SERIES  ARTICLE
Snippets of Physics
T Padmanabhan works at
IUCAA, Pune and is
interested in all areas
of theoretical physics,
especially those which
have something to do with
gravity.
Keywords
Fluid mechanics, electrostatics.
802
9. Ambiguities in Fluid Flow
T Padmanabhan
T h e id e a liz e d ° o w o f ° u id a ro u n d a sp h e r ic a l
b o d y is a c la ssic te x tb o o k p r o b le m in ° u id m e -
c h a n ic s. In te r e stin g ly e n o u g h , it le a d s to so m e
c u r io u s tw ists a n d tu rn s a n d c o n c e p tu a l c o n u n -
d r u m s.
T h e ° ow in d u ced in a ° u id w h en a b o d y m ov es th ro u g h
it is o f trem en d o u s p ra ctica l im p o rta n ce { w ith th e a ir-
p la n e w in g s p rov id in g ju st o n e ex a m p le. In g en era l, n o -
b o d y u n d ersta n d s th e ° ow o f real ° u id s a n d w e h av e
to reso rt to sca led m o d els (e.g ., in w in d tu n n els) o r to
n u m erica l sim u la tio n s to m a k e p ro g ress. B u t th ere a re
so m e idealized m o d els th a t o n e ca n so lv e a n a ly tica lly
w h ich co rresp o n d s, b ro a d ly sp ea k in g , to th e m y th ica l
° u id so m etim es ca lled \ d ry w a ter" . T h ese p ro b lem s a re
su p p o sed ly w ell-u n d ersto o d a n d w e w ill see, in th is in -
sta llm en t, th a t ev en th e sim p lest o f th em ca n lea d to
su rp rises.
W h en a b o d y m ov es th ro u g h th e h y p o th etica l ° u id , th e
resu ltin g ° ow sa tis¯ es th e fo llow in g co n d itio n s: F irst,
th e ° u id ° ow is in co m p ressib le w ith th e d en sity b ein g a
co n sta n t. T h en , th e co n serva tio n o f m a ss, ex p ressed in
th e fo rm o f a co n tin u ity eq u a tio n
+ r ¢ (½ v ) = 0
@½
(1 )
@t
red u ces to th e sim p le co n d itio n r ¢ v = 0 . S eco n d , w e
(in w h ich ½ is th e d en sity a n d v is th e ° u id v elo city )
w ill a ssu m e th a t th e ° ow is irro ta tio n a l (r £ v = 0 )
a sca la r p o ten tia l v = r Á . F in a lly w e w ill ig n o re a ll th e
a llow in g fo r th e v elo city to b e ex p ressed a s a g ra d ien t o f
p ro p erties o f rea l ° u id s lik e v isco sity, su rfa ce ten sio n ,
RESONANCE  September 2008
 SERIES  ARTICLE

so lu tio n s to th e tw o eq u a tio n s r ¢ v = 0 a n d r £ v = 0
etc., a n d w ill trea t th e p ro b lem a s o n e o f ¯ n d in g th e

su b ject to certa in b o u n d a ry co n d itio n s. E q u iva len tly,

r 2 Á = 0 . S o th e p ro b lem red u ces to so lv in g th e L a p la ce

w e ¯ n d th a t th e p o ten tia l sa tis¯ es L a p la ce's eq u a tio n

eq u a tio n w ith v sa tisfy in g th e b o u n d a ry co n d itio n s {

w h ich is th e o n ly n o n triv ia l fea tu re o f th e p ro b lem !
L et u s co n sid er a b o d y o f a n a rb itra ry sh a p e m ov in g
w ith a v elo city u th o u g h th e ° u id . T h en w e n eed to

d itio n n ¢v = n ¢u a t th e su rfa ce, w h ere n is th e n o rm a l

so lv e th e L a p la ce eq u a tio n su b ject to th e b o u n d a ry co n -

to th e su rfa ce. W e w o u ld ex p ect th e ° u id ° ow n ea r th e

b o d y to b e a ® ected b y its m o tio n , b u t a t su ± cien tly
la rg e d ista n ce th is e® ect sh o u ld b e n eg lig ib le. H en ce
th e ° u id v elo city v w ill b e zero a t sp a tia l in ¯ n ity.
In terestin g ly en o u g h th e g en era l fo rm o f th e ° u id v elo c-
ity a t la rg e d ista n ces fro m th e b o d y (o f a rb itra ry sh a p e)
ca n b e d eterm in ed b y th e fo llo w in g a rg u m en t. W e k n ow
th a t th e fu n ctio n 1 = r sa tis¯ es th e L a p la ce eq u a tio n . F u r-
th er, if Á sa tis¯ es th e L a p la ce eq u a tio n , th e sp a tia l d eriv -
a tiv es o f Á a lso sa tisfy th e sa m e eq u a tio n . T h erefo re,
th e d irectio n a l d eriva tiv e o f 1 = r , a lo n g so m e d irectio n
sp eci¯ ed b y a n a rb itra ry v ecto r A w ill a lso sa tisfy th e

b y A ¢ r (1 = r ) a n d w ill fa ll a s 1 = r 2 a t la rg e d ista n ces.

L a p la ce eq u a tio n . S u ch a d irectio n a l d eriva tiv e is g iv en

H en ce, a t la rg e d ista n ces fro m th e b o d y, w e ca n ta k e

th e lea d in g o rd er term s in th e p o ten tia l to b e
µ ¶
Á = ¡ + A ¢r + O (1 = r 3 ):
q 1
(2 )
r r
Y o u w ill, o f co u rse reco g n ize a ll th ese to b e electro sta -
tics in d isg u ise a n d th e ex p a n sio n in (2 ) to b e ju st th e
la rg e d ista n ce ex p a n sio n o f th e p o ten tia l d u e to th e d is-
trib u tio n o f ch a rg es. T h e ¯ rst term is th e m o n o p o le
co u lo m b term a n d th e seco n d o n e is th e d ip o le term .
(In cid en ta lly, th e d ip o le term is ju st th e d i® eren ce in
th e p o ten tia l d u e to tw o ch a rg es k ep t sep a ra ted b y a

RESONANCE  September 2008 803

 SERIES  ARTICLE

d ista n ce A ; clea rly, th e n et p o ten tia l w ill b e th e d irec-

tio n a l d eriva tiv e a lo n g A . T h is is th e q u ick est w ay to
g et th e d ip o le p o ten tia l.) A t su ± cien tly la rg e d ista n ces
w e ig n o re fu rth er term s, o b ta in ed b y ta k in g th e seco n d ,
th ird , ... d eriva tiv es o f 1 = r .
T h e v elo city ¯ eld w ill th en b e th e a n a lo g u e o f th e electric
¯ eld in electro sta tics. F ro m G a u ss' law w e k n ow th a t th e
° u x o f th e electric ¯ eld a t la rg e d ista n ces is p ro p o rtio n a l
to th e `to ta l ch a rg e' q . S in ce w e ca n n o t h av e a n o n -
zero ° u x o f v elo city a t la rg e d ista n ce in o u r p ro b lem ,
it fo llow s th a t q = 0 a n d th e a sy m p to tic fo rm o f th e
p o ten tia l m u st h av e th e fo rm :
A ¢n
µ ¶
Á = A ¢r = ¡
1
; (3 )
r r2
w h ere n is th e u n it v ecto r in th e ra d ia l d irectio n . T a k in g
th e g ra d ien t, w e g et th e v elo city ¯ eld to b e
3 (A ¢ n )n ¡ A
µ ¶
v = (A ¢ r )r
1
= : (4 )
r r3
(T h ese m a n ip u la tio n s a re m o st e± cien tly d o n e u sin g in -
d ex n o ta tio n a n d su m m a tio n co n v en tio n , w ith @ ® r =
(1 = 2 r )@ ® r 2 = x ® = r u sed rep ea ted ly.) T h e a ctu a l fo rm
o f A n eed s to b e d eterm in ed u sin g th e co n d itio n s n ea r
th e b o d y (w h ich w ill b e a m ess fo r a b o d y o f a rb itra ry
sh a p e) b u t it is in terestin g th a t th e ° ow a t la rg e d is-
ta n ces is ¯ x ed en tirely in term s o f a sin g le v ecto r A . In
° u id m ech a n ics, it is a b it o f a su rp rise b u t in electro -
It is interesting that
sta tics it is n o t. If th e m o n o p o le va n ish es, y o u w o u ld
the flow at large
ex p ect th e d ip o le m o m en t to d eterm in e th e b eh av io u r
distances is fixed
o f th e electric ¯ eld a t la rg e d ista n ces.
entirely in terms of a
single vector A. In T h e rea l su rp rise co m es w h en w e try to ca lcu la te th e
fluid mechanics, it is to ta l k in etic en erg y a sso cia ted w ith th e ° u id ° ow g iv en
a bit of a surprise by
but in electrostatics Z
1
it is not. K la b = ½ d 3 x v 2 ; (5 )
2

804 RESONANCE  September 2008

 SERIES  ARTICLE

w h ere th e in teg ra l is ov er a ll sp a ce o u tsid e th e b o d y a n d

th e su b scrip t `la b ' sta n d s fo r th e la b fra m e in w h ich th e
b o d y is m ov in g w ith a v elo city u . (T h e fa ct th a t th e
b o d y is m ov in g is irreleva n t sin ce it o n ly sh ifts th e o rig in
b y u t w h ich is a co n sta n t a s fa r a s th e sp a tia l in teg ra tio n
is co n cern ed .) W h ile th e ° u id ° ow a t la rg e d ista n ces ca n
b e ex p ressed en tirely in term s o f a sin g le v ecto r A , th e
° ow clo ser to th e b o d y ca n b e ex trem ely co m p lica ted .
O n e m ig h t h av e th o u g h t th a t, in su ch a g en era l ca se,
o n e ca n n o t say a n y th in g a b o u t th e to ta l k in etic en erg y
o f th e ° u id . B u t it is in d eed p o ssib le to ex p ress th e to ta l
k in etic en erg y o f th e ° u id ° ow en tirely in term s o f th e
sin g le v ecto r A ev en th o u g h th e ° u id ° ow ev ery w h ere
ca n n o t b e ex p ressed in term s o f A a lo n e. (T h is resu lt,
a s w ell a s eq u a tio n s (8 ) a n d (1 8 ) b elow , a re d eriv ed in
[1 ] b u t n o t d iscu ssed in d eta il in a n y o th er b o o k , a s fa r
a s I k n ow .)

u ) ¢ (v ¡ u ). If w e in teg ra te b o th sid es o f th is eq u a tio n

T o o b ta in th is resu lt, w e u se th e id en tity v 2 = u 2 + (v +

g iv e a co n trib u tio n p ro p o rtio n a l to (V ¡ V b o d y ). In th e

ov er a la rg e v o lu m e V , th e ¯ rst term o n th e rig h t w ill

seco n d term , w e w rite (v + u ) = r (Á + u ¢r ). N ow u sin g

r ¢ v = 0 ; a n d r ¢ u = 0 , w e ca n w rite th e seco n d term
a s a to ta l d iv erg en ce r ¢[(Á + u ¢r )(v ¡ u )]: In teg ra tin g
th is ov er th e w h o le sp a ce, th e seco n d term b eco m es a
su rfa ce in teg ra l ov er b o th th e su rfa ce o f th e b o d y a n d a
su rfa ce a t la rg e d ista n ce. T h a t is, w e h av e p rov ed :
Z I
v d V = u (V ¡ V 0 ) +
2 2
(Á + u ¢ r )(v ¡ u ) ¢ n d S ;
S+ S0
(6 )

w h ere S is a su rfa ce b o u n d in g th e v o lu m e V a t la rg e
d ista n ce a n d S 0 is th e su rfa ce o f th e b o d y. T h e su rfa ce

O n th e su rfa ce o f th e b o d y, (v ¡ u )¢n va n ish es d u e to th e

in teg ra l is ta k en ov er b o th . T h e m ira cle is n o w in sig h t.

b o u n d a ry co n d itio n s a n d h en ce w e g et n o co n trib u tio n s

fro m th ere! T h is is g o o d sin ce w e h a v e n o clu e a b o u t th e

RESONANCE  September 2008 805


1
The electrostatic analogue is
the following. You are given a
distribution of charges with
qtot = 0 and dipole moment p in a
region bounded by a surface S0.
You are also given a constant
vector E0 and told that the nor-
mal component of the electric
field normal to S0 is given by
n . E 0 . Then the electrostatic
energy is proportional to
(4 p.E0–V0 E02) where V0 is the
volume of the region bounded
by S0.
806
SERIES  ARTICLE
p a ttern o f v elo city ° ow n ea r th e b o d y. A t la rg e d ista n ces
fro m th e b o d y, w e ca n u se th e a sy m p to tic fo rm o f th e
v elo city ¯ eld g iv en in (4 ) a n d d o th e in teg ra l ta k in g th e
su rfa ce to b e a sp h ere o f la rg e ra d iu s R . S in ce d S =
a s 1 = R 2 w e ca n a p p rox im a te Á (v ¡ u ) ¢n ¼ ¡ Á u ¢ n o n
R 2 d - in crea ses a s R 2 w h ile v fa lls a s 1 = R 3 a n d Á fa lls
S . H en ce th e su rfa ce in teg ra l in (6 ) o n S b eco m es th e
su m
I I
¡ Á u ¢n R d - +
2
(u ¢ n )(v ¢ n ) R 3 d -
S SI
¡ (u ¢ n )2 R 3 d - : (7 )
S
u sin g th e ea sily p rov ed rela tio n < (A ¢ n )(B ¢ n ) > =
T h e in teg ra tio n ov er a n g u la r co o rd in a tes ca n b e d o n e
(1 = 3 )A ¢ B w h ere < ::: > d en o tes th e a n g u la r average
see th a t th e in teg ra l ov er ¡ (u ¢n )2 R 3 g iv es ¡ u 2 V w h ich
w h ich is 1 = 4 ¼ tim es th e in teg ra l ov er d - . U sin g th is, w e
p recisely ca n cels th e u 2 V in th e ¯ rst term in (6 ). U sin g
(3 ) a n d (4 ) w e g et th e ¯ n a l a n sw er to b e:
½ (4 ¼ A ¢ u ¡ V 0 u 2 ):
1
K la b = (8 )
2
T h u s, if w e k n ow th e m o tio n o f th e ° u id a t v ery la rg e
d ista n ces fro m th e b o d y, w e ca n co m p u te th e to ta l k i-
n etic en erg y o f th e ° u id ° ow w ith o u t ev er k n ow in g th e
v elo city ¯ eld clo se to th e b o d y !1
O n e ca n u se th is to o b ta in a n o th er cu rio u s resu lt. T o
d i® eren t fo rm o f su rfa ce in teg ra l. W ritin g v = r Á , th e
d o th is, w e n o te th a t th e K la b ca n a lso b e ex p ressed in a
ex p ressio n fo r k in etic en erg y red u ces to
Z Z
d 3 x r ¢ (Á r Á );
1 3 2 1
2 V 2 V
K = ½ d x (r Á ) = ½ (9 )
w h ere w e h av e u sed r 2 Á = 0 . U sin g G a u ss' th eo rem ,
th is ex p ressio n ca n b e co n v erted to a su rfa ce in teg ra l
RESONANCE  September 2008
 SERIES  ARTICLE

ov er th e b o d y a n d ov er a su rfa ce a t la rg e d ista n ce. T h e

seco n d o n e va n ish es, g iv in g
I I
K la b = ¡ ½ d S (n ¢ v )Á = ¡ ½ d S (n ¢ u )Á ;
1 1
2 S0 2 S0
(1 0 )

w h ere w e h av e u sed n ¢ v = n ¢ u a t th e su rfa ce. B u t

sin ce w e k n ow w h a t K la b is, th is a llow s u s to o b ta in
a n in teg ra l ov er th e su rfa ce o f th e b o d y o f a p a rticu la r
ex p ressio n , in th e fo rm
I
¡ d S (n ¢ u )Á = (4 ¼ A ¢u ¡ V 0u 2) (1 1 )
S0

ev en th o u g h w e d o n o t k n ow eith er th e sh a p e o f th e
b o d y o r th e v elo city p o ten tia l o n th e su rfa ce!
L et u s n ow sp ecia lize to th e sim p lest o f a ll p o ssib le
sh a p es fo r th e b o d y : a sp h ere o f ra d iu s a . In th is ca se,
th e d ip o le p o ten tia l h a p p en s to b e th e exact so lu tio n
a t a ll d ista n ces o u tsid e th e sp h ere. T h is is n o t d i± cu lt
to u n d ersta n d . G iv en th e sp h erica l sy m m etry, th e o n ly
v ecto r th a t ca n a p p ea r in th e so lu tio n is th e v elo city o f
th e b o d y u . L in ea rity o f th e L a p la ce eq u a tio n (a n d th e
b o u n d a ry co n d itio n ) tells u s th a t th e p o ten tia l m u st b e

fo rm in eq u a tio n (3 ) w ith A / u . U sin g th e b o u n d a ry

lin ea r in th is v ecto r u . H en ce th e so lu tio n m u st h av e th e

co n d itio n n ¢ v = n ¢u a t th e su rfa ce, it is ea sy to sh ow

th a t
1 3
A = a u (1 2 )
2
w h ich co m p letely so lv es th e p ro b lem . W e w ill n ow p lay
a ro u n d w ith th is so lu tio n .
G iv en th e ° u id ° ow p a ttern ev ery w h ere, w e ca n ex p lic-
itly co m p u te th e to ta l k in etic en erg y ca rried b y th e ° ow
u sin g a n y o f th e ex p ressio n s d eriv ed a b ov e. W e g et

RESONANCE  September 2008 807

 SERIES  ARTICLE

Z µ ¶
= ¡ ½ a d - ¡ 2 (A ¢ n )(u ¢ n )
1 2 1
K la b
2 a

½ (4 ¼ ) (A ¢ u ) = m d isp u 2 ;
1 1 1
= (1 3 )
2 3 4
w h ere m d isp is th e m a ss o f th e ° u id d isp la ced b y th e
sp h ere. S o th e to ta l k in etic en erg y is (1 = 2 )[m b o d y +
(1 = 2 )m d isp ]u 2 a n d th e ° u id a d d s (1 = 2 )m d isp to th e ef-
fectiv e m a ss o f th e sp h ere. O f co u rse, o u r g en era l ex -
p ressio n , eq u a tio n (8 ) lea d s to th e sa m e resu lt w h en w e
u se (1 2 ) a n d ev ery th in g seem s ¯ n e.
L et u s n ex t co n sid er th e to ta l m o m en tu m P ca rried b y
th e ° u id w h ich is th e in teg ra l ov er a ll sp a ce o f ½ v . In
a rea so n a b le w o rld , w e w o u ld h av e ex p ected it to b e
(1 = 2 )m d isp u b u t w e a re in fo r a ru d e sh o ck . B y sy m m e-

o n ly n eed to co m p u te th e sca la r P ¢u . B u t sin ce v fa lls

try, th e v ecto r P h a s to b e in th e d irectio n o f u so w e

a s 1 = r 3 a n d th e v o lu m e g row s a s r 3 w e a re in tro u b le!

w ere in teg ra tin g v 2 / 1 = r 6 ov er a ll sp a ce.). E x p licitly,

(T h is d id n o t h a p p en fo r th e k in etic en erg y sin ce w e

w e h av e,
Z
P la b ¢ u = ½ d 3 x 3 [3 (A ¢ n )(u ¢ n ) ¡ A ¢ u ]
1
Z1 Z
r
d - [3 (A ¢ n )(u ¢ n ) ¡ A ¢ u ]:
dr
= ½
a r
(1 4 )
O b v io u sly, o u r p ow er co u n tin g a rg u m en t is co rrect a n d
Whoever would th e r -in teg ra l d iv erg es lo g a rith m ica lly a t la rg e d ista n ces!

su rfa ces g iv es zero b eca u se < 3 (A ¢ n )(u ¢ n ) > = A ¢ u

have guessed that O n th e o th er h a n d th e a n g u la r in teg ra tio n ov er sp h erica l
the simplest
problem in fluid ca n cels th e seco n d term . W h o ev er w o u ld h av e g u essed
flow past a body th a t th e sim p lest p ro b lem in ° u id ° ow p a st a b o d y w ill
will actually lead to
a ctu a lly lea d to a p ro d u ct o f zero a n d in ¯ n ity !
a product of zero If w e d o th e in teg ra l b etw een tw o sp h eres o f ra d ii r = a
and infinity! a n d r = R cen tred o n th e m ov in g sp h ere a t a n y g iv en

808 RESONANCE  September 2008

 SERIES  ARTICLE

in sta n t o f tim e, th en th e a n sw er is in d eed zero b eca u se One feels uneasy

th e a n g u la r av era g e g iv es zero . T h is w o u ld h av e b een about the result being
a n a ccep ta b le resu lt, ex cep t fo r tw o rea so n s. F irst, th e dependent on what
resu lt d ep en d s o n ta k in g th e o u ter b o u n d a ry to b e a one is doing at infinity
sp h ere. If w e ch o o se so m e o th er sh a p e, say, a cy lin d er
especially since the
co a x ia l w ith th e d irectio n o f m o tio n o f th e sp h ere, th e
direction of u breaks
resu lt ca n b e d i® eren t. O n e feels u n ea sy a b o u t th e re-
the spherical
su lt b ein g d ep en d en t o n w h a t o n e is d o in g a t in ¯ n ity
symmetry.
esp ecia lly sin ce th e d irectio n o f u b rea k s th e sp h erica l
sy m m etry.
S eco n d , o n e ca n a rg u e th a t, if y o u p u sh th e sp h ere fro m
rest to let it a cq u ire a v elo city u , th en { in th e p ro cess
{ y o u im p a rt so m e m o m en tu m to th e ° u id . T o d o th is
co m p u ta tio n o n e n eed s to k n ow th e p ressu re w h ich a cts
o n th e sp h ere w h en u is a fu n ctio n o f tim e [2 ]. L et m e

p o in t is th e E u ler eq u a tio n @ v = @ t + (v ¢ r )v = r p = ½ .
b rie° y in d ica te h ow th is ca n b e o b ta in ed . T h e sta rtin g

W h en v = r Á (t;x ), y o u ca n m a n ip u la te th is eq u a tio n
to sh o w th a t r [p + (1 = 2 )½ v 2 + ½ (@ Á = @ t)] = 0 , so th a t
th e p ressu re ca n b e ex p ressed in th e fo rm

p = p1 ¡ ½v ¡ ½ ;
1 2 @Á
(1 5 )
2 @t
w h ere p 1 is th e p ressu re a t in ¯ n ity. (T h is is ju st a tim e
d ep en d en t v ersio n o f B ern o u lli's eq u a tio n .) W e a re in -
terested in th e n et fo rce in th e d irectio n o f m o tio n o f th e
sp h ere, ta k en to b e th e z-a x is, w h ich ca n b e o b ta in ed b y
in teg ra tin g p co s µ ov er th e su rfa ce o f th e sp h ere. F ro m
(4 ) w e see th a t v 2 w ill b e a fu n ctio n o f co s2 µ so th e
co n trib u tio n fro m th e ¯ rst tw o term s w ill va n ish o n in -
teg ra tio n ov er a sp h ere. T h e o n ly su rv iv in g co n trib u tio n
co m es fro m th e la st term w h ich ca n b e ea sily eva lu a ted
to g iv e
Z¼ · ¸
Fz = ¡ 2 1 2 duz 1 duz
2 ¼ a sin µ d µ ½ a co s µ = m d isp :
0 2 dt 2 dt
(1 6 )

RESONANCE  September 2008 809

 SERIES  ARTICLE

C lea rly, th e to ta l m o m en tu m im p a rted is

Z
1
F z d t = m d isp u z ; (1 7 )
2
w h ich m a k es sen se w h en w e rem em b er th a t th e k in etic
en erg y co m es w ith th e e® ectiv e m a ss (1 = 2 )m d isp . S o ,
th is is a n o th er p u rely lo ca l rea so n to b eliev e th e to ta l
m o m en tu m o f th e ° u id ° ow is n o n -zero .
In fa ct, o n e ca n g en era lize th is a rg u m en t a n d o b ta in a
¯ n ite ex p ressio n fo r th e m o m en tu m fo r a n y b o d y m ov -
in g th ro u g h a ° u id [1 ]. O n ce a g a in th e resu lt ca n b e
ex p ressed en tirely in term s o f th e v ecto r A fo r a b o d y o f
arbitrary sh a p e. T o o b ta in th is resu lt, w e ca n u se a trick

m o m en tu m b y th e rela tio n d E = u ¢d P a n d u se th e re-

w h ich rela tes th e in ¯ n itesim a l ch a n g es in th e en erg y a n d

su lt in (8 ). T o p rov e th is rela tio n , let u s a ssu m e th a t

th e b o d y is a ccelera ted b y so m e ex tern a l fo rce F ca u sin g
th e m o m en tu m o f ° u id ° ow to in crea se b y a n a m o u n t

w e im m ed ia tely g et u ¢ d P = F ¢ u d t = d E . G iv en th e
d P in a tim e in terva l d t. F ro m th e rela tio n d P = F d t

fo rm o f E , it is n ow a n elem en ta ry m a tter to v erify th a t

th e to ta l m o m en tu m o f th e ° u id ° ow is g iv en b y

P = 4¼ ½A ¡ ½V 0u : (1 8 )

W e see th a t th is is, in g en era l, n o n -zero . In th e ca se

The need for o f th e sp h ere it d o es g iv e (1 = 2 )m d isp u ; th is is w h a t w e
regularizing the w o u ld h av e n a iv ely ex p ected . O f co u rse, th e a rg u m en t
problem by is d esig n ed to g iv e th is.
introducing a very
large but finite T h e n eed fo r reg u la rizin g th e p ro b lem b y in tro d u cin g a
volume for the total v ery la rg e b u t ¯ n ite v o lu m e fo r th e to ta l ° u id b eco m es
fluid becomes more m o re a p p a ren t w h en w e stu d y th e sa m e resu lt in th e rest
apparent when we
fra m e o f th e sp h ere. In th is fra m e, w e h av e a sp h ere
o f ra d iu s a lo ca ted a ro u n d th e o rig in a n d th e ° u id is
study the same
° ow in g p a st it. T h e b o u n d a ry co n d itio n a t in ¯ n ity is
result in the rest

a co n sta n t va lu e ¡ u a t la rg e d ista n ces. T h is is ea sily

n ow d i® eren t a n d w e ex p ect th e ° u id v elo city to rea ch
frame of the sphere.

810 RESONANCE  September 2008

 SERIES  ARTICLE

a ch iev ed b y a d d in g a co n sta n t electric ¯ eld to a d ip o le in

th e electro sta tic ca se. T h is lea d s to a v elo city p o ten tia l
o f th e fo rm
A ¢n
à = ¡ r ¢u + Á = ¡ r ¢u ¡ : (1 9 )
r2
W e d en o te th e v elo city p o ten tia l in th e rest fra m e b y Ã
to d istin g u ish it fro m th e v elo city p o ten tia l in th e la b
fra m e Á . L et u s n ow a sk w h a t is th e k in etic en erg y o f

° u id v elo city n ow is V = v ¡ u . T h e k in etic en erg y in

th e ° u id in th is fra m e in w h ich th e b o d y is a t rest. T h e

th e rest fra m e w ill b e

Z Z
d x ½ V = ½ d 3 x v 2 + u 2 ¡ 2v ¢u
3 1 2 1 £ ¤
K rest =
Z
2 2
½ d 3 x u 2 ¡ u ¢ P la b + K la b :
1
= (2 0 )
2
W e see th a t th e la st term is th e k in etic en erg y in th e
la b fra m e co m p u ted a b ov e, w h ich is q u ite w ell-d e¯ n ed .
T h e seco n d term is a m b ig u o u s a n d va n ish es if w e u se
sp h erica l reg u la riza tio n w h ile is g iv en b y (1 8 ) if w e u se

K la b ¡ u ¢ P la b = ¡ (1 = 4 )m d isp u 2 is n egative. T h e ¯ rst

lo ca l en erg y co n serva tio n a rg u m en ts. In th e la tter ca se,

term , h ow ev er, w ill b e d iv erg en t if w e ta k e th e v o lu m e o f

th e ° u id to b e in ¯ n ite a n d is p o sitiv e. T h is d iv erg en ce

th en m o st o f it w ill b e m ov in g w ith a v elo city ¡ u in th e

a rises b eca u se if th e ° u id ex ten d s a ll th e w ay to in ¯ n ity

rest fra m e o f th e sp h ere. T h is w ill co n trib u te a n in ¯ n ite

a m o u n t o f k in etic en erg y. W h ile q u ite u n d ersta n d a b le,
it sh ow s th a t G a lilea n in va ria n ce n eed s to b e u sed w ith Address for Correspondence
ca re in th e p resen ce o f a n ex tern a l m ed iu m . T Padmanabhan
IUCAA, Post Bag 4
Suggested Reading Pune University Campus
Ganeshkhind
[1] L D Landau, E M Lifshitz, Fluid Mechanics, Section 10,11, Pergamon,
Pune 411 007, India.
1989.
Email:
[2] T E Faber, Fluid Mechanics for Physicists, Section 4.8, Cambridge
paddy@iucaa.ernet.in
University Press, 1995.
nabhan@iucaa.ernet.in

RESONANCE  September 2008 811