s6o  Papers  relative  to  an  Accident 
move  at  the  rate  of  four  or  five  miles  an  hour,  the  wind 
which  occafions  that  motion  is,  I apprehend,  in  general, 
little  more  than  what  is  ufually  faid  to  be  fenfible : for  it 
has  been  frequently  obferved,  that  a traveller  on  horfe- 
back  muft  ride  with  a pretty  good  fpeed,  to  keep  pace 
with  the  fhadow  from  a cloud,  when  it  chances  to  move 
in  the  direction  of  his  journey,  even  though  the  wind  at 
fuch  a time  can  fcarcely  be  faid  to  blow. 
Now  a thunder-cloud,  as  it  is  ufually  called,  moves,  as 
I apprehend,  in  many  cafes  at  leaft,  with  a much  greater 
velocity;  and  though  there  may  be  a few  inftances, 
where  fuch  clouds  move  with  a lefs,  I think,  if  the  mo- 
tion of  them  be  put,  in  general,  at  four  or  five  miles  an 
hour,  it  ought  to  be  confidered,  in  this  cafe,  as  a very 
moderate  computation.  For  thefe  reafons,  the  weight  to 
draw  the  model  was  adjufted  to  that  velocity. 
In  regard  to  the  quantity  of  the  fluid  required  to 
charge  the  great  cylinder,  I found,  from  many  experi- 
ments, that  twenty  uniform,  and  rather  brifk,  turns  of 
the  wheel,  were  the  moft  favourable  for  the  following 
experiments ; becaufe  half  a turn  in  twenty  made  a far 
lefs  difference  in  the  charging,  than  half  a turn  of  the 
fame  wheel,  when  eight  or  ten  turns  only  were  required, 
and  with  nearly  the  fame  uniform  velocity. 
4 
In 
