JSeio  Improved  Type  of  Chronograph.  401 
with  revolving  drums,  where  those  have  to  be  used  in  view 
of  the  nature  of  the  research  (physiological,  physical,  &c.)  in 
spite  of  their  inaccuracy.  All  that  is  required  for  this  is  to 
fix  to  the  axis  of  the  revolving  drum  a  suitable  wheel  with 
projections  making  electrical  contacts  for  our  chronograph, 
at  as  frequent  intervals  as  is  desirable,  and  to  calibrate  the 
value  of  each  revolution  of  the  pen  of  our  instrument  with  an 
ordinary  watch,  at  a  greater  time  interval. 
The  accuracy  of  the  instrument  as  a  time  indicator  and  the 
length  of  the  spiral  allows  of  its  wide  application  for  all  sorts 
of  scientific  and  technical  purposes  :  e.  g.  by  a  corresponding 
variation  in  the  accessory  instruments  for  more  frequent 
contacts  a  considerably  more  accurate  anemometer,  speed 
indicator,  &c,  is  easily  arranged,  allowing  a  very  much 
greater  accuracy  than  has  hitherto  been  practicable.  We 
shall  illustrate  this  only  on  the  two  instruments  mentioned, 
i.  e.  on  the  anemometer  and  the  speed  indicator. 
The  anemometer  has  to  indicate  the  miles  of  wind  in  the 
F 
unit  of  time,  i.  e.  — .     In  the  present  instrument  two  clocks 
are  usually  used  :  one  moves  the  drum  with  the  speed  of  one 
revolution  in  24  hours  (this  gives  t)  ;  the  other  raises  the 
writing-pen  and  allows  it  to  drop,  thus  extending  the  line  for 
contacts  several  times  (this  gives  F) .  Since  the  length  of 
the  curve  for  time  is  very  short,  the  registration  of  t  is  very 
F 
defective,  •—  must  be  defective   also.      "With  our  instrument 
z 
the  revolvin  g  pen  makes  a  contact  on  our  spiral  for  say  each 
F  miles  of!  wind.     The  spiral  is  about  50  times  longer  than 
with  ordinary  instruments  and  is  used  both  for  the  indication 
of  time  and   for  the  contacts.     The   length  of  line  is  about 
10  times  longer  for  F  and  about  50  times  longer  for  t. 
The  speed  indicator  has  to  give  the  number  of  revolutions 
(of    an   electromagnet,    dynamo,    gas-engine,    steam-engine, 
S 
motor  car,  &c.)  in  the  unit  of  time,  i.  e.  — .     In  all  present 
instruments  time  is  given  by  a  revolving  drum.  The 
indication  of  time  (t)  is  very  wrong,  especially  where  large 
strips  of  paper  are  rolled  and  unrolled  from  drums  and  drawn 
by  a  clock  (in  the  first  instance  on  account  of  enormous 
friction).  The  number  of  revolutions  is  given  by  a  rising- 
fly — this  (s)  is  again  wrong,  because  the  rise  of  the  fly  is  not 
directly  proportional  to  the  speed.  Then  the  pen  writes  not 
in  straight  lines,  but  in  arcs.  The  result  is  that  we  get  an 
idea  of  the  speed,  but  no  quantitative  measurement  of  it,  and 
we  have  no  means  of  knowing  exactly  the  speed  at  anv  given 
