Lutz, Triploid Mutants in Oenotheru. 405 



lateral or sub-lateral branch may differ from the first flowers of 

 the stem or a strong basal branch. 



The indiscriminate pollination of flowers without regard to the 

 position of the pistil- or pollen-flower on the plant, - - such as was 

 practiced in the self- and cross-fertilization of the triploid mutants, - 

 might very well result in the production of offspring with just 

 such a variety of somatic chromosome numbers as that which is 

 indicated by the vegetative characters of the offspring of these 

 mutants. 



The discrepancy in the observations of Gates and Geerts 

 may have resulted from material having been fixed from early buds 

 in one case, and from late in the other 18 ). ("Early'' and "late" of 

 course refers to the first and last buds on the stem and branches, 

 and not to the time of the season at which they were produced). 



These questions can be settled only by extensive studies of the 

 reduction division in the germ cells of flowers from various regions 

 of 21-chromosome plants in connection with the exact determination 

 of somatic chromosome number in a large number of offspring of 

 these plants, self-pollinated. However, it is exceedingly difficult 

 to obtain seed from the self-pollination of 21-chromosome plants, 

 or the pollination of such an individual by another of its kind, and 

 it may not be possible to obtain large cultures of such offspring 



18) I have since had the privilege of discussing this question with Dr. Geerts, 

 and he tells me that his fixations from the 21-chromosome hybrids were made in 

 September and October, and that they were taken from seed-plants, therefore from 

 individuals which had produced their first flowers much earlier in the season. 

 This indicates that Geerts' type of reduction appears in the later flowers, and 

 Gates' probably in the earlier ones. Pollination of the Cold Spring Harbor 21- 

 chromosome mutants was continued from the appearance of the first flowers (dates 

 unrecorded, probably about the last of July) until the end of the second or third 

 week in August. In our climate, seeds will not ordinarily ripen from later fertiliza- 

 tions. It therefore appears that the Geerts type of reduction may occur compara- 

 tively early in the season, though perhaps in flowers of a stem or branch which 

 has been blooming for sometime. It was my custom, in pollinating these 21-chro- 

 mosome mutants, to cover the stigma of one flower with pollen from another. 

 Occasionally, however, the pollen of a flower was applied to its own stigma. If 

 pollen chanced to be taken from au "early" flower, and to be applied to the stigma 

 of a "late" flower, then a 7 -chromosome egg might be fertilized by a 10- or 11- 

 chromosome sperm, and produce a 17- or an 18-chromosome offspring. Likewise, the 

 fertilization of a "late" by a "late" might produce only 14-chromosome offspring. 

 All the Lamarckiana offspring of 5483 may have been derived from the seeds of 

 one capsule, and from the fertilization of a "late" flower with its own pollen, or 

 with that of another "late" flower. Even though it may not be possible to ripen 

 seeds from late fertilizations, it will be of great interest to determine whether they 

 are more commonly produced from the pollination of "late" than "early" flo\vers of 

 21-chromosome plants. 



Geerts' discovery of the degeneration of a portion of the chromosomes during 

 the maturation process of certain triploid hybrids is a most important and significant 

 one. A large field of inquiry is opened up in this connection. 



