W. A. Davis, A. J. Daish AxXD G. C. Sawyer 287 



parallel witli it, a rise in the pentosans. It is clear therefore that the 

 increase of pentosan material, which probably includes the gum-like 

 substances mentioned above (p. 285), must take place at a relatively 

 greater rate than the increase of sugars, since their increase does not 

 mask it. Saccharose, hexoses, pentoses and pentosans are all increasing 

 simultaneously during the first period of the day, that is whilst the tem- 

 perature is rising; it is probable, as stated on p. 281, that the hexoses 

 are converted into pentoses and the latter into pentosans. Thus we 

 find the pentoses rising not so quickly as the pentosans whilst the latter 

 are being formed, but from 2 p.m. -4 p.m. when the hexoses are falling 

 a rapid rise of pentoses occurs, whilst the pentosans from this point 

 up to 6 p.m. have ceased to be formed. It is probable that the sudden 

 apparent /«W followed by a rise of pentoses between 4 p.m. and 8 ]).m. 

 is partly a relative effect owing to the large sudden increase and decrease 

 of the saccharose and hexoses between these points; but this would 

 not account for the magnitude of the change (from 0-G8 to 0-4.5 per cent. 

 and back again to 0-71 per cent.), and the two changes are obviously 

 interconnected and take place in opposite directions. From 8 p.m. to 

 2 a.m. there is an actual rise of pentoses, which must be somewhat 

 greater than it appears because it is partly masked by the large increase 

 of saccharose and hexoses during this interval. The fall of pentosans 

 between 8 p.m. and 4 a.m. is probably only an apparent or relative eflect, 

 owing to the large increase in the other sugars, but the rapid rise of 

 pentosans from 4 a.m. onwards corresponds with the actual fall of 

 pentoses from 2 a.m. onwards, which must be larger than it appears 

 becanse the saccharose and hexoses are falling simultaneously. 



The actual range of pentoses during the day is small, viz. 0-34-0-76 

 per cent. 



Series III. Final Stage of Growth, October llth-12th, 1912. 



Changes during the 24 hours. 



Maltose and Starch are again entirely absent. 



Saccharose and Hexoses. These increase immediately after sunrise, 

 the curves (see Fig. 6) rising at first almost parallel to the temperature 

 curve ; the cane sugar increases in amount until midday but from this 

 point until sunset (5 p.m.) the total quantity of sugars present remains 

 nearly constant, the fluctuations between 11 a.m. and 5 p.m. consisting 

 merely of interconversions of the cane sugar and invert sugar. Thus 

 between 1 p.m. and 3 p.m. cane sugar increases, apparently at the expense 



Journ. of Agric. Sci vii 20 



