Researches  on  the  Electromotive  Force  in  the  Contact  of  Metals.  213 
differ  much  in  their  effects  on  human  health ;  but  this  question 
can  only  be  settled  by  observation  and  experiment.  These  valu- 
able means  of  inquiry,  however,  have  been  rendered  unproductive, 
as  they  have  been  guided  by  the  opinion  that  all  malarious 
emanations  are  produced  by  the  plants  growing  in  the  moist 
localities  where  their  evils  are  manifested.  Yet  we  mast  con- 
sider that,  to  impregnate  the  air  which  changes  so  frequently 
over  them,  marshes  must  have  a  very  abundant  source  of  these 
exhalations;  and  it  seems  impossible  that  so  large  an  amount  of 
poisonous  matter  could  exist  in  their  plants  without  being  dis- 
coverable by  the  resources  of  chemical  analysis.  But  the  diffi- 
culty may  be  removed  by  supposing  that  the  poison  consists  of 
volatile  vegetable  oils  collected  from  wide  areas  to  the  low  seats 
of  moisture,  or  occasionally  transported  away  by  rivers,  and  thus 
enabled  to  make  inroads  on  the  human  health  in  distant  regions. 
Cincinnati,  December  30, 18/1. 
XXVI.  Researches  on  the  Electromotive  Force  in  the  Contact  of 
Metals ,  and  on  the  Modification  of  that  Force  by  Heat,  By  E. 
Eblund. 
[Continued  from  p.  98.] 
§4. 
Iron-Copper. 
I  EXPERIMENT  1.  Intensity  of  the  current  =  tan  23°  38'. 
J     The  current  was  reversed  after  three  quarters  of  an  hour. 
Deviations. 
49-11 
49-0  h 48-43 
46-6 1 
42-6  U5-20 
49-0  J 
46-7    48-03        Mean  47-22 
49-7  J 
The  numbers  under  the  heading  "Deviations"  designate  the 
distances  in  millimetres  passed  through  by  the  index  during  45 
minutes.  Thus,  at  the  first  reversal  of  the  current,  the  index 
traversed  49*1  millims.  in  one  direction.  The  current  having 
been  afterwards  restored  to  its  previous  direction,  the  index  re- 
trograded 49  millims.  After  the  time  fixed,  the  current  was 
again  reversed,  and  the  index  gave  466  divisions  of  the  scale 
towards  the  same  side  as  at  first.  Now,  taking  the  mean  of  the 
first  and  third  deviations,  and  adding  to  it  the  second,  the  sum 
divided  by  2  gives  48-43.     All  the  other  means  were  calculated 
