Two Lata-Families. 281 



fourth row. The plants are transplanted in rows in 

 order to make the best advantage of the availal)le space. 



It should be observed that the pedigree of this culture 

 contains seven successive generations of lata, the first of 

 which arose from a Lamarckiana which I jj^rcw in my 

 garden. This group arose from a single [)lant which 

 was fertilized with the pollen of a plant of Lamarck's 

 Evening Primrose of similar parentage. The origin of 

 this race vv'as therefore as pure as the unisexual ity of 

 lata permitted. 



Seedlings of O. alhida are very easy to recognize. 

 Nothing short of seeing the plants themselves can really 

 give an accurate idea of these phenomena : but the pic- 

 tures may serve to convey the impression of the ])rocess 

 of mutation. The plate shows besides O. alhida, five 

 plants of O. lata and six of O. Lamarckiana. The for- 

 mer can be recognized by their bright green round leaves, 

 the latter by their darker green leaves which are more 

 or less pointed. The three O. alhida are much smaller, 

 of a paler color, and with narrower leaves. 



The plate shows another case besides these. I mean 

 the plant at the right-hand end of the second row. It 

 has narrow leaves and does not look quite like any of the 

 tvpes hitherto recognized in these experiments. Does 

 it represent a new form? I devoted every possible care 

 to its further cultivation: but it sickened and died before 

 its mature characters had developed. The three alhidas 

 also got no further than the rosette stage. 



Another group from the same experiment is faith- 

 fully represented by photography in Fig. 48. The group 

 was taken in the way described above on the 25th of 

 May; it arose from another parent plant, of the same 

 experiment of 1899, which had likewise been fertilized 



