116 Gates, Tetraploid Mutants and Chromosome Mechanisms. 



In another paper (1912 b, p. 533) Stomps describes (c) a single 

 mutant from 0. biennis having 21 chromosomes, which he calls 

 0. biennis semigigas 13 ). 



Miss Lutz (1912) has also found 5 triploid mutants from 

 Lamarckiana and 3 from lata X Lqmarckiana, having 21 chromo- 

 somes in their root tips. 



Both Stomps and Miss Lutz apparently assume that this 

 proves tetraploid mutations to have originated through the union 

 of two diploid (2 X] germ cells and by this method only. Their 

 conclusion in no way follows, however, for if all the pollen grains 

 are haploid (X) as in other plants, or if the diploid pollen grains 

 fail to function, the origin of both triploid and tetraploid mutants 

 can still be easily explained; namely, that the former originate 

 from the fertilization of a diploid egg by a haploid male call, the 

 latter from the apogamous development of a tetraploid megaspore 

 mother cell, such as was observed by Geerts. That a triploid 

 mutant can also be produced by the union of a diploid male cell 

 with a haploid egg, will only be proven when it is shown that 

 diploid pollen grains occur and are functional. At present there is no 

 certain evidence for this, while there is direct evidence for the occur- 

 rence of tetraploid megaspores in the above-cited observation of 

 de Vries' pupil, Geerts. This also, despite the fact that studies 

 of pollen development in the Oenotheras have been much more 

 numerous than studies of megaspore development, and notwithstand- 

 ing the further fact that the number of microspore tetrads observed 

 has been thousands of times greater than the number of megaspore 

 tetrads. 



Races of 0. gigas. 



In the paper already cited (Gates, 1909 a, p. 536) I mentioned 

 that the gigas of de Vries' cultures is all descended from one 

 individual, but that two other mutants more or less resembling 

 gigas were observed. I suggested that the latter perhaps represented 

 a different form, and it now appears that they may have been 

 triploid in character. Two independent cases of the appearance 

 of gigas forms have since occurred, and these I shall now describe. 

 The first I received in 1909 from the Botanical Garden at Palermo, 

 Italy, under the garden name 0. cognata. My first culture from 

 these seeds, grown at the Missouri Botanical Garden, is described 

 in a paper now in press in the Transactions of the Linnean 



13) During the past summer I also (Gates, 1912a) discovered mutants iu a 

 cultivated race of 0. biennis, including 0. biennis lata, 0. biennis laevifolia and 

 0. biennis rubrinervis. These corresponded to, though not agreeing with, the 

 Lamarckiana forms, and were called parallel mutations. 



