A™kiuary*m)sw'}      A  Study  of  Cannabis  Americana.  17 
during  the  growing  of  the  drug  to  weed  out  the  male  plants.  This 
notion,  according  to  our  experience,  is  based  largely  upon  error,  as 
we  have  found  repeatedly  that  the  Indian  drug  which  contains  large 
quantities  of  seed  is  fully  as  active  as  the  drug  which  does  not  con- 
tain the  seed,  provided  the  seed  is  removed  before  it  is  percolated, 
and  the  experiments  are  based  upon  a  fluid  extract  or  other  phar- 
maceutical product  obtained  from  an  equal  weight  of  drug  minus 
the  seeds.  The  seeds  themselves  do  not  contain  the  active  principle 
upon  which  the  therapeutic  properties  of  the  plant  depend,  but  may 
make  up  a  very  large  percentage  of  the  weight  of  the  drug  as  it 
appears  on  the  market. 
Several  years  ago  we  began  a  systematic  investigation  of  American 
grown  hemp.  Samples  were  obtained  from  the  following  localities 
and  studied : 
(1)  August,  1905,  Mr.  Gaumnitz,  of  the  Department  of  Agricul- 
ture, of  the  University  of  Minnesota,  sent  us  samples  of  hemp  grown 
on  the  college  grounds. 
(2)  1906.    Also  supplied  by  Mr.  Gaumnitz. 
(3)  Grown  in  Mexico,  1903.    Sent  in  for  examination. 
(4)  "      "       "  1904. 
(5)  "      "       "  1906. 
(6)  "      "  Kentucky,  1905. 
(7)  "      "        "  1906. 
(8)  "  near  Detroit,  Mich.,  1907. 
From  these  several  samples  of  Cannabis  Americana,  were  prepared 
fluid  extracts  and  solid  extracts  according  to  the  U.S.P.,  which  were 
tested  upon  animals  for  physiological  activity. 
The  method  of  assay,  which  has  previously  been  called  to  the 
attention  of  this  society,  is  that  which  one  of  us  (Houghton)  devised 
and  has  employed  for  the  past  twelve  years.  This  method  consists 
essentially  of  the  careful  observation  of  the  physiological  effects 
produced  upon  dogs  from  the  internal  administration  of  the  prepa- 
ration of  the  drug  under  test,  compared  with  the  physiological 
effects  produced  by  definite  doses  of  a  standard  preparation  of  the 
drug,  according  to  the  following  method.  It  is  necessary  in  selecting 
the  test  animals  to  pick  out  those  that  are  easily  susceptible  to  the 
action  of  cannabis,  since  dogs  as  well  as  human  beings  vary  consider- 
ably in  their  reaction  to  the  drug.  Also,  preliminary  tests  should  be 
made  upon  the  animals  before  they  are  finally  selected  for  test  pur- 
