Você está na página 1de 9

SERIES  ARTICLE

Snippets of Physics
21. Extreme Physics

T Padmanabhan

V a ria tio n a l p rin c ip le s p la y a c e n tr a l ro le in th e o -


re tic a l p h y sic s in m a n y g u ise s. W e w ill d isc u ss , in
th is in sta lm e n t, so m e c u r io u s fe a tu r e s a sso c ia te d
w ith a c o u p le o f v a r ia tio n a l p r o b le m s.
In H erm a n M elv ille's 1 8 5 1 cla ssic M oby D ick th ere is
a ch a p ter ca lled \ T h e T ry -W o rk s" w h ich d escrib es h ow T Padmanabhan works at
IUCAA, Pune and is
th e try -p o ts o f th e sh ip P equ od a re clea n ed . (In ca se
interested in all areas
y o u h av en 't rea d th e b o o k , a try -p o t is a la rg e ca u ld ro n , of theoretical physics,
u su a lly m a d e o f iro n , w h ich is u sed to o b ta in liq u id o il especially those which
fro m w h a le b lu b b er.) In th a t is th e p a ssa g e: \ It w a s in have something to do with
th e left h a n d try -p o t o f th e P eq u o d ..... th a t I w a s ¯ rst gravity.

in d irectly stru ck b y th e rem a rka b le fa ct, th a t in g eo m -


etry a ll b o d ies g lid in g a lo n g th e cy clo id , m y so a p sto n e
fo r ex a m p le, w ill d escen d fro m a n y p o in t in p recisely th e
sa m e tim e."
T h e rem a rka b le fa ct M elv ille w rites a b o u t is rela ted to
w h a t is k n ow n a s th e b ra ch isto ch ro n e p ro b lem (brach-
istos m ea n in g sh o rtest a n d chron os referrin g to tim e)
w h ich req u ires u s to ¯ n d a cu rv e co n n ectin g tw o p o in ts
A a n d B in a v ertica l p la n e su ch th a t a p a rticle, slid -
in g u n d er th e a ctio n o f g rav ity, w ill trav el fro m A to B
in th e sh o rtest p o ssib le tim e. It w a s k n ow n to J o h a n n
B ern o u lli (a n d to sev era l o th ers; see B ox 1 fo r a ta ste
o f h isto ry ) th a t th is cu rv e is (a p a rt o f) a cy clo id if w e
ta k e th e E a rth 's g rav ita tio n a l ¯ eld to b e co n sta n t. T h e
cy clo id a l p a th a lso h a s th e p ro p erty th a t th e tim e ta k en
fo r a p a rticle to ro ll fro m a n y p o in t to th e m in im u m o f
th e cu rv e is in d ep en d en t o f w h ere it sta rted fro m { w h ich
is w h a t M elv ille w a s ta lk in g a b o u t. In o th er w o rd s, a Keywords
p a rticle ex ecu tin g o scilla tio n s in a cy clo id a l tra ck u n - Brachistochrone, cycloid, extre-
d er th e a ctio n o f g rav ity w ill m a in ta in a p erio d w h ich is mum problem, Bernoulli.

RESONANCE  September 2009 907


SERIES  ARTICLE

in d ep en d en t o f a m p litu d e. T h is is q u ite va lu a b le in th e co n stru ctio n o f p en d u lu m


clo ck s a n d th e ea rly clo ck m a k ers k n ew th is w ell. (T h is ea rn ed cy clo id th e n a m es
iso ch ro n e a n d ta u to ch ro n e a s if b ra ch isto ch ro n e is n o t en o u g h !)
T h e cy clo id is th e cu rv e tra ced b y a p o in t o n th e circu m feren ce o f a w h eel w h ich
is ro llin g w ith o u t slip p in g a lo n g a stra ig h t lin e. F ro m th is it is ea sy to sh ow (see
F igu re 1 ) th a t th e p a ra m etric eq u a tio n (x = x (µ ); y = y (µ )) to a cy clo id h a s th e
fo rm

x = a (µ ¡ sin µ ); y = a (1 ¡ co s µ ) ; (1 )

w h ere a is th e ra d iu s o f th e ro llin g circle. W e sh a ll n ow ta k e a clo ser lo o k a t th is


resu lt.
W h ile th e in itia l so lu tio n to th e b ra ch isto ch ro n e p ro b lem in v o lv ed so m e o f th e
in tellectu a l g ia n ts o f th e sev en teen th cen tu ry, it is n ow w ith in th e g ra sp o f a n
u n d erg ra d u a te stu d en t. L et y (x ) d en o te th e eq u a tio n to th e cu rv e w h ich is th e
so lu tio n to th e b ra ch isto ch ro n e p ro b lem w ith th e co o rd in a tes ch o sen su ch th a t
x is h o rizo n ta l a n d y is m ea su red v ertica lly d ow n w a rd s a s in F igu re 1 . L et th e
p a rticle b eg in its slid e fro m th e o rig in w ith zero v elo city. If th e in ¯ n itesim a l a rc
len g th a lo n g th e cu rv e a ro u n d th e p o in t P (x ;y ) is d s = (1 + y 02 )1 =p2 d x , w h ere
y 0 = (d y = d x ), th en th e p a rticle ta k es tim e d t = d s= v , w h ere v = 2 g y is its
sp eed a t P . T o d eterm in e th e cu rv e w e o n ly n eed to ¯ n d th e ex trem u m o f th e
in teg ra l ov er d t w h ich is a stra ig h tfo rw a rd p ro b lem in th e ca lcu lu s o f va ria tio n .
W e w ill, h ow ev er, a n a ly se it fro m tw o slig h tly d i® eren t a p p ro a ch es.
In th e ¯ rst a p p ro a ch , w e sh a ll m a k e a co o rd in a te tra n sfo rm a tio n w h ich sim p li¯ es
th e p ro b lem co n sid era b ly. L et u s in tro d u ce tw o n ew co o rd in a tes ® a n d ¯ in p la ce
o f th e sta n d a rd C a rtesia n co o rd in a tes (x ;y ) in th e ¯ rst q u a d ra n t b y th e rela tio n s

Figure 1.

908 RESONANCE  September 2009


SERIES  ARTICLE

µ ¶ µ ¶
2
¯ ¯ 2
¯
x = ® ¡ sin ; y = ® 1 ¡ co s ; (2 )
® ® ®

w h ere ® > 0 a n d 0 · ¯ · 2 ¼ ® . O b v io u sly, fo r a ¯ x ed ® , th e cu rv e x (¯ );y (¯ ) is


a cy clo id (w h ich tells y o u th a t w e a re ch ea tin g a little b it u sin g o u r k n ow led g e o f
th e ¯ n a l so lu tio n !). T h e sq u a re o f th e v elo city o f th e p a rticle

v 2 = 2 g y = x_2 + y_2 ; (3 )

w h ere ov erd o ts d en o te d i® eren tia tio n w ith resp ect to tim e, ca n n ow b e ex p ressed
in term s o f ¯_ a n d ®_ b y stra ig h tfo rw a rd a lg eb ra . T h is g iv es th e rela tio n
µ ¶2
¯ ¯
2 g y = 2 y ¯_ + 4 2 ® sin
2
¡ ¯ co s ®_ 2 : (4 )
2® 2®
p
T h e term in v o lv in g ®_ 2 is n o n -n eg a tiv e; fu rth er, sin ce y > 0 w e h av e ¯_ · g.
In teg ra tin g th is rela tio n b etw een t = 0 a n d t = T , w h ere T is th e tim e o f d escen t,
w e g et
ZT ZT
_ p p
¯ (T ) = ¯ dt · g dt = g T : (5 )
0 0

It fo llow s th a t th e tim e o f d escen t is b o u n d ed fro m b elow b y th e eq u a lity T =


p p
¯ (T )= g . T h e b est w e ca n d o is to set ¯_ = g a n d ®_ = 0 to sa tisfy (4 ) a n d h it
th e low er b o u n d in (5 ). S in ce th e req u ired cu rv e h a s ® = co n sta n t, it is o b v io u sly
a cy clo id p a ra m eterized b y ¯ .
T h e a n g u la r p a ra m eter µ = ¯ = ® o f th e cy clo id va ries w ith tim e a t a co n sta n t ra te
p
µ_ = ¯_= ® = g = ® . It is clea r fro m th e p a ra m eteriza tio n in (2 ) th a t th e ra d iu s a o f
2
th e circle w h ich ro ta tes to g en era te th e cy clo id is rela pted to ® b y a = ® . H en ce
th e a n g u la r v elo city o f th e ro llin g circle is ! = µ_ = g = a . If th e p a rticle m ov es
a ll th e w ay to th e o th er en d o f th e cy clo id a t a h o rizo n ta l d ista n ce L = 2 ¼ a , th en
th e tim e o f ° ig h t w ill b e T = 2 ¼ = ! = (2 ¼ L = g )1 = 2 . If L is th e 1 0 0 m , th en w ith
g = 9 :8 m s¡ 2 w e g et T ¼ 8 sec w h ich is b etter th a n th e w o rld reco rd fo r 1 0 0 m
d a sh ! G rav ity seem s to d o q u ite w ell.
T h ere is a n o th er in d irect w ay o f a rriv in g a t th e cy clo id a l so lu tio n th a t is o f so m e
in terest. T h is a p p ro a ch u ses th e co n cep t o f h o d o g ra p h w h ich is th e cu rv e tra ced
b y a p a rticle in th e v elo city sp a ce [1 ]. L et u s try to d eterm in e th e h o d o g ra p h

RESONANCE  September 2009 909


SERIES  ARTICLE

co rresp o n d in g to th e m o tio n o f sw iftest d escen t. F o r sim p licity, co n sid er th e fu ll


tra n sit o f th e p a rticle fro m a p o in t A to a p o in t B in th e sa m e h o rizo n ta l a x is
y = 0 . L et th e sp eed o f th e p a rticle b e v w h en th e v elo city v ecto r m a k es a n a n g le
µ w ith resp ect to th e v x ¡ a x is in th e v elo city sp a ce. T h en th e h o d o g ra p h is g iv en
b y so m e cu rv e u (µ ) w h ich w e a re try in g to d eterm in e. U sin g x_ = v co s µ ; y_ =
v sin µ ;y = v 2 = 2 g , w e ca n w rite th e rela tio n s

dv vdv
dt = ; dx = co t µ : (6 )
g sin µ g
W e a re n ow req u ired to m in im ize th e in tegra l ov er d t w h ile k eep in g th e in teg ra l
ov er d x ¯ x ed . In co rp o ra tin g th e la tter co n stra in t b y a L a g ra n g e m u ltip lier (¡ ¸ ),
w e see th a t w e n eed to m in im ize th e fo llo w in g in teg ra l
Z µ ¶
dv 1
I = ¡ ¸ v co t µ : (7 )
g sin µ

T h e m in im iza tio n is triv ia l sin ce n o d eriva tiv es o f th e fu n ctio n s a re in v o lv ed a n d


lea d s to th e rela tio n v = (1 = ¸ ) co s µ w ith ¡ ¼ = 2 < µ < ¼ = 2 . W e ca n n ow tra d e
o ® th e L a g ra n g e m u ltip lier ¸ fo r th e to tal h o rizo n ta l d ista n ce L (o b ta in ed b y
in teg ra tin g d x ) a n d o b ta in ¸ 2 = ¼ = 2 g L . H en ce, o u r h o d o g ra p h is g iv en b y th e
eq u a tio n
r
2gL
v (µ ) = co s µ ´ 2 R 0 co s µ : (8 )
¼
T h is is ju st th e p o la r eq u a tio n fo r a circle o f ra d iu s R 0 w ith th e o rig in co in cid in g
w ith th e left-m o st p o in t o f th e circle.
H ow d o w e g et to th e cu rv e in rea l sp a ce fro m th e h o d o g ra p h in th e v elo city
sp a ce? In th is p a rticu la r ca se, it is q u ite ea sy. S u p p o se w e sh ift th e circu la r
h o d o g ra p h h o rizo n ta lly to th e left b y a d ista n ce R 0 . T h is req u ires su b tra ctin g
a h o rizo n ta l v elo city w h ich is n u m erica lly eq u a l to th e ra d iu s o f th e h o d o g ra p h .
A fter th e sh ift, w e o b ta in th e h o d o g ra p h o f a u n ifo rm circu la r m o tio n w h ich is,
o f co u rse, a circu la r h o d o g ra p h w ith th e o rig in a t its cen tre. H en ce th e m o tio n
th a t m in im izes th e tim e o f d escen t is ju st u n ifo rm circu la r m o tio n a d d ed to a
rectilin ea r u n ifo rm m o tio n w ith a v elo city eq u a l to th a t o f circu la r m o tio n . T h is
is, o f co u rse, th e p a th tra ced b y a p o in t on a circle th a t ro lls o n a h o rizo n ta l
su rfa ce, w h ich is a cy clo id . T h is a p p ro a ch h a s th e a d va n ta g e th a t y o u o b ta in th e
cy clo id n o t in term s o f eq u a tio n s b u t in term s o f its g eo m etrica l d e¯ n itio n .

910 RESONANCE  September 2009


SERIES  ARTICLE

T h ere is a n in terestin g g en era liza tio n o f th e b ra ch isto ch ro n e p ro b lem w h ich h a s


n o t receiv ed m u ch a tten tio n . T h e cy clo id so lu tio n w a s o b ta in ed u n d er th e a s-
su m p tio n o f a u n ifo rm , co n sta n t g ra v ita tio n a l ¯ eld o f a ° a t E a rth . In rea lity,
o f co u rse, th e g rav ita tio n a l ¯ eld va ries a s (1 = r 2 ) a ro u n d a sp h erica l o b ject. T h e
q u estio n a rises a s to h ow th e cu rv e o f sw iftest d escen t g ets m o d i¯ ed w h en w e
w o rk w ith th e (1 = r 2 ) fo rce.
T o ta ck le th is p ro b lem , it is co n v en ien t to u se p o la r co o rd in a tes in th e p la n e o f
m o tio n a n d a p p rox im a te th e g rav ita tio n a l so u rce a s a p o in t p a rticle o f m a ss M
a t th e o rig in . W e a re in terested in d eterm in in g th e cu rv e r(µ ) su ch th a t a p a rticle
sta rtin g fro m a p o in t A (w ith co o rd in a tes r = R a n d µ = 0 ) w ill rea ch a p o in t B
(w ith co o rd in a tes r = r f ; µ = µ f ) in th e sh o rtest p o ssib le tim e. A s u su a l, so m e
n ew cu rio sities creep in .
T h e m a th em a tica l fo rm u la tio n o f th e va ria tio n a l p rin cip le is triv ia l. If v (r) is th e
sp eed o f th e p a rticle w h en it is a t th e ra d ia l d ista n ce r , th en
µ ¶ µ ¶
2 1 1 2 1
v = 2G M ¡ = C ¡ 1 ; (9 )
r R x
w h ere x = r= R a n d C 2 = 2 G M = R . T h e va ria tio n a l p rin cip le req u ires u s to
m in im ize th e in teg ra l ov er d s = v w h ere d s = R d µ (x 02 + x 2 )1 = 2 is th e a rc len g th a lo n g
th e cu rv e w ith x 0 = d x = d µ . T h is, in tu rn , req u ires d eterm in in g th e ex trem u m o f
th e in teg ra l
Z µ 02 ¶ Z
R x + x 2 1=2
T = dµ ´ L (x 0;x )d µ : (1 0 )
C (1 = x ) ¡ 1
T h e E u ler{ L a g ra n g e eq u a tio n w ill a s u su a l lea d to a seco n d o rd er d i® eren tia l
eq u a tio n in v o lv in g x 00(µ ). B u t sin ce th e in teg ra n d is in d ep en d en t o f µ (tim e), w e
k n ow th a t x 0(@ L = @ x 0) ¡ L is co n serv ed (en erg y ). E q u a tin g it to a co n sta n t K
g iv es a ¯ rst in teg ra l th ereb y a llow in g th e p ro b lem to b e red u ced to q u a d ra tu re.
F a irly stra ig h tfo rw a rd a lg eb ra th en lea d s to th e fo rm o f th e fu n ctio n µ (x ) g iv en
b y th e in teg ra l
Zx r
d x¹ 1 ¡ x¹
µ (x ) = ; (1 1 )
1 x¹ ¸ x¹ 3 + x¹ ¡ 1
w h ere ¸ ´ (R = K C ).
U n fo rtu n a tely, th is is a n ellip tic in teg ra l (a n d a p retty b a d o n e a t th a t) w h ich
m a k es fu rth er a n a ly tic p ro g ress d i± cu lt. W o rk in g th in g s o u t n u m erica lly, o n e

RESONANCE  September 2009 911


SERIES  ARTICLE

Figure 2.

ca n p lo t th e releva n t cu rv es w h ich sh ow a v ery in terestin g b eh av io u r (see F igu re


2 ). T o b eg in w ith , o n e n o tices th a t ea ch cu rv e h a s a tu rn in g p o in t x = l,
say, w h ere (d x = d µ ) = 0 . T h is is a p o in t o f m in im u m a p p ro a ch rela ted to ¸
b y ¸ = l¡ 3 (1 ¡ l). W h a t is cu rio u s is th e a sy m p to tic b eh av io u r o f th e cu rv e
after it tu rn s a ro u n d . It is clea r fro m F igu re 2 th a t th e cu rv es n ev er en ter th e
`fo rb id d en reg io n ' b etw een µ = ¡ 2 ¼ = 3 a n d µ = + 2 ¼ = 3 . W ith so m e h a rd w o rk ,
o n e ca n a ctu a lly p rov e th is resu lt a n a ly tica lly fro m th e fo rm o f th e in teg ra l (1 1 ).
(T ry it o u t y o u rself; d eterm in e th e a n g le µ (l) a t th e p o in t o f m in im u m a p p ro a ch
fro m (1 1 ) a n d th en ca refu lly eva lu a te th e l ! 0 lim it o f µ (l)). In fa ct, th e 3
in (2 ¼ = 3 ) o f th e fo rb id d en zo n e co m es fro m th e p ow er law in d ex o f th e fo rce.
F o r th e b ra ch isto ch ro n e p ro b lem in r ¡ n fo rce law , th e fo rb id d en zo n e is g iv en b y
¡ 2 ¼ = (n + 1 ) < µ < 2 ¼ = (n + 1 ). T h e p a th o f q u ick est d escen t fro m r = R ;µ = 0
to a n y p o in t in th e fo rb id d en zo n e m u st n ecessa rily p a ss th ro u g h th e sin g u la rity
a t th e o rig in .
H av in g d escrib ed th e cla ssic va ria tio n a l p ro b lem w h ich sta rted it a ll, I w a n t to
d iscu ss a n o th er o n e w h ich d o es n o t ev en seem to h av e a resp ecta b le n a m e. T h is
p ro b lem [2 ] ca n b e sta ted a s fo llow s. C o n sid er a p la n et o f a g iv en m a ss M a n d
v o lu m e V a n d a co n sta n t d en sity ½ = M = V . W e a re a sk ed to va ry th e sh a p e o f

912 RESONANCE  September 2009


SERIES  ARTICLE

Figure 3.

th e p la n et so a s to m a k e th e g rav ita tio n a l fo rce ex erted b y th e p la n et o n a g iv en


p o in t a t its su rfa ce th e m a x im u m p o ssib le va lu e. W h a t is th e resu ltin g sh a p e?
M o st p eo p le w o u ld g u ess th a t th e sh a p e is eith er a sp h ere o r so m eth in g lik e th e
a p ex o f a co n e. T h e seco n d g u ess is ea sy to refu te sin ce it p u ts a fa ir a m o u n t o f th e
m a ss aw ay fro m th e ch o sen p o in t; b u t a sp h ere rem a in s a n in trig u in g p o ssib ility.
T h e co rrect a n sw er, h ow ev er, is q u ite stra n g e a n d ca n b e o b ta in ed a s fo llow s.
L et th e ch o sen p o in t b e a t th e o rig in a n d let z -a x is b e a lo n g th e d irectio n o f
th e m a x im a l fo rce a ctin g o n a test p a rticle a t th e o rig in . It is o b v io u s th a t
th is z -a x is m u st b e a n a x is o f sy m m etry fo r th e p la n et; if it is n o t, th en o n e
ca n in crea se th e z -co m p o n en t o f th e n et fo rce b y m ov in g m a teria l fro m la rg er
to sm a ller tra n sv erse d ista n ce u n til th e p la n et is a x ia lly sy m m etric. T h u s o u r
p ro b lem red u ces to d eterm in in g th e cu rv e x = x (z ) (w ith 0 < z < z 0 , say ) w h ich ,
o n rev o lu tio n a ro u n d th e z -a x is g en era tes th e su rfa ce o f th e p la n et. (T h e so lu tio n
is p lo tted a s a th ick u n b ro k en cu rv e in F igu re 3 .)
T h e ea siest w a y to ca lcu la te th e z -co m p o n en t o f th e g rav ita tio n a l fo rce a ctin g o n
th e o rig in is to d iv id e th e p la n et in to circu la r d iscs ea ch o f th ick n ess d z , lo ca ted
p erp en d icu la r to th e z -a x is. T o g et th e fo rce ex erted o n a test p a rticle o f m a ss
m b y a n y sin g le d isc, w e fu rth er d iv id e it in to a n n u la r rin g s o f in n er ra d ii x a n d
o u ter ra d ii x + d x . T h e fo rce a lo n g th e z -a x is b y a n y o n e su ch rin g w ill b e g iv en
by
1 z
d F = G m (½ 2 ¼ x d x ) d z 2 2
p : (1 2 )
x + z x 2 + z2
H en ce th e to ta l fo rce is g iv en b y

RESONANCE  September 2009 913


SERIES  ARTICLE

Z z0 Z x (z )
z
F = 2¼ G m ½ dz x dx
(x + z 2 )3 = 2
2
Z0 z 0 µ0 ¶
3G M m z
= dz 1 ¡ : (1 3 )
2a3 0 (x 2 (z ) + z 2 )1 = 2
In a rriv in g a t th e la st ex p ressio n w e h av e ex p ressed th e d en sity a s ½ = 3 M = 4 ¼ a 3
so th a t th e v o lu m e o f th e p la n et is co n stra in ed b y th e co n d itio n
Z z0
2 4¼ a 3
V = ¼ d z x (z ) = : (1 4 )
0 3
Im p o sin g th is co n d itio n b y a L a g ra n g e m u ltip lier (¡ ¸ ), w e see th a t w e h av e to
essen tia lly ¯ n d th e ex trem u m o f th e in teg ra l ov er th e fu n ctio n
z
L = 1¡ 2 2 1 = 2
¡ ¸x2 : (1 5 )
(x + z )
T h is is stra ig h tfo rw a rd a n d w e g et
z 1
= 2 ¸ = ; (1 6 )
(x 2 + z 2 )3 = 2 z 02
w h ere th e la st eq u a lity d eterm in in g th e L a g ra n g e m u ltip lier fo llow s fro m th e
co n d itio n th a t x (z 0 ) = 0 . O u r co n stra in t o n th e to ta l v o lu m e b y (1 4 ) im p lies
th a t z 03 = 5 a 3 th ereb y co m p letely so lv in g th e p ro b lem . T h e p o la r eq u a tio n to
th e cu rv e is r 2 = 5 2 = 3 a 2 co s µ ; fo r co m p a riso n , a sp h ere w ith th e sa m e v o lu m e w ill
co rresp o n d to r = 2 a co s µ .
T h e sh a p e o f o u r w eird p la n et is sh ow n in F igu re 3 b y a th ick u n b ro k en cu rv e
(a lo n g w ith th a t co rresp o n d in g to a sp h ere o f sa m e v o lu m e). It h a s n o cu sp s a t
th e p o les a n d I a m n o t aw a re o f a n y sp eci¯ c n a m e fo r th is su rfa ce. T h e to ta l
fo rce ex erted b y th is p la n et a t th e o rig in h a p p en s to b e
µ ¶1 = 3
27 GM m G M m
F = 2
¼ 1 :0 3 (1 7 )
25 a a2
w h ich is n o t to o m u ch o f a g a in ov er a sp h ere. B u t th en , a s th ey say, it is th e
p rin cip le th a t m a tters.
T h ere is a m in o r su b tlety w e g lo ssed ov er w h ile d o in g th e va ria tio n in th is p ro b lem .
U n lik e th e u su a l va ria tio n a l p ro b lem s, th e en d p o in t z 0 is n o t g iv en to u s a s ¯ x ed
w h ile d o in g th e va ria tio n o f th e in teg ra ls in (1 3 ) a n d (1 4 ). It is p o ssib le to ta k e

914 RESONANCE  September 2009


SERIES  ARTICLE

B o x 1 . A B it o f H isto r y
One of the early investigations about the time of descent along a curve was by Galileo.
He, like many others, was interested in the time taken by a particle to perform an
oscillation on a circular track which, of course, is what a simple pendulum of length L
hanging from the ceiling will do. Today we could write down this period of oscillation as
s Z
¼ =2
L dµ
T = p ; (1)
g 0 1 ¡ k 2 sin2 µ
where k is related to the angular amplitude of the swing. Of course, in the days before
calculus, the expression would not have meant anything! Instead, Galileo used an in-
genious geometrical argument and { in fact { thought that he has proved the circle to
be the curve of fastest descent. It was, however, known to mathematicians of the 17th
century that Galileo's argument did not establish such a result.
The maj or development came when Bernoulli threw a challenge in 1697 in the form
of the brachistochrone problem to the mathematicians of that day with the interesting
announcement:
\I, Johann Bernoulli, greet the most clever mathematicians in the world. Nothing is
more attractive to intelligent people than an honest, challenging problem whose possible
solution will bestow fame and remain as a lasting monument. Following the example set
by Pascal, Fermat, etc., I hope to earn the gratitude of the entire scienti¯c community
by placing before the ¯nest mathematicians of our time a problem which will test their
methods and the strength of their intellect. If someone communicates to me the solution
of the proposed problem, I shall the publicly declare him worthy of praise" .
Bernoulli, of course, knew the answer and the problem was also solved by his brother
Jakob Bernoulli, Leibniz, Newton and L' Hospital. Newton is said to have received
Bernoulli's challenge at the Royal Society of London one afternoon and (according to
second hand sources, like John Conduit { the husband of Newton's niece) , Newton solved
the problem by night-fall. The \solution" , which was simply a description of how to con-
struct the relevent cycloid, was published anonymously in the P h ilo so p h ica l T ra n sa ctio n s
o f th e R o ya l S ociety of January 1697 (back dated by the editor Edmund Halley) . Newton
actually read aloud his solution in a Royal Society meeting only on 24 February 1697.
Legend has it that Bernoulli had immediately recognized Newton's style and exclaimed
\ta n qu a m ex u n gu e leo n em " meaning `the lion is known by its claw' !

th is in to a cco u n t b y a slig h tly m o re so p h istica ted trea tm en t b u t it w ill lea d to


th e sa m e resu lt in th is p a rticu la r ca se.
Suggested Reading

[1] T Padmanabhan, See e.g., Planets move in circles! A different view of orbits, Resonance, Vol. 1, No.9, p.34, 1996
for a description of hodograph in the case of Kepler problem.
[2] W D Macmillan, Theory of the Potential, Dover Publications, 1958.

Address for Correspondence : T Padmanabhan, IUCAA, Post Bag 4, Pune University Campus Ganeshkhind
Pune 411 007, India. Email: paddy@iucaa.ernet.in, nabhan@iucaa.ernet.in

RESONANCE  September 2009 915

Você também pode gostar