( 401 ) 
began to develop vigorously, so that in the autumn scarcely any 
difference could be observed. The reason of this late, very rapid 
development appeared when the plants were dug out. It turned out, 
namely, that some of the strongest roots had reached the underlying 
earth through the layer of sand. As long as the plants only obtained 
their food from the sand, they remained tiny and backward, but 
when the roots had penetrated into the fertile earth they still deve- 
loped vigorously and with great rapidity. Also with Zberis amara 
the roots appeared to have reached the earth underneath but in a 
much less degree. It was difficult here to trace the fine terminals 
of the principal roots as far as the underlying earth, whereas the 
roots of Malea vulgaris, where they passed from the sand into the 
earth below, were strong and penetrated at least a few decimetres. 
Of Ranunculus arvensis only few roots had reached the underground 
with their tips, the same being the case with Scandix Pecten- Veneris 
and Anethum graveolens; the roots of Cardamine hirsuta were restricted 
to the sand, as far as I could see. 
Although with most of the species studied the nutrient material 
was not entirely derived from the sandy soil, yet all these plants 
were in less favourable nutritive conditions than the plants on the 
manured soil. So the experiments will show us the consequences 
of the difference in nutrition. 
For the investigation I chose some characters that are easily 
expressed quantitatively and numerically and took care that the 
determination was made at the same time for both cultures and that 
the same parts of both were always taken. 
In this way I determined in the first place the length of the leaf 
of Lheris amara. In July the length of the five oldest leaves, which 
were already adult then, was measured. Besides in the autumn, after 
the plants had been dug out, the length of the plant was determined 
from the base to the top of the inflorescence of the principal stem ; 
at the same time were counted the number of branches of the second 
order, the number of branches of the third order and the number 
of fruits on the inflorescence of the principal stem. 
Of Malvea vulgaris the number of akenes of the schizocarp, the 
length of the leaf-blade and the length of the leaf-stalk were deter- 
mined. These countings and measurements were made in the beginning 
of July, when a very distinet difference in the development between 
the two cultures was visible, hence probably before the roots of the 
plants on the meagre soil had penetrated the layer of sand, and in 
any case before a better nutrition had any perceptible effect. 
In the case of Anethum graveolens and Scandia Pecten- Veneris the 
28 
Proceedings Royal Acad. Amsterdam. Vol. VII. 
