1912] Goodspeed: Nicotiana Hybrids 143 



degree of fliictnation was practically as great at the beginning, 

 at the middle and at the end of the flowering season. 



Plate 33 exhibits the degree of fluctuation in corolla diameter 

 of varieties I and III and the fluctuation of all the plants repre- 

 senting the cross 111$ X IJ* of each of sixteen different days 

 between September 15 and November 6. The heights of the 

 various columns correspond to the number of flowers for each 

 diameter measured on the various dates. It will be remembered 

 that the parental varieties did not give as vigorous plants during 

 the season of 1911 as in 1910 and did not flower anywhere nearly 

 so profusely as during the previous season. This is the cause 

 of the shortness of the columns throughout for the parental 

 varieties I and III as contrasted with the heights of the columns 

 showing tlie fluctuation of the hybrid plants representing the 

 cross between these two parents. A comparison between plate 

 31 and plate 33 will show that, in general, the production of 

 flowers for the entire group of plants of cross III 5X I c^ (plate 

 33) corresponds closely to the production of flowers on a single 

 plant of this hybrid group (plate 31). 



It will be noted that upon only two occasions was the diam- 

 eter 29 mm. found for variety I, while the limits of fluctuation 

 of the hybrid, 17 and 27 mm., appeared on six occasions for 

 the diameter 17 mm. and on seven different days for the diameter 

 27 mm. Diameter 15 mm. was noted on the flowers of variety III 

 on onl}^ two occasions, both previous to September 15, and for 

 this reason diameter 15 mm. is not included in the columns of 

 variety III. It is also plain that the number of flowers in the 

 hybrid group exhibiting these extremes of fluctuation was large 

 as compared with the number of flowers giving the limit of 

 fluctuation 29 mm. in the parent variety I. Also, from plate 33 

 and the above table (p. 142), it can be seen that there was no 

 appreciable difference in the average size of corolla diameters as 

 the growing season advanced ; a difference which might have 

 been expected to exhibit itself because of an increase of strictly 

 vegetative growth at certain times and a consequent inhibition of 

 normal floral development. 



