Studies of Teratological Phenomena. 101 



its floral parts while Table 2 gives this same information for the number of 

 leaves. Table 3 is a summation table showing the frequency distribution 

 in number of flower parts per flower for the race. Adding together 

 the total number of flower parts of all the flowers from all the plants 

 of each race examined, and dividing this sum by the total number of 

 flowers examined a constant is obtained by which to more accurately 

 compare the difference between the two races (301- — 1 and 303 — 1). 

 Thirty three 301—1 plants with a total of 825 flowers, gave a constant 

 of 6 -509+, while 303—1 with a total of 850 flowers from 34 plants, 

 gave 7 "152 4-, the difference in abnormalness being 0*643+ in favor 

 of the 303 — 1 race. This difference in abnormalness is apparent in the 

 averages calculated for all the flower parts, and is also true of the 

 stem-flattening and the number of leaves. The average number of 

 leaves per plant for 32 of these same 301 — 1 plants is 34-18+, while 

 that for 36 (303 — 1) plants is 45*52+ leaves per plant. Table 4 gives 

 the range of variation in flower parts per flower for each plant of the 

 two races. The range in individual plant variation is about the same 

 for the two races with a slight advantage in favor of 303 — 1. 

 The modes for the variation in number of parts per flower are also 

 the same, with the exception of the ovary -locules, 303—1 having 

 almost one more locule per average flower then 301 — 1. Selection so 

 far seemed to be producing results, so four plants were selected from 

 the progeny of 301—1 and 303—1 and selfed. These were 301—1—2, 

 303—1—14, 301—1—29 and 303—1—12. The two former were 

 approximately the least abnormal progeny of their respective families 

 grown in 1910, while the two latter were approximately the most 

 abnormal. The extent of their abnormalities in floral structures may 

 be noted in Table 4, and changes in leaf number in Table 2. 



Selfed seed of these four selections was grown in 1911 and the 

 plants matured under about the same environment as surrounded the 

 301 — 1 and 303 — 1 cultures of 1910. Time did not allow me to make 

 an elaborate examination of the 239 plants thus produced, nor of the 

 303 — 1 and 301 — 1 plants that grew beside them, serving as checks. 

 However, by going through them at maturity, I was able to classify 

 them roughly by the extent of their stem-fasciation into slightly abnormal 

 and abnormal classes (see Table E). 



Slightly abnormal simply means that stem fasciation only appeared 

 in the region of the inflorescence. This was the stem condition of the 

 parents 301—1—2, 301—1—29 and 303—1—14. Parent 303—1—12 

 had an exceedingly abnormal, bent flattened stem. 



