568 DOUBLED CHROMOSOMES OF OENOTHERA 



almost exactly alike, with a very typical form of the radical leaves 

 and the same stem and spike reaching the same height at the time 

 of flowering, in the middle of August. One specimen with 16 chromo- 

 somes slightly deviated in its characters; moreover, it did not 

 produce any good pollen, whereas all the others were rich in this 

 respect. It was also weaker and of lower stature. The main form 

 reached about one-half the height of the normal Lamar ckiana, and 

 had the same stature as this, with a more condensed spike and a 

 less branched stem. Of all the mutants, the pulla differed the least 

 from 0. Lamarckiana. The flower buds reached the same size, al- 

 though theu were not as completely filled. The fruits were as broad 

 and stout, and as lich in seeds as those of the parent species. In 

 the spring the radical leaves of this form had been quite a puzzle, 

 since the type was unknown at that time. These leaves resembled 

 those of 0. gigas in being broad, with a rounded tip and a sharp 

 contraction of the base, which separated the blade from the long 

 and winged petiole. They were given a special name, but proved 

 later to correspond to the type now called pulla. 



On account of the small and hardly striking differences from the 

 parent species, we assume that the factors of pulla are to be sought 

 for in the same chromosome which decides the main characters of 

 0. Lamarckiana itself, bearing the determiners for laeta in some of 

 the gametes and those for velutina in others. We have called this 

 chromosome the central one, but the outward features of pulla are 

 accompanied by a doubling of one of the nuclear rods. Five speci- 

 mens had 15 chromosomes, whereas three had 16, and two 17 such 

 bodies. The mutations previously ascribed to the central chromo- 

 some, such as dwarfishness and brittleness, are not accompanied 

 by an increase of the number of nuclear rods. If our conception is 

 true, the pulla completes the series of the seven cases of chromatic 

 doubling which must cooperate in the formation of semigigas. From 

 this point of view it is interesting to note that the pulla occurred 

 in about the same number of individuals (12.5 per cent) as most 

 of the other main types of mutations. 



Lamar ckiana-like plants. — In the reduction division of semigigas, 

 the extreme case is that the seven doubled chromosomes go to one 

 gamete and the seven single ones to the other. After fertilization 

 with 0. (biennis x Lamarckiana) velutina the first group would 

 produce germs with 21, but the other group would give embryos 

 with 14 nuclear rods. Those with 21 would constitute semigigas 

 plants, but their chance of survival is small, and in our experiment 



