96 
UPWARD GROWTH OF THE HEAD, AND [Part II. 
straight down and came through the wire into 
the air. The gemmules or heads of five also 
came down, and, on cutting away the wire, all 
five grew horizontally to the light at the window, 
and then grew diagonally upward and to the 
light. The gemmules or heads of the rest grew 
upward through about 8J inches of earth, were 
drawn by the light with unerring precision 
through the hole of the flower-pot, and one 
through an accidental hole of about a quarter of 
an inch in diameter. They then turned short 
towards the light at the window. 
In November, on taking the flower-pot off, 
I found that branch-roots had passed over the 
top of the ball of earth, 8^- in, high. I replaced 
the flower-pot with a chimney-pot 2 ft. 7 in. 
high, and filled it with earth, leaving only the 
five plants whose heads had come out below. 
I placed the experiment out of doors supported 
from below. The five plants grew in 1845. In 
the spring of 1846 I cut off all except one plant, 
and placed a second chimney-pot above the first 
one. 
In March, 1850, I placed a third chimney-pot 
on the second, making the column of earth 7 ft. 
5 in. in height. The roots had already reached 
the top of the second chimney-pot, about 5 ft. in 
