400 De Vries. — On Bias trep sis in its 
which were made with my twisted breed of Dipsacus sylvestris. 
As I have already remarked, this breed, with proper treat- 
ment, gives a progeny 30-40 per cent, of which have well- 
marked twisting of the main stem, besides other individuals 
which show less marked twisting of the branches, and others 
again which have their leaves in whorls of three. If, however, 
the treatment is unsatisfactory as regards any one of the 
essential conditions mentioned above, this high percentage 
of monstrous forms is not obtained ; there may, in fact, be 
none at all. 
The scientific results of my experiments may be summed 
up by saying that Biastrepsis, especially in the case of Dipsacus 
sylvestris, is in a high degree dependent upon the conditions 
which obtain during and subsequently to the germination of 
the seed. Conditions which favour the vigorous development 
of the plants promote also the size and the number of the 
portions of the main stem and of the branches which show the 
twisting. 
The practical result of my experiments is, I consider, the 
recognition of the fact that the cultivation of monstrosities 
should either be carried out strictly according to rule, or not 
be attempted at all. Without the requisite care the attempt 
only results in doubt being cast upon the goodness of the 
seed and the hereditary properties of the breed. 
I treat of the subject under the following heads : — 
A. The normal cultivation of the breed. 
B. The influence of space. 
C. The influence of the soil. 
D. Summer-sowings. 
E. Autumn-sowings in the open. 
F. Autumn-sowings under glass. 
G. Cultures in other Botanic Gardens. 
A. The Normal Cultivation of the Breed. 
In my monograph on the subject (1891), to which allusion 
has already been made, I have given an account of the 
morphology of biastrepsis in Dipsacus sylvestris , and of the 
