416 é BOTANICAL GAZETTE [DECEMBER 
the model was withdrawn, and into the canal was inserted the radicle 
of another seedling, so that the tip of the latter almost touched the 
horizontal hypocotyl (fig. 2). The inserted seedling was then almost 
entirely covered with gypsum, and when hard the preparation was 
placed in sawdust. 
The block of cork 
under the hypocotyl 
was of course with- 
drawn, and thus the 
side of the hypocotyl 
opposite the entering 
root was not covered 
with gypsum, and if 
‘Fic. 2.—Arrangement of seedlings for penetration the entering root pene- 
awtn — the line-shaded portion representing trated the hypocotyl it 
: could find exit unob- 
Many such tests i i bee bipee 
= were made in which the main and lateral roots of 
Vicia F aba, Lupinus albus, and Phaseolus multiflorus were used in 
the various combinations, but the results were always negative. The 
preparations were allowed to stand undisturbed for various periods 
up to paliaaie days, but the result was always negative. The radicle 
of the inserted seedling turned its tip as far as the canal would allow 
— than the diameter of the radicle a mm. above the tip) and 
. — imbedded in the callus subsequently formed. A stroné 
print of the blunted radicle was always found on the hypocotyl, 
but Heicrpscopic examination showed the cuticle to be uncorroded 
= aed intact. It is evident that if the radicle or the lateral 
oe natag © enzyme, such enzyme has no digestive action Ph 
— = pparently the mechanical push of the advancing roo 
ee ong ier to break through the cuticle. However, 9° 
that in er nei ne . drawn here. It is significant, aap 
with a small Pee oe eee slightly wounded, as by piercing 
ame es ass point, either a main root or a lateral root . 
sale ce - s entirely through the hypocotyl without formation © 
: er visible evidence of obstructed passage. Of come 
is much easier to make an exit than an entrance through intact 
