Tricotyly and the Arrangement of Leaves. 373 
away as soon as they reverted to the decussate arrange- 
ment of their leaves, but seven of these trees are still 
alive, two with ternary stems (p. 370), two with a de- 
cussate arrangement of the leaves, and one with a much 
flattened main stem. This last one began with three 
cotyledons; it then became decussate, and in its second 
year (1889) became ternary again. In the autumn of 
the following year it began to flatten out, formed three 
five-leaved whorls and began to split when laying down 
the winter bud. I then broke off all the terminal buds 
except one, which in the spring of 1891, during sub- 
sequent growth, split into three flat twigs of which 
again two were removed. In the following summer the 
fasciation recurred, and also, after splitting in the win- 
ter, in the next year (1892), and again in 1893. Every 
time the forked branches were reduced to one. The 
divisions became much rarer, later on, and the older sec- 
tions of the stem which were at first flat gradually be- 
came cylindrical, as usually happens in fasciation when 
it affects trees. 1 
On tricotylous specimens I also observed flattened 
stems in Antirrhinum majus, Artemisia Absynthinm, Sca- 
biosa atropurpiirea, Dianthus plumarms, Collinsia hetero- 
phylla, C. grandi flora, C. violacea and Tetragonia e.v- 
pansa (Fig. 78) ; and amongst tetracotyls in Scrophn- 
larla nodosa and Collinisia violacea and other species. 2 
In many cultures I have observed that fasciations are 
more common amongst tricotylous plants than amongst 
dicotyloiis ones, but I shall only deal in detail with an 
experiment on Asperula azurea. In the spring of 1892 
I selected the hemi-tricotylous, tricotylous and tetra- 
*Kruidk. Jaarb. Gent, 1894, Plate XI, (Abies excelsa). 
*Ber. d. d. bot. Ges., Vol. XII, p. 38. 
