Numerical Solutions of Three: Classes of Nonlinear Parabolic Integro-Differential Equations 1st Edition Jangveladze
Numerical Solutions of Three: Classes of Nonlinear Parabolic Integro-Differential Equations 1st Edition Jangveladze
Numerical Solutions of Three: Classes of Nonlinear Parabolic Integro-Differential Equations 1st Edition Jangveladze
com
https://ebookmass.com/product/numerical-
solutions-of-three-classes-of-nonlinear-
parabolic-integro-differential-equations-1st-
edition-jangveladze/
ebookmass.com
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
https://ebookmass.com/product/asymptotic-perturbation-methods-
for-nonlinear-differential-equations-in-physics-attilio-maccari/
https://ebookmass.com/product/schaums-outline-of-differential-
equations-fifth-edition-bronson/
https://ebookmass.com/product/boundary-value-problems-for-
systems-of-differential-difference-and-fractional-equations-
positive-solutions-1st-edition-johnny-henderson/
https://ebookmass.com/product/first-course-in-differential-
equations-11ed-differential-equations-and-boundary-value-
problems-9ed-solutions-manual-dennis-zill-roberto-martinez/
Advanced Differential Equations 1st Edition Youssef
Raffoul
https://ebookmass.com/product/advanced-differential-
equations-1st-edition-youssef-raffoul/
https://ebookmass.com/product/nonlinear-differential-equations-
in-micro-nano-mechanics-application-in-micro-nano-structures-and-
electromechanical-systems-1st-edition-ali-koochi/
https://ebookmass.com/product/techniques-of-functional-analysis-
for-differential-and-integral-equations-1st-edition-paul-sacks/
https://ebookmass.com/product/mathematical-physics-with-
differential-equations-yisong-yang/
https://ebookmass.com/product/transforms-and-partial-
differential-equations-t-veerarajan/
Numerical Solutions of Three Classes of Nonlinear
Parabolic Integro-Differential Equations
Numerical Solutions of Three Classes
of Nonlinear Parabolic Integro-Differential
Equations
Temur Jangveladze
Ilia Vekua Institute of Applied Mathematics of
Ivane Javakhishvili Tbilisi State University
Tbilisi, Georgia
&
Georgian Technical University
Tbilisi, Georgia
Zurab Kiguradze
Ilia Vekua Institute of Applied Mathematics of
Ivane Javakhishvili Tbilisi State University
Tbilisi, Georgia
Beny Neta
Naval Postgraduate School
Department of Applied Mathematics
Monterey, CA, U.S.A.
Academic Press is an imprint of Elsevier
32 Jamestown Road, London NW1 7BY, UK
225 Wyman Street, Waltham, MA 02451, USA
525 B Street, Suite 1800, San Diego, CA 92101-4495, USA
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK
This book and the individual contributions contained in it are protected under copyright
by the Publisher (other than as may be noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and
experience broaden our understanding, changes in research methods, professional
practices, or medical treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge
in evaluating and using any information, methods, compounds, or experiments
described herein. In using such information or methods they should be mindful of their
own safety and the safety of others, including parties for whom they have a professional
responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or
editors, assume any liability for any injury and/or damage to persons or property as a
matter of products liability, negligence or otherwise, or from any use or operation of
any methods, products, instructions, or ideas contained in the material herein.
ISBN: 978-0-12-804628-9
Temur Jangveladze,
Zurab Kiguradze,
Beny Neta
ACKNOWLEDGMENTS
Introduction
Abstract
The description of various kinds of integro-di erential equa-
tions and a brief history of their origin and applications are
given. The importance of investigations of integro-di erential
models is pointed out as well. Classi cation of integro-di eren-
tial equation is given. The main attention is paid on parabolic
type integro-di erential models. In particular, three types of
integro-di erential equations are considered. Two of them are
based on Maxwell’s equations describing electromagnetic eld
penetration into a substance. The third one is obtained by
simulation of heat ow. At the end of the chapter, as at the
end of each chapter, the comments and bibliographical notes is
given, which consists of description of references concerning to
the topic considered.
t
@ 2 U (x; t) @ 2 U (x; t) @ 2 U (x; )
Z
= + (t; )d ; (1.1)
@t2 @x2 0 @x2
" #
2
@ 2 U (x; t) @ 2 U (x; t)
Z
2 @U (x; t)
+ dx = 0; (1.2)
@t2 0 @x @x2
CHAPTER 1. INTRODUCTION 7
"Z #
2
@ 2 U (x; t) @U (x; t) @ 2 U (x; t)
a dx =0 (1.3)
@t2 0 @x @x2
and
@ 2 U (x; t) 1
a A 2 U (t) AU (t) = 0; (1.4)
@t2
where a(S) a0 = const > 0 is a known function of its argu-
ment and A is a self-adjoint positive operator, i.e., A = A > 0.
The norm used in (1.4) is the one de ned on the range of the
operator A.
In investigating (1.3) and (1.4) type models it is su cient
to mention the following publications: [9], [24], [36], [37], [55],
[61], [65], [130], [309], [327], [328], [349], [355], [356], [357], [373],
[380], [383], [390], [394], [395] though this list is not complete.
Let us note that equations (1.3) and (1.4) are also called Kirch-
ho equations. They, along with some similar equations, de-
scribe important physical processes, among which are linear and
nonlinear dynamics of di erent dimensional bodies (see, for ex-
ample, [36], [355], [373] and [471]).
In other questions connected with hereditary phenomena,
one nds the integro-di erential equations of a parabolic type,
which were investigated by Evans [160]. These equations look
like
Z t 2
@U (x; t) @ 2 U (x; t) @ U (x; )
2
+ A(t; )d = 0; (1.5)
@t @x t0 @x2
@W (x; t)
+r [a(S(x; t))r W (x; t)] = 0;
@t (1.6)
r W (x; t) = 0;
where Z t
S(x; t) = jr W (x; )j2 d : (1.7)
0
In system (1.6), r W and r W are the usual vector oper-
ators with respect to the variables x = (x1 ; x2 ; x3 ): Even one-
dimensional scalar version of this model is very complicated and
its investigation has been possible yet only for special cases. The
one-dimensional scalar case of the model (1.6), (1.7) has the fol-
lowing form
" Z ! #
t 2
@U (x; t) @ @U (x; ) @U (x; t)
= a d ; (1.8)
@t @x 0 @x @x
began in the works [138], [187], and [188]. Since then many
scienti c publications were devoted to the investigation of exis-
tence and uniqueness of their solutions under various types of
initial and boundary conditions. In this respect, especially sig-
ni cant are the works [49], [50], [141], [146], [147], [219], [220],
[238], [254], [256], [302], [303], [304], [305], [306], [307], [322],
[333], [334], and reference therein. Authors of this book have
also made contribution in this direction, for example, see [137],
[138], [139], [140], [141], [145], [146], [147], [186], [187], [188],
[219], [220], [223], [238], [247], [248], [253], [254], [256], [261],
and reference therein.
Making certain physical assumptions in mathematical de-
scription of the above-mentioned process of penetration of elec-
tromagnetic eld into a substance, Laptev [306] has constructed
a new integro-di erential model, which represents a general-
ization of the system introduced in [187] and [188]. Founded
on Maxwell’s system the following parabolic integro-di erential
model is obtained
Z tZ
@W (x; t)
=a jr W (x; )j2 dxd W (x; t): (1.9)
@t 0
Z t
@U (x; t)
=r a jrU (x; )j2 d rU (x; t) (1.10)
@t 0
and
Z tZ
@U (x; t)
=a jrU (x; )j2 dxd U (x; t); (1.11)
@t 0
respectively.
In equations (1.10) and (1.11) we have x = (x1 ; : : : ; xn ) and
the vector operator rU is given by
@U (x; t) @U (x; t)
rU = gradU = ;:::; = (D1 U; : : : ; Dn U ) :
@x1 @xn
Some generalizations of the models (1.6)-(1.11) are given in
the works [137], [139], [145], [219], [305], and [306]. One kind of
these models has the forms:
n
@U (x; t) X
Di a(S(x; t)) jrU (x; t)jq 2
Di U (x; t)
@t i=1 (1.12)
= f (x; t)
and
n
@U (x; t) X
a(S(t)) Di jrU (x; t)jq 2
Di U (x; t)
@t i=1 (1.13)
= f (x; t);
CHAPTER 1. INTRODUCTION 11
where Z t
S(x; t) = jrU (x; )jq d
0
and Z tZ
S(t) = jrU (x; )jq d :
0
The models of type (1.6)-(1.13) are complex and have been
intensively studied by many authors. The existence and unique-
ness of global solutions of initial-boundary value problems for
equations and systems of type (1.6) - (1.13) were studied in
[49], [50], [137], [138], [139], [140], [141], [145], [186], [187],[188],
[219], [223], [238], [247], [248], [253], [261], [302], [303], [304],
[305], [306], [307], [322], [333], [334], and in a number of other
works as well.
The existence theorems that are proved in [137], [138], [139],
[140], [141], [145], [187], [188], and [219] are based on a-priori
estimates, modi ed Galerkin’s method and compactness argu-
ments as in [327], [328], [461], and [462] for nonlinear elliptic
and parabolic equations.
For equation (1.8) with nonhomogeneous right-hand side and
a(S) = 1 + S, or for equation (1.13) in the one-dimensional
case with q = 2, such theorem for rst initial-boundary value
problem is proved in section 3.6.
The asymptotic behavior as t ! 1 of the solutions of such
models has been the object of intensive research in recent years,
see [28], [29], [31], [32], [33], [34], [35], [139], [141], [145], [146],
[147], [217], [218], [219], [220], [223], [227], [228], [229], [231],
[232], [233], [234], [235], [236], [237], [238], [239], [240], [241],
[242], [243], [245], [246], [247], [248], [251], [252], [253], [254],
[256], [257], [261], [276], [277], [278], [280], [282], [283], and ref-
erence therein.
Another model considered in this book and studied by one
of the authors of this monograph [376] is
12 CHAPTER 1. INTRODUCTION
Z t
@U (x; t) @ @U (x; )
= a(t ) d + f (x; t): (1.14)
@t 0 @x @x
[452], [453], [454], [470], [475], [476], [477], [485], [486], [487],
[489], [491], and references therein.
As we have already pointed out, by using Maxwell’s system
[300] for mathematical simulation of the process of electromag-
netic eld penetrating into a substance, new class of integro-
di erential models (1.6), (1.7), (1.8), and (1.10) arises (see [187]
and [188]).
Integro-di erential equations arise in many other practical
processes as well, see, for example, [11], [21], [22], [36], [37],
[55], [59], [74], [84], [119], [127], [128], [164], [185], [187], [188],
[295], [298], [306], [310], [311], [312], [355], [356], [357], [363],
[373], [380], [384], [405], [425], [459], [463], [465], [466], [467],
[468], [469], [471], [484], and in a number of other works as well.
The motivation for studying integro-di erential problems
comes from the many physical models in such elds as elec-
tromagnetic wave propagation, heat transfer, nuclear reactor
dynamics, and thermoelasticity. Besides the integro-di erential
equations arise in many spheres of human activity as well. For
example, the second order fully nonlinear integro-di erential
equations are derived from the pricing problem of nancial deriva-
tives and optimal portfolio selection problem in a market [59].
In [84] nonlinear integro-di erential equations that arise from
stochastic control problems with purely jump Levy processes
are considered.
Many problems of modern science and engineering can be
described by partial integro-di erential equations. Since quite
a lot of these problems are time-dependent, most of them are
evolution equations and especially nonlinear evolution parabolic
equations, see [93], [177], [179], [180], [191], [216], [310], [311],
[312], [322], [332], [345], [346], [358], [451], and references therein).
Many scienti c works are devoted to investigation and nu-
merical solution of parabolic integro-di erential models, see [6],
[18], [45], [53], [62], [63], [69], [74], [83], [86], [87], [93], [98], [103],
16 CHAPTER 1. INTRODUCTION
[108], [115], [118], [128], [137], [139], [145], [155], [156], [163],
[175], [177], [181], [182], [191], [198], [247], [253], [263], [265],
[266], [267], [276], [279], [281], [306], [308], [310], [311], [312],
[313], [319], [320], [322], [323], [327], [353], [358], [370], [376],
[378], [389], [393], [427], [430], [432], [433], [447], [448], [449],
[450], [451], [456], [479], [483], [488], [492], [497], and references
therein.
Study of the models of type (1.6), (1.7), and (1.8) has begun
in the works [138], [187] and [188]. In these works, in particu-
lar, the theorems of existence of solution of the initial-boundary
value problem (with rst (Dirichlet) boundary conditions) for
scalar equation with one-dimensional space variable are proved.
Investigations of higher space dimensions for model (1.10) car-
ried out initially in [137] and [140].
In [306] some generalization of the system of type (1.6), (1.7)
was proposed. In particular, assuming that the temperature
of the considered body depends on the time variable, but in-
dependent of the space coordinates, then the same process of
penetration of a magnetic eld into the material is simulated
by the averaged (as the author of [306] has named it) integro-
di erential models (1.9) and (1.11).
Study of the models of type (1.9) and (1.11) has started in
the works [217] and [219].
One must note that some works were devoted to the study
of modeling of physical process of electromagnetic eld pene-
tration in the case of cylindrical conductors. In this case, the
above-mentioned integro-di erential model (1.6), (1.7), written
in cylindrical coordinates, was given in [148]. The work [333]
is devoted to the investigation of periodic problem for one-
dimensional (1.8) type model in cylindrical coordinates.
Interest in the above-mentioned integro-di erential (1.6) -
(1.11) models is increasing. Some generalizations of (1.10) and
(1.11) models, which have the forms (1.12) and (1.13) corre-
1.1. COMMNETS AND BIBLIOGRAPHICAL NOTES 17
Mathematical Modeling
Abstract
The chapter consists of six sections. The chapter concerns
mathematical modeling of the investigated in the monograph
equations. Some mathematical features of these models are
studies as well. In the rst section general statement of di usion
process is given. Sections two and three are dedicated to a re-
duction of Maxwell’s equations to the integro-di erential mod-
els. Consequently, two types of integro-di erential equations
are obtained which are called Model I and Model II accordingly.
Both models in di erent physical assumptions describe process
of penetration of an electromagnetic eld into a substance. In
the third section Model III is considered, which represents a
special model for one-dimensional heat ow in materials with
memory. This model arises in the theory of one-dimensional
viscoelasticity as well. Next two sections are devoted to some
mathematical features of all three models above. The existence
and uniqueness properties of the solutions as well as asymp-
totic behavior of solutions of the appropriate initial-boundary
value problems are presented. This chapter closes with some
concluding remarks and bibliographical overview of the three
investigated models.
1 @( H)
= r E; (2.1)
c @t
r ( H) = 0; (2.2)
4
E = r H; (2.3)
c
where E = (E1 ; E2 ; E3 ) and H = (H1 ; H2 ; H3 ) are vectors of
electrical and magnetic elds, respectively, is the magnetic
permeability, is the conductivity of substance, and c is the
speed of light in vacuum.
In (2.3), following an assumption of quasi-stationarity, the
term proportional to @E=@t is omitted and Ohm’s law is used,
relating the vector E to a vector of density of a current J by
the following relation
J = E: (2.4)
2.1. ELECTROMAGNETIC DIFFUSION PROCESS 21
@
cv = EJ = E 2 ; (2.5)
@t
where cv is the speci c heat capacity of the environment. Thus
the factor of heat capacity can also depend on temperature (as
a rule, in a power-like form).
Equations (2.1), (2.3), (2.5) form the closed system for def-
inition of an electromagnetic eld and temperature under an
appropriate initial and boundary conditions.
The questions of existence and uniqueness of the solutions of
linear di erential problems ( = (x)), in general enough state-
ment, are considered in [136], [297], and in a number of other
works. Thus, in [297] the movement of environment is supposed
within the framework of magnetic-hydrodynamical approach. In
these works the transition from classical statement to the gener-
alized one is made. The requirement that the functions satisfy
equation (2.2) and boundary conditions is replaced with the re-
22 CHAPTER 2. MATHEMATICAL MODELING
@
0 cv = J 2: (2.7)
@t
Here is density of the substance, 0 is the density at time t = 0,
and cv (as was already mentioned in 2.1.1) is its speci c heat
capacity. Generally they also depend on temperature. Equation
(2.7) does not take into account transfer of heat due to heat
conductivity and radiation. Number of other physical e ects
are not considered. However, from the mathematical point of
view system (2.6), (2.7) is complex enough.
Now we shall begin the reduction of system (2.6), (2.7) to a
system of nonlinear integro-di erential equations.
Let us rewrite equation (2.7) in the following form
cv @
0( ) = J 2:
( ) @t
we have
@S
= J 2:
@t
Let us assume that the process begins at the moment of time
t = 0 in which the temperature 0 of the substance is constant.
Integrating this equation on a segment [0; t]; we shall nd
Z t
S( (x; t)) S( 0 ) = J 2d :
0
24 CHAPTER 2. MATHEMATICAL MODELING
t
c2
Z
@H c 2
+ r r H d r H = 0;
@t 4 0 4
r H = 0:
Let us introduce the notations
c2 c
a(S) = ( (S)); W = H
4 4
and rewrite the system in the following form:
Z t
@W
+r a jr W j2 d r W = 0;
@t 0 (2.8)
r W = 0:
2.1. ELECTROMAGNETIC DIFFUSION PROCESS 25
@U @U
r W = ; ;0
@x2 @x1
and (2.8) takes the form
Z t
@U
=r a jrU j2 d rU : (2.9)
@t 0
Let us note again that equations such as (2.8), (2.9) for the
rst time have arisen in [187] and [188]. In these works, as well
as in [137], [138], [139], [140], [141], [145], [186], [219], [223],
[238], [247], [253], and [261] together with other questions, the
uniqueness of the solutions of the initial-boundary value prob-
lems for equations (2.8) and (2.9) is given, under a rather general
assumptions on the function a = a(S).
Assume that 0 and cv are constants. Let us give examples
of function ( ), inducing a(S), see, e.g., [145], [304], and [306].
If ( ) = , > 1, then
a(S) = C1 (C0 S) 1;
where C0 and C1 are some positive constants. Thus, the power
growth of the resistance ( ) gives a coe cient determined only
on a nite interval. Let us notice that the physical substances
do not have such property.
If ( ) = , then
a(S) = C1 eC0 S ;
with positive constants C0 and C1 , i.e., the linear growth of the
function ( ) gives exponential function a(S). The linear growth
26 CHAPTER 2. MATHEMATICAL MODELING
a(S) = C1 (C0 + S) 1 ;
with positive constants C0 and C1 . Thus, under-linear growth
of the function ( ) gives sedate growth of the coe cient a(S).
If ( ) = , < 0, then
da c2 d d c2 d =d c2
= = = ( ) 0 ( ):
dS 4 d dS 4 dS=d 4 cv
Z tZ
@W 1
+r a jr W j2 dxd r W = 0;
@t j j 0
28 CHAPTER 2. MATHEMATICAL MODELING
r W = 0:
Here j j is the volume of the substance. Due to averaging the
coe cient of this equation depends only on the variable t, and
consequently the equation can be rewritten in the form
Z tZ
@W
=a jr W j2 dxd W: (2.15)
@t 0
r r W = W + r(r W ):
For a at eld W = (0; 0; U ), where U = U (x1 ; x2 ; t) is a
function of two spatial variables, system (2.15) becomes
Z tZ
@U
=a jrU j2 dxd U: (2.16)
@t 0
@U @ 2U
=k 2;
@t @x
where U (x; t) is the absolute temperature and k > 0 is a con-
stant called thermal di usivity. This parabolic equation allows
a thermal disturbance at any point in the medium to be in-
stantly felt at other points, see [196]. Gurtin and Pipkin [196]
2.3. MATHEMATICAL MODEL III 29
which is (1.14).
" Z ! #
t 2
@U @ @U @U
a d = f (x; t) (2.22)
@t @x 0 @x @x
and
2.4. SOME FEATURES OF MODELS I AND II 31
!
Z tZ 1 2
@U @U @ 2U
a dxd = f (x; t): (2.23)
@t 0 0 @x @x2
Z Z
@U
+ A(U ) V dxdt = f V dxdt; (2.26)
QT @t QT
@ 2V @ @V
2
= a ;
@t @x @x
which generally does not possess global solutions.
The following existence and uniqueness statement takes place
for problem (2.22), (2.24), (2.25) (see [138]).
2.4. SOME FEATURES OF MODELS I AND II 33
Theorem 2.1 If
@ 2U
2 L2 (QT );
@x2
p @ 2U
T t 2 L2 (QT ):
@t@x
If
2a0
;
C
where C here and below is a positive constant independent of t,
the following estimate takes place
t=2
jjU ( ; t)jj e jjU0 jj :
Another random document with
no related content on Scribd:
Not wanting to put a high poultry netting on my fence to keep my
poultry at home, I devised the method shown, and was thus enabled
to use 24-in. wire instead of 48-in., thereby effecting a saving, as
well as not marring the appearance of the fence. Chickens will not
attempt to fly over this wire but once or twice, as they endeavor to
alight on the pickets of the fence instead of the wire, and
consequently are thrown back by striking the wire. The wire should
be supported at each post, as shown in the sketch.—Roy H. Poston,
Flat River, Mo.
A Fort Built of Snow Blocks Made in Box Form
The Making of the Snow-Block Fort Is Nearly as Much Fun as Daring the
Enemy to Come and Capture It or Capturing the Enemy’s Barricade
What boy does not enjoy a regular snowball fight—a battle, in fact
—especially when he has a fort of his own building to defend, and
his friends have one which he wants to capture? A way to build a
snow fort that will not crumble easily when the enemy charges it is to
pack the snow into a box form and build the structure of the snow
blocks, as shown in the photograph. Portholes can be provided and
when the snow melts slightly, and freezes again, a very solid wall
results. The form for the block, as detailed in the sketch, is made of a
box, arranged so that it is hinged at one corner and has a catch at
the corner diagonally opposite. The form is set on a wooden board
and when the snow is packed in solidly, the catch is opened and the
form lifted away from the block. It is then carried into position and
another block made on the same board. Several boards are handy in
making the block quickly.—H. S., Chicago, Ill.
¶In applying varnish over an old wood surface, or on new wood that
has become soiled, care should be taken to remove all traces of oil
or grease with benzine.
Model Steam Tractor Made by Boy
This Model Steam Tractor Is 28 Inches Long and Drives Itself, Wood and Coal
being Used as Fuel
This Toboggan with Stout Runners Made from Buggy Shafts can be Used
like One of the Ordinary Type, but Works Best on Frozen Snow
When the Front Wheel of One of the Bicycles Broke Down, They were
Rigged into a Tandem, and a Distance of Five Miles to Town was Covered
On a bicycle trip, when about five miles from town, the front axle of
my partner’s bicycle broke, and as neither of us wanted to walk, we
had to find some way of getting us both home. We removed the front
wheel from one of the bicycles, and set the front fork on the rear axle
of the undamaged machine, as shown in the sketch. The axle was
just long enough so that the sides of the fork could be bolted
securely. By careful riding we made the distance to town with only
the slight inconvenience of my partner having to carry the extra
wheel on his back—J. F. G., Mountain Grove, Mo.
¶Do not touch the inner surface of a tent or shelter in the rain, as the
water will drip from the spot touched.
Test Exposures for Bromide Enlargements
To avoid expensive waste of paper in making large bromide
enlargements the following device is useful: Cut an oblong sheet of
stiff cardboard, 5¹⁄₄ by 4 in., with a central opening, measuring 3³⁄₄ by
1¹⁄₂ in. Along the lower edge of the central opening mark off five
spaces of ³⁄₄ in. each, and number them from 1 to 5. Glue a strip of
cardboard, 1 in. wide, at the top and the bottom of the oblong, on the
back. Now glue a second strip of cardboard over each of those just
attached, making the second strips slightly wider to form grooves
similar to those in a sliding-top box.
Cut a piece of cardboard, wide enough to slide easily in these
grooves, and about 7 in. long. In the center of it, cut an opening, 1¹⁄₂
in. long and ³⁄₄ in. wide, with the longer dimension extending across
the strip. By inserting this slide into the grooves, the opening in it
may be shifted to correspond to any of the five ³⁄₄-in. indications at
the lower edge of the opening in the oblong frame.
Focus the enlargement on the easel and put the cap on the lens.
Then pin a 1¹⁄₂-in. strip of bromide paper where a good range of
detail will fall upon it. Pin the oblong card over the bromide strip and
make five successive exposures, of increasing duration, by sliding
the strip so that its opening is opposite the number indications on the
oblong. Note the particulars of exposure at the time and later
develop the strip of bromide paper. Suppose the exposures were of
8, 16, 24, 32, and 40 seconds’ duration, and upon development it
was found that the first and second exposures were much
underexposed and that the fifth was overexposed; the third and
fourth being only slightly under and overexposed, respectively. Then
the mean of these two, that is 24 plus 32 divided by 2, or 28
seconds, will be approximately the correct exposure. The use of this
device will effect a considerable saving.—H. J. Gray, Lewes, Sussex,
England.
Scale on Vaulting Pole Indicates Points of Grip
Indications on the Vaulting Pole Aid the Athlete in Finding the
Proper Grip
The Current of the Canal Causes the Wheel to Revolve, Carrying the Water
to the Trough
Denting of Tire Pumps is Prevented by the Use of a Case Like That Shown
The brass cylinders of many tire pumps are made of thin metal,
and a slight blow may dent them. Such injuries interfere with the
action of the plunger, making the pump almost useless. A case which
will afford protection to the pump is shown in the sketch. It was made
of two strips of wood cored out on the inner sides to fit the pump and
provided with small hinges and hooks, to lock it into place.
A Feather as an X-Ray Lens
The outline of the bones in the hand may be seen by holding it
before a strong light and looking through a lens made of a piece cut
from the tip of a feather. The sketch shows such a lens, which is
mounted in a small piece of cardboard that may be carried in the
pocket, and which affords diversion wherever exhibited.
The device is made as follows: Procure a soft white feather from a
fowl, and cut off the tip as indicated in the upper sketch. Cut a piece
of cardboard, about 2 in. wide and 6 in. long. Fold it to form a folder,
3 in. long, and cut a hole, ¹⁄₄ in. in diameter, through both parts of it.
Glue the tip of the feather between the folder at the hole, taking care
that the fibers lie flat and that no rib is exposed. Glue the edges of
the folder, and the device is ready for use. Hold the hand, or other
object, to be examined about 15 in. from the lens, and place the
latter close to the eye. Looking through the hole toward an electric,
or other strong, light, it will be observed that the bones of the hand,
for example, may be seen clearly. They will be outlined with a band
of varicolored light. A frosted globe is better than one which exposes
a filament.—Leander L. Droesch, Chickasaw, Ohio.
The Small Pipe Fixed to the Water-Jacket Outlet Siphons the Bilge Water
from the Bottom of the Motorboat