The tips of 11 radicles were protected by moistened gold-beaters’ skin, which adheres closely, for a length varying from 1 ½ to 2 ½ mm.  After 22 h. to 24 h., 6 of these radicles were clearly bent towards or had come into contact with the sieve; 2 were slightly curved in this direction, and 3 not at all.  All had grown well.  Of 14 control specimens observed at the same time, all excepting one had closely approached the sieve.  It appears from these cases that a cap of goldbeaters’ skin checks, though only to a slight degree, the bending of the radicles to an adjoining damp surface.  Whether an extremely thin sheet of this substance when moistened allows moisture from the air to pass through it, we do not know.  One case indicated that the caps were sometimes more efficient than appears from the above results; for a radicle, which after 23 h. had only slightly approached the sieve, had its cap (1 ½ mm. in length) removed, and during the next 15 ½ h. it curved itself abruptly towards the source of moisture, the chief seat of curvature being at a distance of 2 to 3 mm. from the apex.

Vicia faba.—­The tips of 13 radicles were coated with the grease for a length of 2 mm.; and it should be remembered that with these radicles the seat of chief curvature is about [page 183] 4 or 5 mm. from the apex.  Four of them were examined after 22h., three after 26 h., and six after 36 h., and none had been attracted towards the damp lower surface of the sieve.  In another trial 7 radicles were similarly treated, and 5 of them still pointed perpendicularly downwards after 11 h., whilst 2 were a little curved towards the sieve; by an accident they were not subsequently observed.  In both these trials the radicles grew well; 7 of them, which were at first from 4 to 11 mm. in length, were after 11 h. between 7 and 16 mm.; 3 which were at first from 6 to 8 mm. after 26 h. were 11.5 to 18 mm. in length; and lastly, 4 radicles which were at first 5 to 8 mm. after 46 h. were 18 to 23 mm. in length.  The control or ungreased radicles were not invariably attracted towards the bottom of the sieve.  But on one occasion 12 out of 13, which were observed for periods between 22 h. and 36 h., were thus attracted.  On two other occasions taken together, 38 out of 40 were similarly attracted.  On another occasion only 7 out of 14 behaved in this manner, but after two more days the proportion of the curved increased to 17 out of 23.  On a last occasion only 11 out of 20 were thus attracted.  If we add up these numbers, we find that 78 out of 96 of the control specimens curved themselves towards the bottom of the sieve.  Of the specimens with greased tips, 2 alone out of the 20 (but 7 of these were not observed for a sufficiently long time) thus curved themselves.  We can, therefore, hardly doubt that the tip for a length of 2 mm. is the part which is sensitive to a moist atmosphere, and causes the upper part to bend towards its source.