total of 1,621 

 offspring seedlings in 1913. All these progenies were relatively 

 normal, containing a total of only 45 abnormal plants or 2.78 

 per cent, as compared with sensibly 100 per cent in the so-called 

 tetracotyledonous race. 

 \ Thus the origin of this race of plants has all the characteristics 

 )of a de Vries ian rn u tation. 



III. Quantitative characterization of the race 

 The precision of a quantitative description of the race is limited 

 by technical difficulties in counting in the case of the discrete 

 variates and by the possibility of personal equation in the case 

 of the characters which must be described in arbitrary categories. 

 Because of the pressure of other experiments and the difficulties 

 of classifying plants with such a high degree of abnormality the 

 description of the 1915 germinations of these' lines was necessarily 

 incomplete. To have expressed adequately the characteristics of 

 several thousands of plants of such complexity of structure would 

 have required many weeks' work, and in the present state of our 

 knowledge of the morphology of the bean seedling would have 

 been of doubtful value. 



I have therefore limited myself to a detailed study of a portion 

 of the material only. 



