904 
Journal of Agricultural Research 
Vol. XXIII, No. ix 
through the winter. Although resting buds had been organized at the 
bases of the leaf petioles, these did not develop, and the general appear¬ 
ance above ground was one of comparative inactivity. Eventually, 
however, new shoots emerged from the soil as offsets. It is apparent 
that the light conditions were such as to permit of limited vegetative 
activity, thus preventing leaf fall, but these conditions were unfavorable 
for strong apogeotropic response. The appearance of the plants in 
January under natural winter illumination only and with the added 
electric illumination is shown in Plate 13, A and B. 
Similar tests were made with dwarf sumac (Rhus copallina L.). Under 
natural illumination only, the leaves were only slightly delayed in coloring 
and falling, the colors being somewhat less intense than usual, and the 
plants did not awaken from dormancy during the winter. With the 
addition of the weak electric illumination, coloring and falling of the 
leaves was delayed for several weeks, although eventually the leaves were 
shed and the plants became dormant. 
Experiments were carried out with the tulip tree (Liriodendron tulipfera 
(L.) in a similar manner. In this case transfer to the electrically illumi¬ 
nated greenhouse in September was promptly followed by renewal of 
active growth of aerial parts with abundant development of new leaves. 
There was no definite period of leaf fall, only individual older leaves 
gradually dying back. There was no abscission, and even after the leaves 
were dead the petioles remained firmly attached to the stem. In this 
manner the plants continued active vegetative development till the 
experiment was discontinued at the end of 18 months without having 
shown any indication of abscission or definite period of shedding of the 
leaves. The control plants soon lost their leaves through abscisson and 
remained dormant through the winter. The appearance of the plants 
under the two conditions of illumination is shown in Plate 14, A and B, 
Thus, in the three species considered it is seen that weak electric light 
(3 to 5 foot-candles) as a supplement to the sunlight of short winter days 
has ranged in effectiveness from simply delaying leaf fall to causing more 
or less complete resumption of normal vegetative activities. 
In the foregoing experiments it is shown that natural shortening of the 
day length in fall and winter results in leaf fall through abscission. It is 
not to be inferred from these results, however, that abrupt change from 
a long to a very short day will accomplish the same result. It already 
has been pointed out (p. 902) that in a large class of plants exposure to an 
illumination period slightly longer than that causing etiolation leads to 
development of an abnormally dark green color in the leaves. Under 
these conditions flowering occurs with difficulty or not at all and visible 
evidence of activity is found chiefly at or below the surface of the soil. 
A somewhat longer illumination period, on the other hand, favors repro¬ 
ductive activities followed by leaf fall and dormancy or death. In such 
plants, therefore, abrupt change from a long to a very short day will not 
lead to shedding of leaves but will result in diverting developmental 
activities downward in the plant, the leaves remaining intact and doubt¬ 
less functioning in some measure. The above-mentioned two sumacs and 
tulip and other woody species were exposed to a io-hour day during the 
summer months. Under these conditions none of the plants shed their 
leaves but, instead, assumed a generally unthrifty appearance and ceased 
upward growth. The leaves became dark green in color and to the touch 
appeared to contain but little sap. These results indicate that a gradual 
