Amoc°TUri^arm'}    British  Pharmaceutical  Conference.  531 
are  capable  of  acting  as  antacids,  but  also  to  mireral  waters  containing  mere 
traces  of  lithium  compounds  and  those  in  the  form  of  chloride  or  sulphate.  In 
a  series  of  experiments  made  to  ascertain  how  far  this  faith  has  a  chemical 
foundation,  he  found  that  the  solvent  action  of  solutions  of  lithium,  sodium 
and  potassium  carbonates  on  a  given  weight  of  uric  acid,  pure  or  as  deposited 
from  urine,  or  on  a  given  weight  of  urinary  sediment  consisting  of  acid  urates, 
under  exactly  similar  conditions,  is  strictly  proportional  to  the  ratio  of  the 
molecular  weights  of  these  solvents.  Further  that  lithium  chloride  and 
lithium  sulphate  have  no  solvent  action  on  uric  acid  and  acid  urates,  and 
lastly  that  the  degree  of  alkalinity  of  urine  produced  by  the  internal  adminis- 
tration of  medicinal  doses  of  lithium  citrate  is  nor  greater  than  that  produced 
by  equivalent  doses  of  potassium  citrate. 
Arsenic  in  Glycerin. — Consequent  upon  some  statements  that  have  been  made 
recently,  respecting  the  occurrence  of  arsenic  in  glycerin,  Mr.  Siebold's  atten- 
tion has  been  directed  also  to  that  subject,  and  he  reported  the  results  cf  the 
examination  of  a  number  of  samples  for  the  presence  of  such  a  contamination. 
The  experiments  were  not  extended  to  low  qualities  of  glycerin,  but  were  con- 
fined to  colorless  and  odorless  sanples  sold  for  toilet  and  pharmaceutical 
purposes.  The  majority  of  the  samples  tested  proved  to  contain  arsenious 
acid  varying  in  proportion  from  1  part  in  4000  parts  to  1  part  in  6000.  A  few 
samples  exceeded  this  proportion,  and  one  contained  as  much  as  1  part  in  2500. 
Some  samples  that  were  found  free  from  arsenic  were  traced  all  to  one  process 
of  manufacture,  which  was  described  as  being  that  by  which  as  a  rule  glycerin 
used  for  dispensing  purposes  is  understood  to  be  made.  The  process  followed 
in  testing,  which  was  demonstrated  by  Mr.  Siebold,  was  a  modification  of  one 
recently  recommended  by  Yulpius  and  Fliickiger,  which  depends  upon  the 
action  of  the  gas  given  off  by  a  mixture  of  the  glycerin  with  zinc  and  very 
dilute  hydrochloric  acid  in  a  test  tube  upon  a  paper  cap  that  has  been  touchf  d 
with  a  drop  of  a  solution  of  mercuric  chloride,  a  yellow  spot  beirjg  the  result 
of  the  presence  of  minute  traces  of  arsenic.  A  further  observation  by  Mr. 
Siebold  was  consistent  with  arsenic  being  sometimes  introduced  into  glycerin 
from  the  glass  of  the  containing  bottle,  and  it  is  proposed  to  further  investigate 
this  point ;  but  there  was  a  general  agreement  among  the  speakers  that  such 
quantities  of  arsenic  as  those  found  in  some  cases  were  more  probably  intro 
duced  during  the  manufacture. 
Lead  in  Water.— The  special  object  of  the  paper  next  read,  by  Mr.  R.  Rey- 
nolds, was  to  point  out  that  the  report  of  a  chemical  analyst  on  the  presence  or 
absence  of  lead  in  any  particular  sample  of  drinking  water,  she uld  be  so 
framed  as  to  guard  against  it  being  taken  to  mean  more  than  it  stated.  It  was 
pointed  out  that  the  quantity  of  lead  taken  op  by  an  acid  water  might  vary 
considerably  with  the  conditions,  and  that  therefore  the  results  of  an  analysis 
of  a  water  that  had  been  subjected  to  a  minimum  influence  might  have  no 
relation  to  the  results  obtained  with  the  same  kind  of  water  after  it  had  been 
submitted  to  more  energetic  or  prolonged  action.  This  was  illustrated  by  the 
results  obtained  in  the  examination  of  a  sample  of  water  drawn  at  7  A.  M., 
when  it  had  probably  lain  in  contact  with  the  pipes  for  some  hours,  which 
gave  0  35  grain  of  lead  per  gallon,  and  one  drawn  three  hours  after,  which 
gave  only  0'1  grain  per  gallon.  It  was  therefore  suggested,  among  other  ways, 
that  relative  certainty  should  be  secured  by  the  analyst  directing  that  samples 
