108  Standardization  of  Digitalis  Bodies.  {AmMarch,'iP9i3rm' 
cause  many  of  them  to  die.  Haskell's  observations  on  the  seasonal 
variations  in  the  reaction  of  guineapigs  to  a  fixed  drug,  ouabain, 
show  an  enormous  range  of  variation  in  response  for  different 
months.  Thus,  a  series  run  in  August  gave  the  fatal  dose,  in  terms 
of  grams  of  the  drug  per  gram  of  animal  weight,  as  0.000,000,28 ; 
in  October  the  dose  was  0.000,000,36;  in  November  and  December 
0.000,000,52 ;  in  January  0.000,000,52-f- ;  in  February  0.000,000,35 
to  0.000,000,40;  and  in  March  it  had  come  back  to  0.000,000,36. 
From  the  lowest  fatal  dose  of  a  series  to  the  highest  there  is  a 
difference  of  0.000,000,24  gram  per  kilo  of  animal;  that  is  about 
100  per  cent,  of  the  lowest  and  50  per  cent  of  the  highest.  Haskell 
concludes,  "  From  an  examination  of  these  tables,  it  is  evident  that 
in  assaying  a  preparation  of  the  digitalis  group  upon  guineapigs  a 
standard  preparation  is  as  necessary   as  is  the  case  with  frogs." 
As  is  to  be  expected  the  cat  method  is  also  influenced  by  certain 
adventitious  factors.  Among  those  which  have  been  brought  up 
against  it  is  the  use  of  an  anesthetic.  This  feature  is  one  which 
we  feel  has  given  rise  to  a  number  of  the  great  variations  which 
some  have  reported  from  the  use  of  this  method,  for  it  is  not  a 
difficult  matter  to  kill  a  cat  with  an  overdose  of  ether,  and  some 
of  those  who  report  these  great  variations  have  called  attention  to 
the  fact  that  some  of  their  animals  died  from  respiratory  failure. 
Such  a  mode  of  death  is  exceedingly  uncommon  in  our  experience, 
the  heart  practically  invariably  coming  to  a  standstill  before  respira- 
tion fails,  and  while  the  respiratory  center  is  still  very  active.  This 
fact  is  confirmed  by  the  observations  made  by  Eckler,  who  found 
the  heart  to  have  stopped  in  nearly  every  case  when  the  chest  was 
opened  at  once  after  apparent  death.  Out  of  his  69  experiments  he 
found  the  heart  beating  in  only  seven  animals,  "  and  in  these  the 
contractions  were  very  feeble."  The  tracing  shown  was  taken 
from  the  carotid  artery  and  the  trachea  of  a  cat  and  shows  the  stop- 
page of  the  heart  prior  to  that  of  respiration,  and  also  partly 
explains  the  beating  of  the  heart  found  in  the  seven  cats  reported 
by  Eckler.  This  tracing  is  but  one  of  a  large  number  taken  in  this 
laboratory  and  represents  the  usual  conditions  of  digitalis  death. 
Much  stress  has  been  laid  upon  the  necessity  for  the  continuance 
of  an  absolutely  uniform  grade  of  anesthesia  during  the  course  of 
the  assay.  In  years  of  use  of  the  cat  method  almost  all  of  the 
possible  changes  have  been  rung  on  the  grade  and  uniformity  of 
the  anesthesia  and,  short  of  killing  the  animal  with  the  anesthetic, 
such  variations  seem  to  have  little  or  no  effect  on  the  accuracy  of 
