1900] PERENNIAL HERBS 183 
specimens, grown in a special culture case, from their base with 
lines of India ink 5™™ apart. These 5™™ spaces shortened to 
4"" each within six weeks, which equals a contraction of 20 per 
cent. At the same time the uppermost line, that next the root 
base, moved about 6™™ downwards, placing the base of the 
shoot so much deeper. 
Also in Aguilegia Canadensis 1 observed a considerable short- 
ening of the root, which results likewise in drawing down the 
growing point into the earth. In seedlings of Aquilegia vulgaris 
L. I noticed that the growing point of the stem, which immedi- 
ately after germination stood 8™™ above the ground ( fig. 17) 
was found at the end of the first summer, in consequence of the 
contraction of root and hypocotyl, about 6™" below the certace 
of the earth ( fig. 72). The upper lateral roots become bent 
down during this process in a very characteristic manner, (7, 
fg. 12). The transverse wrinkling of the surface in the older 
oe Aquilegia and Allionia is due to the same cause, and 
“orresponds to the folding of the roots of Arisaema, Hypoxis, 
and others, 
There can scarcely be any doubt that in Lithospermum angusti- 
sah Nothocaleis cuspidata, Peucedanum foeniculaceum, Astragalus 
pam te Kuhnistera candida, and K. purpurea the root is also 
Pe a although I have not had the opportunity to measure 
the und A This seems to be indicated by the following points: 
beac aunty. cea of the innermost vascular bundles, the 
the up = of the outer bark, the disturbed position of 
bales we ateral roots, and finally the position of the esate 
prolon ee the surface of the earth, in spite of the continua 
in Kulnia be re: perennial stem-portion towards above. Also 
to occur a A oe and Grindelia squarrosa contrachee seems 
it possibl ough in . less degree, while in Lacinaria punctata 
eee sf cad - exist at all. In pieces of roots tree 
observed a = tum and Ki uhnia eupatorioides lying in water, 
In addition seh i a considerable shortening. a. 
rosa (L.) Hill might say something concerning Lacinaria squar- 
; Which, although resembling Z. punctata very 
