Generalized observers and velocity

measurements in General Relativity

arXiv:gr-qc/0111073v1 21 Nov 2001

1,2 1,3
Paulo Crawford and Ismael Tereno
1. Departamento de Fisica da Universidade de Lisboa
2. Centro de Astronomia e Astrofisica da Universidade de Lisboa
3. Institut dAstrophysique de Paris

Email address:[email protected]

Email address:[email protected]

1
 Abstract

To resolve some unphysical interpretations related to velocity measurements by

static observers, we discuss the use of generalized observer sets, give a prescription

for defining the speed of test particles relative to these observers, and show that,

for any locally inertial frame, the speed of a freely falling material particle is always

less than the speed of light at the Schwarzschild black hole surface.

2
1 Introduction
The radial motion of a test particle falling in a Schwarzschild black hole was treated by
several authors [1, p.298], [2, p.93],[3, pp.19,20],[4, p.342], [5, 6] who reached the same
conclusion that the particle velocity v approaches the light velocity as the test particle
approaches the surface of the black hole, namely the locus r = 2m (with a suitable choice
of units), also known as the event horizon or Schwarzschild radius. All these authors have
in common the use of observers whose worldlines are the integral curves of a hypersurface
orthogonal Killing vector field, that is, static observers (called shell observers in [7, p.2-33])
and, as such, at rest with respect to the mass creating the gravitational field. For example,
Zeldovich and Novikov say that the velocity they use has direct physical significance. It
is the velocity measured by an observer who is at rest (r, , , constant) at the point the
particle is passing. [2, p.93]. The particles motion here is referred to the Schwarzschild
coordinate system in which the line element takes the form
1
2m 2m
    
d 2 = 1 dt2 1 dr 2 r 2 d2 + sin2 d2 , (1)
r r
in geometric units (c = G = 1).
Following along the same lines, Frolov and Novikov recently [3, pp.19,20] add that
The physical velocity v measured by an observer who is at rest in the Schwarzschild
reference frame situated in the neighborhood of the freely moving body is
" #1/2
dx g11 dr [E 2 1 + rg /r]1/2
v= = = . (2)
d |g00 | dt E

If the falling body approaches rg (= 2m), the physical velocity v = dx/d constantly
increases: v 1 as r rg . In Eq.(2), E = (1 2m/r)dt/d is a constant of motion
which we may interpret for timelike geodesics as representing the total energy per unit
rest mass of a particle following the geodesic in question, relative to a static observer at
infinity [8, p.139].
In their very well known textbook [4, p.342], Shapiro and Teukolsky also produce a
similar statement: . . . the particle is observed by a local static observer at r to approach
the event horizon along a radial geodesic at the speed of light . . .

3
 All these statements have contributed to the wrong and widespread view [9] that
makes its way into the literature [10], of a test particle approaching the event horizon
at the speed of light for all observers, and not as a limiting process for a static observer
sitting at r, as r 2m. At the first sight, this view seems quite logical since we expect
the particle to cross the black hole surface in a finite proper time. And if one accepts that
a particle has the speed of light with respect to a static observer (at r = 2m), using locally
the velocity composition law from special relativity, he (or she) concludes that the particle
has the same speed of light with respect to all observers. This is certainly something that
conflicts with the physical observation that, in a vacuum, no material particle travels as
fast as light. This has been very nicely explained by Janis who established that the test
particle does indeed cross r = 2m with a speed less than the speed of light [11]. Here we
take a similar view, and go one step further in obtaining a general expression for geodesic
radial observers in terms of the constants of motion of both observer and test particle.
In Sec. 2 we discuss the different mathematical status of coordinate charts and ref-
erence frames, and compare this present attitude with the early days when it was quite
common the use of curvilinear four-dimensional coordinate system and frame of ref-
erence interchangeably. In Sec. 3, we review some standard results and definitions of
reference frames and observer sets, and give a prescription for the speed at some space-
time point relative to generalized observers at that point. We find the speed of any
material particle to be strictly less than the speed of light. In Sec. 4 we apply this general
prescription to the Schwarzschild field and reproduce Eq.(2) for shell observers, then we
recall that there is a static limit, and we obtain an expression, valid at r = 2m, for the test
particles square speed as a function of the constants of motion of the observer and the
particle, when both follow radially inward geodesics. Finally, we give a brief discussion of
the results.

2 Coordinate Systems and Reference Frames

One of the underlying principles of general relativity is the freedom of choice of coordinates
in the mathematical description of laws and physical quantities. Indeed, the outcome of

4
physical measurements depends, in general, on the reference frame, that is, on the state
of motion of the observer, but cannot depend on the coordinate system chosen, which
may be completely arbitrary and should be selected for convenience in the intermediate
calculation. Of course, certain coordinates may be preferred over other coordinates in the
sense that they are simpler or better adapted to the symmetries of the gravitational field
under consideration.
The association of an arbitrary coordinate system with an arbitrary frame of reference
became standard in the literature for many decades after the advent of general relativity.
Then, it was quite common the use of curvilinear four-dimensional coordinate system
and frame of reference interchangeably, as Bergmann explains in [12, pp.158,159]: . . . we
have always represented frames of reference by coordinate systems . . . . This point is even
stressed when he adds: The equivalence of all frames of reference must be represented
by the equivalence of all coordinate systems.
In our discussion, we find necessary to make a distinction between reference frames
and coordinate systems. By a reference frame we shall mean an observer set by which
measurements are directly made. For example, a set of radially moving geodesic observers
would comprise a frame of reference. On the other hand, a coordinate system refers
to a set of numbers assigned to each point in the space-time manifold. That is, we
follow a common view in which . . . coordinates charts are today given a quite different
mathematical status than that of the frames of reference [13, pp.419-434], whereas they
were previously considered suitable for a given reference frame rather than for an extended
view of the whole manifold.
In Newtonian physics a reference frame is an imagined extension of a rigid body and a
clock. We can then choose different geometrical coordinate systems or charts (Cartesian,
spherical, etc.) for the same frame. For example, the earth determines a rigid frame
throughout all space, consisting of all points which remain at rest relative to the earth
and to each other. One can associate an orthogonal Cartesian coordinate system with
such a reference frame in many ways, by choosing three mutually orthogonal planes and
using the coordinates x, y, z as the measured distances from these planes. As soon as a
time coordinate t is defined one is ready to label any physical event. It should be stressed

5
that this choice of coordinates presupposes that the geometry in such a frame is Euclidean.
But what is precisely a reference frame in general relativity? And how does it differ
from a special relativity inertial frame? To build a physical reference frame in general
relativity it is necessary to replace the rigid body by a fluid [14, p.268] or a cloud of point
particles that move without collisions but otherwise arbitrarily. In more mathematical
terms, one can define [3, p.627] a reference frame as a future-pointing, timelike congruence,
that is, a three-parameter family of curves xa (, y i), where is a parameter along the
curve and y i is a set of parameters that labels the curves, such that one and only one
curve of the family passes each point. If specific parameters and y i are chosen on the
congruence, we define a coordinate system. Of course, this choice is not unique. Thus,
in general, a given reference frame can give rise to more than one associated coordinate
system. And a particular coordinate system may or may not be associated with an obvious
reference frame.
Let us define an observer in a space-time as a material particle parameterized by
proper time [15, p.36]. An observer field (or reference frame) on a space-time M is a
future-pointing, timelike unit vector field. Observers enjoy sending and receiving mes-
sages, and keep close track of their proper time. In special relativity a single geodesic
observer can impose his (or her) proper time on the entire Minkowski space-time, but in
general relativity, a single observer is so local that only cooperation between observers
gives sufficient information[16, p.52], that is, a whole family of observers is needed for
analogous results.

3 Generalized Observers
Given the four-velocity field, u, of an observer set O we parametrize the world lines of
O with the proper time measured by a clock comoving with each observer (wrist-watch
time), so that we have gab ua ub = ua ua = 1; ua is a geodesic reference frame iff in
addition it is parallel propagated along itself: u u = 0. The integral curves of u are
called observers in u (or u-observers, for short). All observers in a geodesic reference
frame are freely falling.

6
 An observer field u on M is stationary provided that exists a smooth function f > 0
on M such that f u = is a Killing vector field, that is, the Lie derivative of the metric
with respect to the vector field vanishes

L gab c c gab + gcba c + gac b c

= a b + b a = 0. (3)

If the one-form corresponding to is also hypersurface orthogonal

a a ,

where and are two scalar fields, then each u-observer is static (i.e., u is integrable).
In this case the integral manifolds u are three-dimensional, spacelike submanifolds that
are isometric under the flow and constitute a common rest space for the u-observers.
Let us consider a test particle given by its 4-velocity vector field ta = dxa /d . We
can decompose ta into a timelike component and a spacelike component by applying a
time-projection tensor, (ua ub), and a space-projection tensor, hab gab + ua ub :

tak = ua ub tb , ta = hab tb . (4)

One can easily verify that tak e ta are timelike and spacelike, respectively. Then we rewrite
the space-time distance ds2 between two events xa and xa + dxa of the test particles
wordline as
ds2 = (ua dxa )2 + hab dxa dxb = dt2 + d2 . (5)

That is, separation of time and space is always possible infinitesimally, and an (instan-
taneous) observer in xa , with four-velocity ua , measures between the two events xa and
xa + dxa of the particles wordline a proper space and proper time given respectively by

d = (hab dxa dxb )1/2 , (6)

and
dt = ua dxa . (7)

The asterisks in Eqs.(6) and (7) denote that the quantities so indicated are not, in general,
exact differentials. The minus sign in Eq.(7) gives dt the same sense as dx0 .

7
 There is a natural way for an u-observer to define the speed of any particle with
four-velocity ta as it passes through an event p M. As the observer has instantaneous
information at p that allows him (or her) to break up the tangent space Tp (M) at p into
time t (parallel to u) and space u , he (or she) will measure
d 2 (gab + ua ub )ta tb
v2 = ( ) = , (8)
dt (ua ta )2
for the square of the speed of the particle at p, which can be written as,
1
v2 = 1 . (9)
(ua ta )2
Whatever is the particles four-velocity, ta , one can always write it as

ta = tak + ta = ua + a , where a ua = 0. (10)

Since ta should be timelike, ta ta = 2 + ||2 < 0 (notice that ||2 = a a = hab ta tb =

(ta ) (ta ) ), and since both ta and ua are future-pointing, = ua ta > 0, and || < =
(||2 + 1)1/2 . From this last equation and from (9) one immediately concludes that under
these conditions v 2 < 1. The number represents the instantaneous rate at which the
observers time is increasing relative to the particles time, and || is the rate at which arc
length d in ua is increasing relative to the particles time, that is,
dt d
= , || = . (11)
d d
Thus the u-observer measures the speed of the t-particle at event p as
d d /d ||
v= = = < 1. (12)
dt dt /d
Notice that, from Eq. (12), v = 1 iff the t-particle is lightlike (ta ta = 0); otherwise, for
timelike particles, v < 1.

4 The Schwarzschild Field Case

Having dealt with this problem in a very general way and proved that the velocity v
of any massive particle with respect to any physical observer is always smaller than the
velocity of light: v < 1, let us apply these ideas to the Schwarzschild gravitational field
and find a general prescription for evaluating v when both the particle and the observer
are geodesic.

8
4.1 Geodesic test particle
Let us suppose that our test particle follows a radially ingoing geodesic in a Schwarzschild
field. Its geodesic equation of motion is the Euler equation for the Lagrangian 2L =
gab x a x b , which is given by
2L = t2 + 1 r 2 , (13)
1
where = g00 = g11 = 1 2m/r, for the Schwarzschild metric Eq.(1) with = const.
and = const., and the dot, as usual, denotes differentiation with respect to proper time.
Along the orbit
2L = 1, (14)

for the particles proper time is given by

d 2 = dt2 1 dr 2 . (15)

From this we could also write

d 2 = dt2 (1 v 2 ), (16)

where !2
2 1 dr
v = 2 , (17)
dt
is, accordingly to Eq.(8), the velocity of the particle with respect to a static observer
(r = constant); i.e. while the particle travels a proper distance 1/2 dr the observer
measure a proper time given by 1/2 dt.
Eq. (13) shows that t is a cyclic coordinate, and
L
= (1 2m/r)t = const. =: E, (18)
t
is the constant of motion along the geodesic associated with the Killing vector field /t;
that is, if the particles 4-velocity ta is geodesic, t t = 0, then: t [g(t, /t)] = 0, which
equally implies Eq. (18).
Inserting Eqs.(17) and (18) into Eq.(14) gives
!2
dr
= E 2 , (19)
d

9
and from this we obtain
1 2m/R
E2 = = , (20)
1v 2 1 v02
where (R, v0 ) are initial conditions; R is the radial coordinate at which the fall begins,
and v0 is the initial velocity.
Now, from Eqs.(18) and (19) we obtain the components of the 4-velocity ta of a radially
ingoing geodesic particle
E 2
 
a
t = , E , 0, 0 , (21)

written in terms of its constant of motion E.

4.2 Static limit

In Landau and Lifchitz [1, p.250] the velocity is measured in terms of proper time, as
determined by clocks synchronized along the trajectory of the particle, as they say. Their
prescription leads to the following expression
!2
2 dx1 
2
 dx1 2
v = g00 + g01 0 g01 g00 g11 ( ), (22)
dx dx0

for the square of the velocity of a radially moving particle.

We have seen earlier, that there is a natural way for the observer ua to measure the
speed of any particle with four-velocity ta as it passes through an event p M, which is
coordinate free, and given by Eq.(9). For a static observer the 4-velocity has the following
components
ua ua = 1 ua = (g00 )1/2 ga0 ,

and for the test particle, its tangent vector to radially inward, timelike geodesics may be
written as !
dx0 dx1
ta = , , 0, 0 .
d d
Inserting these last two 4-vector components in Eq.(9) leads to Eq.(22), which must be
understood as a specialization of Eq.(9) for static observers.

10
 When applied to the (geodesic) radial motion of a free falling particle, Eq.(22) leads
to Eq.(17) which can be rewritten as
" #1/2
1 2m/r
v = 1 , (23)
E2
which is equivalent to Eq.(2). In the case when E = 1, corresponding to R = or v0 = 0,
it reduces to
2m 1/2
 
v= , (24)
r
which coincides with the Newtonian expression
For either expression, v approaches the speed of light at the event horizon (r = 2m)
and they seem to to predict faster-than-light speeds inside the black hole[17]. It is easily
seen that
lim v = 1 and lim v = , (25)
r2m r0

for both Eqs. (23) and (24).

Taken at face value the previous statements would imply that the particles trajectory
should become lightlike in the limit r 2m. However, as the trajectory can be continued
through the event horizon, it seems clear that it must remain timelike there, otherwise we
had to conclude that the particles velocity would overcome the light speed as its worldline
becomes spacelike.
Since this is an unacceptable result and we know that the Schwarzschild coordinate
system is not suitable for describing the manifold at r = 2m it is rather tempting to blame
the coordinate system for this malfunction. But we should ask first, could it be possible
to find a coordinate system that does not have this defect? The answer is obviously no,
since the result is independent of the choice of coordinates, as we have proved in the
third section of this paper, and as will be even clearer at end of this section. Indeed,
even if we use a coordinate system that has no difficulties at r = 2m, like the advanced
Eddington-Finkelstein coordinates, we would still end up with the same result v 2 1 as
r 2m.
We can easily see this by introducing the Eddington-Finkelstein metric [18, p.828],
2m
   
2
ds = 1 dw 2 + 2dwdr + r 2 d2 + sin2 d2 , (26)
r

11
where
r
2m 
 
w(t, r) = t + r + 2m ln  , (27)
2m 
in Eq.(22), valid for static observers. Then, at r = 2m, where g00 = 0, we obtain v 2 = 1.
Thus the real issue here is the choice of frame not the choice of coordinates. For
instance, the process of synchronizing clocks, used by Landau and Lifchitz[1, p.250],
involves the determination of simultaneous events at different spatial locations, which is
a frame-dependent prescription.
Notice also that an observer cannot stay at rest in a Schwarzschild field at r = 2m,
where gab ua ub = 0, for he (or she) cannot have there a timelike four-velocity field tangent
to its worldline. This means that only a photon can stay at rest at r = 2m, and with
respect to this photon-frame all particles have v 2 = 1, as it should be expected.
Another argument that could be given, although it is closely related to the later discus-
sion, is provided by the study of the acceleration of a static observer in a Schwarzschild
field [19]. Whereas a static Newtonian observer is considered to be at rest in its own
proper inertial frame, in general relativity an observer at rest is not geodesic and is
accelerated. To make it clear(er) lets evaluate the acceleration of a static observer in a
spherically symmetric and static gravitational field. Starting with its four-acceleration
field components in Schwarzschild coordinates,

ac = uc; b ub = uc, b + ua cab = g ca g00, a g00

1
/2,

one finds that all nonradial components vanish

1 m
a0 = a = a = 0, and ar = g00,1 = 2 . (28)
2 r
However, ar , which is the radially inward acceleration as calculated using Newtonian
gravity, is a coordinate-dependent quantity and it is not a scalar field. The invariant
acceleration magnitude that we require is

a = (ac ac )1/2 = (g11 )1/2 g00,1 /2.

For the Schwarzschild field this gives

1/2
m 2m

a= 2 1 . (29)
r r

12
The factor (g11 )1/2 , by which the GR and Newtonian accelerations differ, can be neglected
in most cases (r 2m), such as apply e.g. on the surface of planets or even on the surfaces
of normal stars. For instance, for the Sun (2m/R)J = 4.233 106 . But on the surface
of a neutron star (g11 )1/2 may exceed unity by a very large factor, and for a black hole

a , as r 2m.

It follows that a particle at rest in the space at r = 2m would have to be a photon

[20, p.149]. This makes it very clear that we should define a static limit of a black hole,
that is, the boundary of the region of space-time in which the observer can remain at
rest relative to any observer in the asymptotically flat space-time. In plain words, as any
observer must follow a timelike worldline, the static limit is given by

g00 (r) = 0 (static limit).

This emphasizes the point that one cannot use expressions like (17) or (22) at the surface
r = 2m. In other words, there is no observer at rest on that surface. As Taylor and
Wheeler put it in their recent textbook [7, p.3-15]: Shelland shell observerscannot
exist inside the horizon or even at the horizon, where the spherical shells experiences
infinite stresses.

4.3 Radial observers

Considering that the particle and the observer are both in free fall (inward, timelike
geodesics), we can use Eq.(21) and write respectively
E1 q 2 E2 q 2
   
a a
t = , E1 , 0, 0 , u = , E2 , 0, 0 . (30)

Then inserting these into Eq.(9) the following expression is obtained,
2
v2 = 1  2 , (31)

q q
E12 E22 1 1 E12
1 E22

and since = 1 2m/r, it follows that

lim v 2 = 1 0/0.
r2m

13
 1

0.8

0.6

0.4

0.2

0
0 2 4 6 8 10

Figure 1: Test particles square speed v 2 at r = 2m as function of the observers constant

of motion, for two cases (particle with E1 = 2 and E1 = 4).

However, (1 /E 2 )1/2 may be expanded if r 2m since /E 2 1,

1 2
1/2 3
1
 
1 2 =1 +O .
E 2E 2 8E 4 E2
This leads to 1
2 !2
2 (E12 + E22 )

v2 = 1 2 2 2 2
+O (32)
E1 E2 2E1 E2 E2
and we now obtain an exact expression for the velocity at r = 2m,

2 4E22 E12
v (r = 2m) = 1 2 (33)
(E2 + E12 )2

which shows (see figure1) that the value of the velocity at r = 2m is smaller than 1
unless either E1 or E2 are zero or infinity. In particular, when E1 = E2 , we see that
v 2 (r = 2m) = 0. This means that particle and observer have the same initial conditions
at some space-time point p with r > 2m, and from that event onwards they are both on
the same local inertial frame.
Notice that for each particle there are 2 observers who measure the same value of v 2 .
For example observers A and B for one particle and B and C for the other. Notice also

14
that the constant of motion of the particle is always in between the two values of E2 of
those observers. This means, picturing the first particle and the observers A and B all
free falling, that when they all meet at r = 2m, the particle reaches the observer A with
the same velocity that the observer B reaches the particle with(and the same for B,C and
the other particle). In other words, the value of v is positive in the first branch of the
plots (before the minimum) and negative in the second branch.
Considering now the 2 particles, we notice there is an observer (B), with E in between
the values of the particles constants of motion, who measures the same value of v 2 for
both particles. Once again this means that at r = 2m, B catches one of the particles (the
first one) with the same speed with which it is caught by the other.
In fact, owing to symmetry between test particle and observer, both situations are
equivalent. And we could say that figure 1 refers to the observers square speed v 2 at
r = 2m as function of the particles constant of motion.
Now, we highlight the fact that only in the limits E 0 and E (from what we
have just seen, it is indifferent if E refers to the particle or to the observer), we obtain
v = 1. In these cases we conclude the hypothetical observer (or test particle) is in a
photon-frame. In fact referring to Eq.(20) we see these two limits correspond to either
v0 = 1 or R = 2m. With respect to this (unphysical) frame all particles travel at the
speed of light v = 1.
Expressions similar to Eq.(33) can be found [21, 22] for the velocity of a free falling
particle in the Schwarzschild field, derived for diverse non-static observers.
As an example, let us consider a Kruskal observer, an observer which follows an orbit
defined by
dt dx
!
a
u = , ,
d d
with dx = 0, where (t , x ) are the Kruskal coordinates.
For r > 2m, these coordinates (x , t ) relate to the Schwarzschild ones by,
 
r2m


x2 t2 = 2m
er/2m
  (34)
t


t = tanh x


4m

15
and the metric takes the form [18, p.832],

32m3 r/2m  
ds2 = e (dt2 + dx2 ) + r 2 d2 + sin2 d2 . (35)
r
From here we see that an observer which maintains the space-like coordinate x constant,
verifies, !2
32m3 dt
= 1. (36)
rer/2m d
Differentiating Eq.(34) we get,

dx dt dt dx
! !
dr 8m2 dt 8m2
= r/2m x t , = x t . (37)
d e r d d d d d er/2m (r 2m)

Using dx = 0 and Eq.(36) we can write the following equation :

!2 1 !2
2m dt 2m dr 2m(x2 t2 )
  
1 1 = = 1, (38)
r d r d er/2m (r 2m)

which shows this observer follows a radial trajectory.

Consider now a material particle along a radial ingoing geodesic. From Eq.(8), its
velocity, measured by a Kruskal observer is
dx
v= , (39)
dt
since Eq.(35) is diagonal with gx x = gt t , which is analogous to Eq.(17). Dividing one of
the equations Eq.(37) by the other and solving for v we obtain,
dt
1 + tanh(t/4m) dr (1 2m/r)
v= dt , (40)
tanh(t/4m) + dr (1 2m/r)

where dt and dr refer to the movement t(r) of the particle.

We can now introduce the geodesic dt/dr followed by the particle, from Eq.(19) which
we can also explicitly integrate to obtain an expression for t(r) to substitute in Eq.(40).
The details can be found in [22], where the behavior of the v against E plot was found to
be identical to the one presented in this paper.

16
5 Conclusions and Discussion
We have seen that the speed of any material particle following a radially inward geodesic
is strictly less than 1 with respect to any physical (timelike) observer. We recalled that
there is a limit for the use of static observers in a Schwarzschild field given by: g00 (r) = 0.
Thus, we stress the point that one can only use static observers in the space-time region
characterized by r > 2m. We found a formula for the physical velocity of a test particle
in a radially inward, timelike geodesic, measured by an observer in free fall (which crosses
the event horizon simultaneously with the particle) valid at r = 2m.
We conclude that all free falling observers crossing the black hole surface measure the
speed of light (standing still photons at r = 2m) to be v = 1, and they measure the
speed of any material particle to be strictly less than 1.

6 Acknowledgments
During the last part of this work, one of us (PC) visited the Physics Department of the
University of Boston and the Center for Einstein Studies, and he doubly thanks that
university, and in particular John Stachel, for hospitality. He also thanks J. Stachel for
a critical reading of an earlier version of the manuscript and for bringing some references
to his attention. We also thank Rosa Doran for many stimulating discussions on this and
related issues.

17
References

[1] Landau, L. and Lifschitz, E. (1971) The Classical Theory of Fields (Addison-Wesley,

Reading, Mass., 3rd Ed.).

[2] Zeldovich, Ya. and Novikov, I. (1971) Relativistic Astrophysics (Univ. of Chicago
Press, Chicago, Vol. 1).

[3] Frolov, V. and Novikov, I. (1998) Black Hole Physics, (Kluwer Academic Publishers,

Dordrecht).

[4] Shapiro, Stuart L. and Teukolsky, Saul A. (1983) Black Holes, White Dwarfs and
Neutron Stars: The Physics of Compact Objects, (John Wiley & Sons, Inc., NY).

[5] Cavalleri, G. and Spinelli, G. (1973)Motion of Particles Entering a Schwarzschild

field, Nuovo Cimento Lett. 6, 5-8.

[6] Cavalleri, G. and Spinelli, G. (1977) Note on Motion in the Schwarzschild field,
Phys.Rev. D15, 3065-3067.

[7] Taylor, Edwin F., and Wheeler, John A. (2000) Exploring Black Holes, (Addison
Wesley Longman).

[8] Wald, R.M. (1984) General Relativity, (University of Chicago Press, Chicago).

[9] Mitra, A. (1999)Kruskal Coordinates and Mass of Schwarzschild Black Holes, astro-

ph/9904162 electronic preprint.

[10] Mitra, A. (2000) Non-Ocurrence of Trapped Surfaces and Black Holes in Spherical
Gravitational Collapse, Found. Phys. Lett. 13 543-579.

18
[11] Janis, A. (1977) Motion in the Schwarzschild field: A reply, Phys.Rev. D15, 3068-

3069.

[12] Bergmann, P.G. (1942) Introduction to the Theory of Relativity, (Prentice-Hall, New
York, 1942).

[13] Vilain, C. (1992) Spherical Coordinates in General Relativity from 1915 to 1960:
A Physical Interpretation. In Jean Eisenstaedt and A.J.Kox, eds., The Center for
Einstein Studies, vol.3, Studies in the History of General Relativity (Birkhauser,
Boston).

[14] Mller, C. (1972) The Theory of Relativity, (2nd ed. Oxford University Press, Delhi,
1972. First published in 1952 by Clarendon Press, Oxford).

[15] ONeill, B. (1995) The Geometry of Kerr Black Holes (A K Peters, Wellesley, Mas-
sachusetts).

[16] Sachs, R. and Wu, H. (1977) General Relativity for Mathematicians, (New York:
Springer-Verlag).

[17] Sabbata, V., Pavsic, M. and Recami, E. (1977) Black Holes and Tachyons, Lett.
Nuovo Cim. 19, 441-451.

[18] Misner,C., Thorne,K. & Wheeler,J. (1973) Gravitation, (Freeman, San Francisco).

[19] Doughthy, N.A. (1981) Acceleration of a static observer near the event horizon of a
static isolated black hole, Am.J.Phys. 49 412-416.

[20] Rindler, W. (1979) Essential Relativity, (Springer-Verlag, 2nd revised edition).

[21] Janis, A. (1973) Note on Motion in Schwarzschild field, Phys.Rev. D8, 2360-2362.

19
[22] Tereno, I. (1999) Another view on the velocity at the Schwarzschild horizon, astro-

ph/9905298 electronic preprint.

20