Rodolfo Amedeo Lanciani.

The American journal of science and arts online

. (page 76 of 109)
Online LibraryRodolfo Amedeo LancianiThe American journal of science and arts → online text (page 76 of 109)
Font size
QR-code for this ebook

tion on the line between the helices that it stood at 0° when the
circuit was open or closed, and a current passed from 12 cells
of a Gaiffe bichromate battery, which was rapidly decreasing
in electro-motive force, the needle remained at for over 20
minutes without change or tremble.

Now if we can find the law which shows the variation of the
intensity of action of the electro-magnet on the needle with a
change of its distance from the magnetized core, we will have a
system of measurement for the electro-magnetic forces which
will exceed in accuracy and in delicacy, though be similar in
arrangement, to the best photometric processes.

The advantages of this method, as mw as my knowledge of it
extends, is that you can thus subject the needle to two opposing
actions of great intensity and thus any minute difference in the
causes which alter their relative intensities will, on account of
the nearness of the needle to the helices, give a deflection which
would be inappreciable in the methods heretofore used ; and
above all, the same current traversing both helices will, even if
inconstant and fitful, affect both proportionately and the needle,
as seen above, will preserve a fixed position at 0° under these

The law of variation of the intensity of the force with a change of
distance. — ^To find the law of variation of tne force with the dis-
tance from the pole of the core (which as is well known exists
in the plane of the end-coils of the helix) the following experi-
ments were made.

A core 10 in. long and 1'64 in. in diameter composed of 400
soft well annealed iron wires, was placed in helix E and the
plane of the ends of the wires facing the needle were brought
into the plane of the end coils of the helix. The center of the
compass needle was placed at distances of 4, 5, 6, 7 and 8 ft
from the end of the core ; the needle brought to 0** at each of
these positions by the action of terrestrial magnetism and the
current passed from seven new, fi^eshly amalgamated Bunsen-
cells ; as soon as the needle had come to rest the angle of deflec-
tion was read and at the same time the tangent galvanometer
was noted.

Digitized by VjOOQ IC

198 A. M. Mayer — Besearches in Electro-Magnetiam.

The following table gives the results, each nuinber being the
mean of four experiments.

Column (1) gives the distance of the center of needle from
core ; (2) the readings of the tangent galvanometer ; (8) the de-
flection of the compass needle for the distances given in column


Diet from

TaD. GaL

. (3)

Nat Tan. <
Defl. d*




4 ft. or 1

5 '* " 1-26

6 *' " 16

7 " " 1-75

8 " »' 2

51° 45'
51° 45'
51° 30'
61° 46'
51° 30'

38° 15'


14° 30'
9° 46'
6° 44'

-788336 j -788336
-424475 , -829042
-258618 i -872835
-171831 -917888
-118063 -944504



(1) ; the nat tangs, of the angles of (8) ; (5) the products of the
nat tans, of the angles of deflection bv the cubes of the distances
in (1) ; these products being obtained by multiplying the nat
tan. of each an^le of deflection bv the corresponding distance
from the core raised to the cube, snould, if the law followed the
reciprocals of the cubes, be equal ; column (6) ^ves the pro-
ducts of the nat tangs, of the needle deflections mto their cor-
responding distances from the core raised to the 2'7404 power ;
(7) contains the differences of each product of (6) fix>m •Y88836
the nat tan. of 88° 15'. If the experiments had been more ex-
tended these differences would have been smaller ; as it is, they
are not great, and greater approximation cannot be obtained witn
three decimals in the exponent of the power, and this d^ree of
accuracy is all the determinations wOl allow. The sum of the
H- departures equals -008857, that of the — departures equals
•008661, their difference is -f000206.
We can therefore assume that the intensity of the Jbrce (for this

apparatus) varies inversely as the 2'740^ power {or as cP''^^^'^ of
the distance from the core.

Experiments on the comparative strengths of cores formed of in-
sulate and uninsulated wires, — The wire used in these experi-
ments was of the best soft iron and of two diameters, ^^ ana jV
incL The wire was cut into lengths of 10 in. and then placed
in iron tubes, (to keep them fix)m the action of the aiA and
heated to bright redness ; the tubes were then covered thickly
with lime and allowed to cool slowly. This process was twice
repeated on all the wire and iron tubes and cylinders "used in
this investigation, and they were found after magnetization to
have very little residual nu^etism. The caution however
was taken always to present the same end of a core toward the
needle ; thus all the cores were placed in the same circumstances.
Tlaree cores were made of each size wire. Each core of the yV
wire contained 100 pieces, while each core of the jV ''^^ con-
sisted of 400 wirea The wires of one core of each size ime^
(after having been tested, as in Ex. 7, that they were equally

Digitized by VjOOQ IC

A. M, Mayer — Researches in Electro-Magnetism. 199

affected in the same circumstances of magnetization), were insu-
lated by dipping them separately into parafine, melted in a tube
about 12 in. lonff, and were thus uniformly coated with a firm
layer of that substance. They were then closely tied together
into bundles by means of silken thread. The helices, in all the
experiments on the cores, were placed 8 feet apart

To abbreviate the descriptions of the experiments we will always
place the E helix east oi the compass and the W helix west,
and to indicate the deflection of the needle we will use such ex-
pressions as **S. 17® W." which means, S. end of Tieedk deflect-
ed 17° W. or toward helix W. In the case of N. poles facing
the compass the above expression would show that the core of
helix W was stronger than that of helix E. If the current was
reversed it would mdicate that the E helix was the stronger.
When not otherwise stated it wiU be understood that the N.
poles of the helices face the needle, and therefore the expression
given of the deflection will always show which of the cores has
the greater action on the needla To designate the different
wire-cores we will use " 100 or 400 TJC " or " 100 or 400 I C,"
which respectively mean, the uninsulated core of 100 or 400 wires
and the insulated core of 100 or 400 wires. T. G. stands for tan-
gent galvanometer.

Eo^eriment 1. Helices alone, without cores. Needle placed
midway between helices and brought to 0®. Current passed from
lOBunsencelk Needle S. 4° E. T. G. 58^

Bkc 2. Reversed current S. 4° W. T. G. 58^

Ex, 8. Moved needle toward W. helix so that it stood at
0** when circuit was open or closed. Distance from E helix
= 52-77 in. T. G. 58. Expts. 1, 2 and 3 show that E helix is
the stronger.

Ec 4 Three cores of 400 wires each, one of insulated wires,
were compressed to exactly the same diameter of 1*64 ins. An
uninsulated core was placea in each helix, and the needle brought
to 0°, midway between the helicea Current passed S. 27** E.
T. G. 50^

Ex, 5. Current reversed S. 27** W. T. G. 50^

Ex, 6. Other circumstances remaining as in (4) moved needle
toward W helix until it stood at 0** when circuit was open or
closed. Needle distant from E helix 55;6 in. T. G. 46^
Expt repeated many times, with current direct and reversed,
witn the same result •

Ex, 7. Interchanged the cores in the helices. Current passed.
Needle still at 0^ T. G. 48^ Current reversed Needle at 0^
This shows that the cores were equally affected in the same

Ex. 8. Replaced E core by the insulated core of 400 parafined
wires. Needle having remained in same position as in (6) and
(7). Current passed. S. 55' W. T. G. 46^

Digitized by VjOOQ IC

200 A. M, Mayer — Researches in JElectro-JfagneHsm,

Ex. 9. Current reversed S. 56' E. T. G. 46^

Expts. 8 and 9 repeated several times with same results, show-
ing that the insulated core was the weaker.

Mc. 10. Moved needle toward E helix, which contained the
400 insulated wires, until it stood at 0*^ when circuit was open
or closed. In this position needle was 65*4 in. fix>m E helix.

The 100 wire-cores substituted for the 400 wire-cores gave
similar results only differing in degree.

It is very important to remark that errors may be introduced
into such determinations as the above, by not taking care to
have the various cores which we compare of exactly the same
diameter. To show the effect of a change of diameter in the
same bundle of wires the following experiment was made.

Mc 11. The two cores of 400 uninsulated wires of 1*64 in.
diameter were placed in the helices and needle brought so that
it stood at 0^ when circuit was open or closed. The E core was
now so firmly compressed that its diameter was reduced to 1-48,
and replaced in E helix, otherwise the circumstances were the
same. Current passed. S. 1° 15' W., showing a consider-
able diminution in the magnetic force of the compressed core ;
and also, beautifully exhibiting the surface action of magnetic

It will be observed, in the above comparative experiment,
that the two uninsulated wire-cor^ were of exactly the same
strength when equally magnetized, (Ex. 7.) and that one of them
remamed in the W helix as unit of commtrison to the uninsu-
lated and insulated cores placed in the E helix. This method
is necessary when the magnetizing effect of the helices are not
the sama The uninsulated core which remains in the W helix
stands in the place of the " unit-candle " in Bunsen's method of
photometry, and the needle which is moved to such a j>osition
Detween the helix that it remains at 0^ when the circuit is open
or closed is the equivalent of the paper screen with its central
translucent circle. But if it should nappen that when the needle
is placed midway between two similar helices that it remains
at 0° when the circuit is open or closed, then two cores would
answer the purpose.

Ckilcv^lation of the relative forces of the uninsulated and insulated
cores. — ^The data for the calculation are given in Experiments 6
and 10.

W.UC : E.UC : : 40-4»^^ : 55 #^*^ : 1 : 2-89925.
W.UC : E.IC :: 40-6*^ : 55 ^^^ : 1 : 2-84874.

Hence Uninsulated Core \ Insulated ftwe::!-: 0-9768.

The insulated core is therefore -02818 weaker than the unin-

In Expts. 8 and 9 it was seen that the above difference of
•02818 in magnetic force caused a deflection of 55' in the com-

Digitized by VjOOQ IC

A. M. Mayer — Researches in Electro-Magnetism, 201

pass needle, and as 8' or even less of deflection can be accurate-
ly read it follows that a diflFerence of tot^Ii in the magnetic
iorce of the two cores can be detected and measured by this
method. This example will convey an idea of the delicacy and
accuracy of this plan of measurement, which no doubt can be
increased by the careful selection of the proper force of current,
distance of nelices and other related functions of the apparatus
It thus appears that this result sides with the theory of Am-

Jfere, and also, to my mind, points to that postulate of the atomic
pyothesis which states that the atoms of bodies are separated
from each other by distances which are great when compared to
the size of these atoms.

Attempts to determine what thickness of tube infraction ofitsdi-
ameteTj is equal to a solid cylinder of tlie same diameter, — ^According
to the theory of Ampere, the magnetization of a bar of steel, or
of an iron core subjected to a helical current of electricity, is due
to the surface currents, which are the resiiltants of the interac-
tion of all the molecular currents in the interior of the bar. It
is therefore exceedingly interesting in a theoretic point of view
to determine accurately what thickness of tube is equivalent to
a solid cylinder of the same length and diameter. The experi-
ments of Barlow, Harris, de Haldat, and of Du Moncel have
shown the truth of this deduction-
There is evidently a certain thickness required to develope
an equivalent action, but a certain portion of the interior of an
electro-magnetic core can be removed without diminishing its
force when in a condition of ^^ saturation." This determination
will therefore inform us, (accepting the hypothetical theory of
Ampere), what central currents are prevented by their inter-
action from aflfecting the resultant simace currents of the bar ;
and it appears to me that we shoiild express the result in terms
of the central cylinder which can be removed and not of a certain
thickness of tube which remains.

Out of well annealed sofl iron a series of tubes, 10 in. long,
were turned so that they fitted closely into each other ; into the
innermost tube slid a solid iron core of "88 in. diameter. The
diameter of the outer tube was 1-68 in. With this construction
of core, tubes of various thicknesses could be rapidly produced
% and compared in effect Two such cores were constructed.
JEx, 12. 400 uninsulated wire-core in W helix ; in E helix
was placed the combination of tubes and core. Needle brought
to such a position that it stood at 0° when circuit was open or
closed. T. G. 46^

JSx. 18. Removed inner core and tubes, and drilled out the
innermost remaining tube until the thickness of the combina-
tion tube was } of its diameter. Current passed. Needle re-
Am. Joub. Sol— Seoond Sebibs, Vol. L, Na 149.— Sept., 1870.

Digitized by VjOOQ IC

102 A. M. Mayer — Researches in Eleciro-Magnetism,

mained at 0^ On fiirther decreasing the thickness the S. end
of the needle was deflected toward the W helix. T. Or. 46°.

It thus appears that a tube of this size having a thickness of
} of its diameter is, when magnetized with this strength of cur-
rent (8 cells, T. G. 46°) equal in eflfect to a solid cylinder of
the same diameter.

Kt, 14. Another tube was constructed having a diameter of
1*8 in. and a thickness of } the diam. Into tms tube slid a
solid core. Placing the uninsxdated wire-core of 400 in W. he-
lix and tube with inner core in E. helix, the needle was placed
so that it stood at 0° when circuit was open or closed. T. G.

Mc 15. Withdrew inner core. Current passed. S. 40' W.
T. G. 46°. Thus a tube of the above diameter, having a thick-
ness of J of its diameter is not equal to a similar solid cylinder.
This subject will therefore have to be the object of further in-

Thmking that in the case where a current did not saturate
the above tube it would equal the solid cylinder, the following
experiments were mada

JSc 16. The tube used in (14) was placed with inner core
in E. helix, and a similar tube and core in W. helix^ and the
needle placed so that it stood at 0° when circuit was open or
closed. A current from 4 cells was passed. T. G. 82°.

Mc 17. Solid core removed from tube in helix E. Needle
remains at 0°. Hence for this strength of current (T. G. 82°)
the tube is equal to an equivalent solid cylinder.

Ex, 18. Other things remaining as in (17) a current from 6
cells, T. G. 87°, was passed through helices. Needle S. 2' W.

Ec. 19. Current passed from 6 cella T. G. 40^ Needle S.
15' W.

Ex. 20. Current passed fix)m 7 cells. T. G. 42°. Needle
S. 20' W.

Ec 21. Current passed from 8 cells. T. G. 44°. Needle & 40^
W. It is clearly seen from above experiments that with a current
of T. G. 82° the tube is eaual to the similar solid cylinder, but
on gradually increasing tne current up to T. G. 44° that the
tube fialls more and more below the strength of the cylinder,
and the needle is again deflected S. 40' W. toward the cylinder
with that current Therefore in determining what portion of
the interior of a cylinder can be removed so that what remains
shall equal the solid cylinder we should see liiat the tube is
. " saturated," otherwise tne comparison does not solve the prob-
leuL Whether this condition was fulfilled in Expt 18 was not

In Daguin's Traiti de Phystmie, voL iii, p. 615, Paris, 1861,
we read : " M. Du Moncel finds that it is necessary, in order

Digitized by VjOOQ IC

A. M, Mayer — Researches in Electro-Magnetism. 108

that the magnetization [of a tube] may remain the same as a
solid cylinder, that the thickness should, be at least | that of the
radius of the cylinder." I have not seen Du Moncel's research
on this subject and do not know the method he adopted in
comparing the magnetization, but I think that his estimated
fraction of i of the diameter is too low, as Ex. 18 clearly shows.
The cause of this error, I imagine, resides in the fact that he
used a current not sufficient to saturate the tube and cylinder,
and thus his determination would be explained by Expta 15
to 21.

This subject is worthy of careftd research, and its theoretical
bearings on the subject of magnetism are very interestiuj?. My
experiments thus far cause me to doubt if any constant traction
of the diameter will be found, but that this fiction will depend
upon the diameter of the tube ; being larger as the tube is smaller.

Experiments on a Tube slit longitudinally, — To determine the
influence, if any, of destroying the continuity of the exterior
surfiwe of a tuhie by a longitudinal slit, I obtained a tube of weU
annealed iron, 1*68 inch in exterior diameter, 16 in. thick and
10 ins. long, and slit it in its whole length by an opening ^V
inch wide. This opening could be closed at pleasure by plac-
ing in it a tightly ntting piece of iron 10 in. long.

Ex. 22. The above tuoe, with the sUt closed, was placed in
the E. helix and the 400 U. C. in the W. helix, and needle
brought to such a position between them that it stood at 0°
when circuit was open or closed. Eight cells of Bunsen. T. G.

Ex. 28. Slit opened by removing the piece of iron. The
needle remains at 0^ T. Gt. 46°.

Expts. 22 and 28 repeated several times with the same result

In the accompanvinff diagrams is shown, according to the
theory of Ampfere, the o&rection of the circulation of the exterior
and interior surfSswe-currents. It is seen that in both cases the
interior surface current circulates in an opposite direction to the
exterior current ; and that when the tube is slit longitudinally
the circulation is not cut off but Jacilitated by the joining of the
exterior and interior surfaces. It is therefore natural to suppose
that on account of this facility afforded to the circulation that
a quicker magnetization and demagnetization will occur in the
split than in the closed tube ; but no difference is found in the
magnetic force of the two tubes, for the loss in surfiace by the
slit seems made up in the greater facility of circxdation.

I would therefore suggest that the cores of the electro-mamets
of astronomical chronographs and of telegraph instruments be of
soft well annealed Norway iron, made into tubes having a thick-
ness of i of their diameters, and sHt longitudinally by a narrow

Digitized by VjOOQ IC

204 A. M. Mayer — Researches in Electro-Magnetism.

On the Tnagnetic condition of the interior surfojce of a tube. —
Referring to fig. 2 it will be seen liiat the Amperean currents on
the interior surfSace of a tube flow in a direction contrary to
those on the exterior. The magnetic condition of this sumce
was therefore an extremely interesting point for investigation ;
it has been the subject of over two hundred experiments,
and is by far the most difficult research mentioned in this

In my first experiments, steel rods and iron cores were intro-
duced into soft ux)n tubes which were then placed in helix R
The current was then passed and the tube with enclosed core
very slowly withdrawn. The steel or iron core was then re-
moved fix)m the tube and tested by placing it to the W. of the
center of a compass needle. The magnetic force of the core
was then compared with what it had before it was introduced
into the tube. This method was thoroughly tested by withdraw-
ing tube and core both before and after contact was broken in
the helix, but no decisive conclusions could be drawn because
the effect of the inductive action which took place when the
core was withdrawn from the tube changed the magnetic rela-
tions which existed while the core was in the tuba

The following method was then devised. Two tubes of very
soft iron were constructed, having both an internal diameter of
'84 inch, and one an external diam. of 108 inch, the other of
1*81 inch. Both tubes were 9 in. long. One hundred and
twenty well annealed iron-wires a little less than 9 in. long
were tightly bound together by silken cord and then wrapped in
six thicknesses of writing paper, so that the wire core fitted
neatly the interior of the tubes.

The wire core was now deprived as far as possible of magnet-
ism by passing a weak magnet over it, and could thus be so
reduced that it caused a deflection of only 1** when placed 7*6
in. W. of the center of a surveyor's compass needla By now

Digitized by VjOOQ IC

A, M. Mayer — Researches in Ekctro-Magv^etiem. 205

inverting the core in the line of the dipping-needle this small
residual magnetism could be reduced to 10' or 15' deflection,
and in some experiments even a reversed magnetization was
thus produced. The core was now handled very carefully to

Erevent all vibrations or blows, and placed in an E. and W.
orizontal Una

The tube was placed in helix E and the current passed. It
was then very slowly and steadily withdrawn, so as gradually to
let down its magnetism, the tulJe removed several feet from the
helix and the current broken. This operation was repeated
until the tube had attained as high a degree of magnetization as
could be given to it with the same constant current in the he-
lix, and a farther repetition of the process caused no increased
deflection in the needle. The tube was then careftdly placed
in an E. and W. horizontal line and the core slowly introduced,
and so that the S. pole of the core was in the S. ena of the tuba
It is important in all these manipulations to keep the core and
tube in a horizontal plane, for if they are inclined to the horizon,
during the experiment, the magnetic condition is changed and
the operations nave to begin anew. The degree of susceptibil-
ity of very soil iron to a change of inclination can only be ap-
preciated by those who make the triaL

The tube with enclosed wire-core was now introduced into
the helix and the current passed ; then withdrawn, and this re-
peated (as in the above process on tubes alone), until the needle-
deflection showed no increase of magnetic force. This deflec-
tion was then compared with that previously produced by the
tube acting alona

Now if the interior electric currents really flow as Ampere's
theory states, then when the core is in the tube and the maxi-
mum magnetization reached, the combined effect of tube and
enclosed core on the needle should be less that given by the
tube alone without the central core ; for, in the former case the
core has a magnetism given it the reverse of the tube; i a, sup-
posing marked end of tube S and marked end of introduced
core S, then, after magnetization in the helix, the marked end
of core is N and will therefore neutralize a portion of the tubers
action on the needle. Now this was found to be invariably the
case in all the experiments made ; and how can it otherwise be
explained. It is Known that the combined force of two mag-
nets with like poles placed together is less than the sum of the
forces of the separate bars separately measured ; but the force
of the two together is always greater than that of one alona
But we here have a tube and enclosed insulated core with like
poles together giving, after magnetization (in the same conditions
exactly as that given to the tube alone) a less effect than that
given by the tube alona Therefore these fects show, to my

Digitized by VjOOQ IC

106 A. M Mayer — Besearches in Mleclro-Magnetiam,

mind, that the deduction fix)m Ampere's theory is verified by
experiment ; and that we have succeeded in giving to a core in
the interior of a tube a polarity the reverse of that given to the
tube itself

We here present the following experiments, which are typical
of all made ; the others only diflfering in q^uantity with a cnange
of distance from the needle and a change m the intensity of me
current in the helix.

The marked ends of tubes were S. poles, and S. 10° E.
means that when tube or core or both were placed W. of the

Online LibraryRodolfo Amedeo LancianiThe American journal of science and arts → online text (page 76 of 109)