4 
B.  Helland-Hansen. 
[No.  15 
acquired  thereby  further  than  what  we  could  have  discovered  by 
a study  of  the  salinilies  and  temperatures. 
The  exhaustivé  researches  carried  out  by  the  International 
Ocean  Research  Confederation  in  recent  years  have  led  to  systematised 
methods  being-  now  adopted,  which  appear  to  answer  all  expectations. 
Several  very  good  stream-gauges  too  have  been  invented.  We  have 
confined  ourselves  in  the  Norse  investigations  entirely  to  the  use  of 
Ekman’s  propeller  current-meter,  and  all  the  observations  to 
be  mentioned  subsequently  have  been  made  with  this  apparatus.  It 
is  described  by  Dr.  Eeman  in  “Publications  de  Circonstance”  No.  24, 
to  which  we  would  refer  the  reader.  We  have  found  it  an  excelleut 
instrument,  and  it  has  been  very  seldom  indeed  that  it  has  not 
worked  entirely  to  our  satisfaction.  On  the  few  occasions  when 
difficulties  arose  we  were  able  to  get  everything  into  order  again 
almost  immediately.  The  most  common  of  its  minor  defects  was 
that  the  shot  were  sometimes  inclinod  to  stick,  though  this  could 
always  be  set  right  without  much  trouble.  And  no  doubt  even 
this  could  be  entirely  avoided  if  the  shot  were  well  cleaned  beforo 
using  and  if  the  notches  in  the  cog-wheel  into  wdiich  the  shot  fall 
were  sufficiently  spacious  and  presented  to  the  shot  as  little  contact 
surface  as  possible.  It  is  a very  great  advantage  of  Ekmaisi’s 
current-meter  that  it  gives  several  determinations  of  the  direction 
of  the  movement.  If  this  were  not  the  case  the  mean  direction  of 
swift  currents  could  not  in  most  cases  be  found  with  any  accuracy. 
The  apparatus  used  by  us  in  1906  was  No.  7.  The  velocity  of 
the  current  was  calculated  from  the  formula  v = 0.7  0.36  n, 
where  v is  the  velocity  in  centimeters  per  second  and  n the  number 
of  revolutions  of  the  propeller  per  minute.  The  formula,  which  was 
found  by  experiments  in  the  Central  Laboratory,  was  verified  by  us 
in  different  ways  for  velocities  between  15  and  25  cm./sec.  and  proved 
to  be  correct  in  these  cases. 
One  of  the  main  difficulties  in  using  a current-gauge  is  getting 
an  absolutely  fixed  point.  Nansen  and  Ekman  i-ecommended  an- 
choring  the  apparatus  and  kceping  it  afloat  by  means  of  a buoy, 
— the  buoy  riding  free,  and  only  the  anchor  being  connected  with 
