204 Atkinson. 



garden for the winter and others were wintered in a cold frame. At 

 the time of publication of the note just cited, the rosettes had 

 passed the winter safely. They formed stems and flowers during the 

 summer of 1913, and one of the plants formed an autumnal rosette 

 again at the base of the stem. This was allowed to stand in the gar- 

 den during the winter of 1913 — 14, but as the winter was severe the 

 rosette died. 



Another interesting feature of Oe. nutans is its amenability to 

 vegetative propagation. The ability to produce new shoots from the 

 mature and bare stems indicates its possible fitness for propagation 

 by cuttings. At my request several cuttings were made by Mr. R. Shore, 

 the head gardener of the Department of Botany, from the mature plants 

 wintered in the green house during 1912 — 13. When the new young 

 shoots from the mature stems were sufficiently hardened they were cut 

 away with a small portion of the old stem, the tip removed and the 

 cutting then placed in the earth. They rooted well and each formed a 

 cluster of leaves. They were transplanted to the garden in the spring 

 of 1913, where they produced stems and flowers from July until checked 

 by the November cold. One of these formed a rosette at the base of 

 the stem, but it was killed during the severe 1913 — 14 winter. In its 

 fitness for vegetative propagation by the ordinary method of cuttings, 

 Oe. nutans appears to form an exception among the Oenotheras so far 

 as at present known. De Vries (25, 1913) propagates them by using 

 the lateral shoot buds of rosettes, but says that for other methods of 

 vegetative propagation the Oenotheras are not adapted. (See also Lode- 

 wyks, 1908). 



The Fl hybrids from Oe. pyciiocarpa and Oe. nutans. 



In the cultures of 1913, the Fi hybrids of reciprocal crosses be- 

 tween Oe. pycnocarpa and Oe. nutans presented three types as stated 

 above. When pycnocarpa was the mother (pycnocarpa X yndans) 

 there were two types, which according to de Vries' terminology are 

 twin hybrids. These twins were exactly the same as those obtained 

 in the cultures of 1911 — 12 from the same cross, so that the results 

 of this earlier culture are confirmed. The uncertainty of the results 

 obtained in the 1911 — 12 cultures related to the rosette stage which 

 was so seriously parasitized by Peronospora that an analysis was im- 

 possible, although the adults of 1912 were readily separated into two 



