On another night two pots with several plants were exposed for between 35 m. and 40 m. to a clear sky, and perhaps to a rather lower temperature, for a thermometer on a dial, 3 feet high, close by stood at -3.3o to -3.8o C. In one pot three leaves were pinned open, and all were badly injured; of the 44 free leaves, 26 were injured, that is, 59 per cent.  In the other pot 3 leaves were pinned open and all were killed; four other leaves were prevented from sleeping by narrow strips of stiff paper gummed across them, and all were killed; of 24 free leaves, 10 were killed, 2 much injured, and 12 unhurt; that is, 50 per cent. of the free leaves were either killed or much injured.  Taking the two pots together, we may say that rather more than half of the free leaves, which were asleep, were either killed or injured, whilst all the ten horizontally extended leaves, which had been prevented from going to sleep, were either killed or much injured.

Cassia floribunda.—­A bush was exposed at night for 40 m. to a clear sky, the temperature on the surrounding grass being -2o C., and not a leaf was injured.* It was again exposed on

* Cassia laevigata was exposed to a clear sky for 35 m., and C. calliantha (a Guiana species) for 60 m., the temperature on the surrounding grass being -2o C., and neither was in the least injured.  But when C. laevigata was exposed for 1 h., the temp. on the surrounding grass being between -3o and -4o C., every leaf was killed. [page 290]

another night for 1 h., when the temperature of the grass was -4o C.; and now all the leaves on a large bush, whether pinned flat open or free, were killed, blackened, and shrivelled, with the exception of those on one small branch, low down, which was very slightly protected by the leaves on the branches above.  Another tall bush, with four of its large compound leaves pinned out horizontally, was afterwards exposed (temp. of surrounding grass exactly the same, viz., -4o C.), but only for 30 m.  On the following morning every single leaflet on these four leaves was dead, with both their upper and lower surfaces completely blackened.  Of the many free leaves on the bush, only seven were blackened, and of these only a single one (which was a younger and more tender leaf than any of the pinned ones) had both surfaces of the leaflets blackened.  The contrast in this latter respect was well shown by a free leaf, which stood between two pinned-open ones; for these latter had the lower surfaces of their leaflets as black as ink, whilst the intermediate free leaf, though badly injured, still retained a plain tinge of green on the lower surface of the leaflets.  This bush exhibited in a striking manner the evil effects of the leaves not being allowed to assume at night their normal dependent position; for had they all been prevented from doing so, assuredly every single leaf on the bush would have been utterly killed by this exposure of only 30 m.  The leaves