s on the celluloses themselves
previously isolated by other processes showed that they were profoundly
and variably attacked by the 'Lange' treatment, wood cellulose losing 50
per cent. of its weight, and filter paper (cellulose) losing 15 per
cent. Further, a specimen of jute yielded 58 per cent. of cellulose by
this method instead of the normal 78 per cent. It was also found that
the celluloses isolated by the process, when subjected to a second
treatment, underwent a further large conversion into soluble
derivatives, and in a third treatment further losses of 5-10 per cent
were obtained. The authors attach value, notwithstanding, to the process
which they state to yield an 'approximately pure cellulose,' and they
describe a modified method embodying the improvements in detail
resulting from their investigation.

_Gabriel's_ method of heating with a glycerin solution of alkaline
hydrate is a combination of 'Hoenig' and 'Lange.' An extended
investigation showed as in the case of the latter that the
celluloses themselves are more or less profoundly attacked by the
treatment--further that the celluloses isolated from lignocelluloses and
other complex raw materials are much 'less pure' than those obtained by
the Lange process. Thus, notably in regard to furfural yielding
constituents, the latter yield 1-2 p.ct. furfural, whereas _specimens of
'jute cellulose'_ obtained by the Gabriel process were found to yield _9
to 13 p.ct. furfural_.

_Cross and Bevan._--Chlorination process yielded in the hands of the
authors results confirming the figures given in 'Cellulose' for yield of
cellulose. Investigation of the products for yield of furfural, gave 9
p.ct. of this aldehyde showing the presence of celluloses, other than
the normal type.

_Conclusions._--The subjoined table gives the mean numerical results for
yield of end-product or 'cellulose' by the various methods. In the case
of the 'celluloses' the results are those of the further action of the
several processes on the end-product of a previous process.

| Methods
| F. Schulze | Weende | Lange | Gabriel | Cross
| | | | | and Bevan
--------------------------------------------------------------------------
Wood cellulose | 98.51 | 91.52 | 48.22 | 55.93 | --
Filter paper cellulose | 99.62 | 95.63 | 78.17 | 79.77 | --
Swedish filte