304 The Pedigree Families. 



In the autumn of 1888 I raised in my garden, from 

 seed gathered from apparently normal examples in the 

 field, besides numerous Lamarckianas seven examples of 

 Jata', of which four developed stems, one remained in 

 the rosette stage, whilst the two others did not come up 

 till late in the summer. In the same summer I found a 

 beautiful flowering specimen of Jata in the field and one 

 or two other young plants which were immediately rec- 

 ognizable as belonging to this new species. 



In 1894 two flowering plants and one rosette of lata 

 were again found in the field. 



0. cUiptica. I found one rosette in 1886 and another 

 was found in 1894. 



O. nanella was obtained in 1889 from seed collected 

 in the field in the previous autumn : there were three 

 rosettes which however did not survive the winter. One 

 of these was a lata-nanella; that is to say it combined 

 the characters of both forms, a phenomenon which has 

 occurred again in my experiments. Another dwarf was 

 found in the field at Hilversum in 1894. 



0. nihrinervis. One rosette was raised from the 

 seed collected in 1888, which has been so often referred 

 to already. 



O. spathnlata. This form was collected as a rosette 

 in the field at Hilversum in 1886 and 1894. 



Five of the new species therefore, either appeared 

 in the field at Hilversum or from seed collected there. 

 They appeared rarely, but repeatedly, and moreover their 

 appearance extended over the course of several years. 

 It was, however, not possible for the ones that arose 

 later to have descended from the earlier ones, since the 

 latter had not flowered. 



