506 MISS ISABEL MCCLATCHIE ON 
produced in large numbers down the side of the second internode and yet 
were absent or of only occasional occurrence in the first internode. In other 
plants the reverse was the case. One plant produced roots from every node 
up to the fifth, and yet all the internodes were bare except the third, 
which was rather bent and had a great number of roots growing from its 
convex side. 
The roots resembled subterranean lateral roots in appearance ; they were 
unpigmented, slender, and produced lateral roots freely. In structure they 
were polyarch and like aerial adventitious roots, except for the absence of 
tannin cells, 
One very interesting feature was that the roots, though given off from the 
side away from the w: all, did not make a right angle with it, but were in all 
twelve plants given off towards the north-north-east. This was possibly due 
to the weight of the soil and the position from which it was thrown during 
the banking-up process, causing the plants to incline very slightly in that 
direction. 
(&)—Six plants were taken about the middle of August and were pegged 
down so that three or four internodes lay along the soil and the upper 
surface remained uncovered. They were left in this position until the end 
of September. 
Results. 
Where the nodes were in contact with the soil they became swollen, and 
in five out of the six plants roots developed from the underside. These 
roots penetrated the soil and produced lateral roots. In four plants there 
were signs of roots at the bent node, just above the soil-level, where the 
plant regained its upright position. No roots were produced along the inter- 
nodes (text-fig. 14). 
(c)—Six plants were pegged down horizontally, as in experiment (4), at 
the beginning of July, and the horizontal portion was buried in soil to a 
depth of several inches. They were left undisturbed until the end of 
September. 
Results. 
There was a profuse development of roots, but from the underside only. 
In some cases roots were confined to the nodes, but in others they appeared 
throughout the hypocotyl, the first, the second, and the third internodes 
(text- fig. 15). 
(¢)—A number of plants which had developed roots from the underside 
at the first node were gradually bent until the plant leaned in the opposite 
direction. 
Result. 
Although no adventitious roots were developed on the underside, those 
which had arisen before the experiment ceased to grow (text-fig. 16 a). 
