AmMSch,  r8^8.rm' }   A n  Exudation  from  Larix  Occident alis.  153 
was  of  a  brownish-yellow  color,  somewhat  porous,  and  possessed 
only  a  moderately  sweet  taste  with  a  terebinthinous  flavor.  It  was 
freely  soluble  in  water  on  warming.  The  resulting  solution  was 
neutral  to  litmus  paper,  reduced  Fehling's  solution,  and  darkened 
on  heating  with  sodium  hydrate.  The  product  from  heating  with 
the  latter  when  supersaturated  with  nitric  acid  gave  an  odor  of  cara- 
mel, which  pointed  to  the  presence  of  some  dextrose.  On  heating  a 
weighed  portion  of  the  sample,  after  solution  in  water  and  filtration 
from  the  adhering  particles  of  bark,  with  Fehling's  solution,  19  38 
per  cent,  of  a  reducing  sugar  were  obtained. 
Another  portion  similarly  prepared  and  heated  with  dilute  acid, 
and  then,  after  making  alkaline,  with  sodium  hydrate,  treated  with 
Fehling's  solution  gave  a  total  sugar  value  of  88  07  per  cent ,  indi- 
cating 68-69  Per  cent,  of  non-reducing  sugar  or  similar  substance. 
Treatment  of  a  moderately  strong  aqueous  solution  with  three  or 
four  times  its  volume  of  alcohol  caused  an  abundant  precipitate, 
which  indicated  that  the  non-reducing  portion  was  precipitated, 
while  the  reducing  portion  remained  in  solution;  the  respective 
weights  of  these  two  portions,  obtained  by  alcohol,  confirmed  the 
foregoing  by  Fehling's  solution.  The  following  will  summarize  the 
composition  of  the  specimen: 
Per  Cent. 
Reducing  sugar    .  I9'3% 
Non-reducing  sugar  68  '69 
Moisture  at  too0  C  5  02 
Ash    044 
Wood  fibre,  etc.,  removed  by  filtration  •  •  6*47 
100  00 
The  reducing  sugars  had  many  of  the  properties  of  dextrose, 
but  gave  only  a  slight  precipitate  with  phenylhydrazin.  The 
non-reducing  portion  closely  resembled  dextrin.  There  was  not 
enough  of  the  sample  to  make"  a  complete  study  of  these  two  sub- 
stances. I  naturally  inferred  on  commencing  this  investigation  that 
this  substance  would  closely  resemble  the  Briancon  Manna  and  con- 
sist of  melezitose,  but  the  absence  of  a  very  sweet  taste  and  a  num- 
ber of  other  properties  showed  it  to  be  entirely  different  from  that 
substance.  The  portion  soluble  in  alcohol  yielded  only  traces  of 
an  osazone  with  phenylhydrazin,  but  the  major  part  of  the  original 
substance  was  not  precipitated  by  that  reagent. 
