Turning next to the element of variation of temperature we are at once struck by the fact that the curve of movement of the tree is practically a replica of the thermographic curve (Fig. 3). The fall of temperature is seen to induce a rise in the tree and vice versâ. There is a lag in the turning points of the two curves; thus while temperature began to rise at 6 a.m., the tree did not begin to fall till 7 a.m. There is in this case a lag of an hour; but the latent period may, sometimes, be as long as three hours. The delay is due to two reasons; it must take some time for the thick trunk of the tree to attain the temperature of the surrounding, and secondly, the physiological inertia will delay the reaction. As a result of other investigations, I find that the induced effect always lags behind the inducing cause. It is interesting in this connection to draw attention to the parallel phenomenon, which is described below, of lag in the variation of sensibility of Mimosa in response to variation of temperature. In this case the lag was found to be about three hours. Returning to the Palm, the tree continues to fall in the forenoon with rising temperature. At about 2-30 p.m. the temperature was at its maximum after which it began to decline; the movement of the tree was not reversed into erection till after 3-15 p.m., the lag being now 45 minutes nearly.
I may state here that the movement of the tree is not primarily affected by the periodicity of day and night, but by variation of temperature. In spring and in early summer the rise of temperature during the early part of the day and the fall of the temperature from afternoon to next morning, are regular and continuous; the corresponding movements of the tree are also regular. But at other seasons, owing to the sudden change of direction of the wind, the fluctuations of temperature are irregular. Thus at night there may be a sudden rise, and in the earlier part of the day sudden fall of temperature. And the
