514 Chibnall.—Diurnal Variations in the 
Sawyer ( 4 ) in their estimations of the diurnal change in sugars and starch 
have all used the percentage dry weight basis. But in these cases it must 
be remembered that the major part of the loss in total dry weight of the 
leaf at night is to be ascribed to the translocation of sugars either pre-exist¬ 
ing as such in the leaf, or formed during the night by the hydrolysis of 
starch. Consequently, as starch and sugars make up only 20-30 per cent, 
of the total dry weight, it follows that the percentage loss in these sub¬ 
stances will always be much greater than that of total solids. Their results 
then, though they show a lesser fall than that which actually occurred, are 
of the utmost value. 
Method C. When using this method it is necessary to consider not only 
changes in the total solids, as has been done above, but in the water-con¬ 
tent of the leaf as well. The diurnal change in the water-content of leaves 
has been the subject of several researches, but only two are worth considering 
here. Livingston and Brown ( 7 ) obtained the water-content of the leaf by 
drying in an oven at 105°. Analysis of the tables given by them for several 
different types of plants shows that the chief fluctuation in the water-content 
took place around midday, when it was natural to expect that the rate of 
transpiration from the stomata would be a maximum. During the night 
the leaves remained more or less saturated, and the values that they give for 
the percentage of water over the hours 6-10 p.m. do not differ by more 
than 1 per cent, from those over the hours 3-6 a.m. Their work was carried 
out in the summer at the Desert Laboratory at Tucson, Arizona, U.S.A., 
and as they themselves point out, in cooler or more humid climates, such J as 
are experienced in England or Germany, smaller fluctuations can reasonably 
be expected. 
Knight ( 5 ), using the same procedure, has recently demonstrated that 
this is indeed so, and his figures for Eupatorium adenophorum indicate 
a change of about o*i per cent, between 5.30 p.m. and 8.30 a.m. It is by 
no means certain, however, that these authors are justified in assuming any 
change in the water-content of their leaves at all. All that they really show 
is that there is a change in the dry weight of the leaf expressed as a percent¬ 
age of the fresh weight. This may well be brought about by translocation 
of solids away from the leaf, for Parkin’s ( 9 ) work on the snowdrop and 
Davis, Daish, and Sawyer’s ( 4 ) on the marigold both show a fall in total 
sugar at night of 3 per cent, of the dry weight, equivalent to about 0-5 per 
cent, of the fresh weight of the leaf. It would appear, then, that the only 
conclusion to be drawn from these results is that the variation in the water- 
content of the leaf at night must be very small. This view is confirmed by 
another observation of Knight’s ( 5 ). He removed a series of leaves from 
a plant and allowed them to wilt for ten, twenty, and thirty minutes respec¬ 
tively before weighing for water determinations. In these cases no trans¬ 
location of solids away from the leaves was possible, consequently his obser- 
