[Vou. 2 
792 ANNALS OF THE MISSOURI BOTANICAL GARDEN 
colloidal iron (Iron dialysed, Merck) were pipetted in and the 
tube well shaken, the iron then being flocked out with .25 gram 
of NazSO4. Upon the addition of this latter the mixture was 
again thoroughly shaken and the iron precipitate thrown 
down by centrifuging, the resulting clear, supernatant liquid 
then being decanted off through a small filter. This filtrate 
was entirely free of proteins or other substances which, 
through oxidation later, would lead to errors in the perman- 
ganate values. Ten ce. of this filtrate were placed in a 
50 ec. lipped centrifuge tube, and standard Fehling’s solu- 
tion added, the copper content of which was in excess of that 
reducible by the sugar present.t The tube was then placed 
in a boiling water bath for 10 minutes, at the end of which 
time it was centrifuged at a moderate speed for 2 minutes, 
the supernatant unreduced Fehling’s carefully decanted off, 
a like volume of distilled water added, and the cuprous oxide 
again thrown down by a 2-minute centrifuging. All but 1 or 
2 ce. of this wash water was carefully decanted off, and the 
copper dissolved in the smallest amount necessary of a mix- 
ture of equal parts of 10 per cent ammonium ferric sulphate 
and 50 per cent sulphuric acid. It was found that if the cop- 
per were stirred up with a glass rod just before dissolving, it 
went into solution more readily. The dissolved copper was 
titrated directly in the centrifuge tube against N/50 KMn0Os.2 
By calculation it is found that 1 ce. of N/50 KMn0Osz is 
equivalent to 1.27 milligrams of copper, and for the conver- 
sion of this into glucose use was made of the table prepared 
by Shaffer.* 
As stated by Shaffer, care must be observed on the three 
following points: (1) to eliminate all oxidizable substances 
other than sugar, (2) to titrate the cuprous oxide immedi- 
ately after dissolving, (3) to use poor conductors of heat as 
containers of the centrifuge tubes in the water bath, else many 
broken tubes will result. As employed here, circular wire 
1In the determinations made here this amount never exceeded 10 ee. 
* It is necessary to titrate immediately after dissolving because of the danger 
of oxidation of the cuprous oxide. If larger amounts of sugar are concerned, 
N/10 KMnO, may be used. 
