John Elihu Hall.

The American journal of science online

. (page 6 of 61)
Online LibraryJohn Elihu HallThe American journal of science → online text (page 6 of 61)
Font size
QR-code for this ebook


in contact, or nearly in contact, with the bed-rock: (4) the
origin of local tensile stresses within this bed-rock, near its sur-
face, by virtue of frictional resistance between it and the frag-
ments ; (5) a gentle slope of the overridden surface toward the
direction from which the ice is coming ; and (6) the position
of this surface near the crest of a hill, that is, between slopes
opposed to, and dipping with, the ice motion, and conse-
quently where there should be in process material changes in
tne relations of the interacting forces.

Harvard University, Cambridge, Mass., Oct. 24, 1011.

♦Gilbert, G. K., op. cit., p. 804.

f The writer finds no mention, in the literature, of the exact relation
between the topography and other occnrrences of the crescentic fractures.
The conclusion just presented is drawn from the Nor they Hill locality only.



Digitized by VjOOQ IC



Mixter — Seat of Formation of Titanium Dioxide. 45



Akt. VII. — The Heat of Formation of Titanium Dioxide /
by W. G. MixTEB.

[ContribntionB from the Sheffield Chemical Laboratory of Tale University.]

In 1909 the writer determined the heat of formation of tita-
nium dioxide by the sodium-peroxide method. The value
found at that time was 215,600°.* In an article f which
appeared shortly after the writer's, Weiss and Kaiser gave the
results obtained by them by burning titanium in oxygen. The
average of their experiments is 97,772'^ for the heat of formar
tion of titanium dioxide. It will be seen that the value found
by them is less than 50 per cent of the value found in this
laboratory. This great difference led the writer to attempt the
burning of titanium in oxygen.

A preliminary test showed that a platinum tray or cup would
not answer for holding the metal in the bomb since the heat of
oose only between Na, Al,Si,0,,,
NaAlSijO, and NaAlSi,Oe, of which the most likely is, as we
have seen, albite. It is probable that there are natural nephe-
lites whose composition would lie above the soda-nephelite,
orthoclase line and for which albite would again be the only
choice.:]:

♦ This Journal, xxxi, 25, 1911. f Jo^r- Wash. Acad. Science, Sept. 19, 1911.

t Nephelite from Denise, Hante Loire, France (anal. 7, Dana's System),
has a composition which would place it barely above the orthoclase, soda-
nephelite line. The low summation in this analysis makes it probable,
however, that no great significance can be attached to this fact.



Digitized by VjOOQ IC



N. Z. Bowen — The CompoHtion of NepheliU, 53

For a great many nephelites, then, mere expression of com-
position in terms of three components may be accomplished in a
number of different ways, but for a systematic method which
we may hope to apply to all possible nephelites we must use
the three molecules JNaAlSiO,, KAlSiO,, and NaAlSi,0„ that
Schaller has suggested. The artificial nephelite A enables one
to come to this decision.

The ability to come to this definite decision also makes it
possible to predict when nephelite may be expected to be

* saturated' with silica. The presence* of orthoclase in a
rock may be considered as having a * silicifying ' effect on
nephelite, but this effect would be limited, for trie resultant
composition could never pass above the orthoclase-soda-nephe-
lite line. The artificial nephelite A shows that the composi-
tion may lie above that line. It is, therefore, only in the
presence of albite itself that nephelites may be expected to be

* saturated ' with silica.

It would appear that the presence of jadeite should bring
about a like result, but this possibility may be disregarded in
nature. The presence of the jadeite molecule in dilute solu-
tion in a hornblende or pyroxene would not have a comparable
effect.

Indeed this remark applies equally well to albite itself. If
the albite molecule were present as oligoclase or andesine it
would, theoretically, not have an equal ' silicifying ' effect on
nephelite, for a nephelite saturated with albite would be in
equilibrium only with a plagioclase saturated with albite and
the only plagioclase saturated with albite is albite itself. On
the other hand, a nephelite found associated with anorthoclase
saturated with albite should itself be saturated with albite.
The conditions necessary for the satumtion of nephelite with
albite are so unlikely to 'occur that it may be safely said that
natural nephelites are probably never saturated.

Nummary,

1. A diagram is given which shows that the composition
of nephelites cannot be explained by assuming mutual solid
solution of any two components. It is necessary to imagine
solid solution among three components.

2. Mere expression of the composition of nephelites in
terras of three components may be accomplished in a number
of different ways.

3. From the results of synthetic work it becomes possible
to decide upon one of these methods as the only one applica-

•By ** presence" it ia necessary to understand intimate association daring
the process of crystallization, not mere proximity.



Digitized by VjOOQ IC



54 A, F. Rogers — BaddeUyite from Montana.

ble to all poseible nephelites. Tha decision falls upon the
three components NaAlSiO,, KAlSiO,, and NaAJSi.Og, sug-
gested by Schaller.

4. Conclusions are drawn as to when nephelite maj be
expected to be saturated with silica.

Mineralogical Laboratory,

Massachasetts Institute of Technology, Boston.



Abt.IX. — Baddeleyite from Montana; by Austin F. Booers.

On some corundum specimens from Montana, obtained from
Ward's Natural Science Establishment, the writer noticed a
black submetallic mineral. The nature of the mineral was
not evident at sight, but it was soon identified by physical and
chemical tests as baddeleyite or native zirconia, ZrO,.

Baddeleyite is a very rare mineral, first described from the
ffem-bearing gravels of Ceylon by Fletcher.* It had also been
found in decomposed jacupirangite (magnetite-pyroxenite) at
Jacupiranga, Sao Paulo, Brazil, and was called brazilite by
Hussak,t who later:|: withdrew his name in favor of baddeley-
ite. Hus8ak§ has also described baddeleyite from Alno,
Sweden, where it occurs in magnetite-olivine segregations in
nephelite-sjrenite.

The specimens obtained of Ward's were from Montana, but
the exact locality was not stated. Similar specimens contain-
ing baddeleyite purchased from Mr. R. M. Wilke of Palo Alto,
CaJifomia, are from the property of the Bozeman Corundum
Company, which according to Pratt] is fourteen miles south-
west of JBozeman, Montana.

Occurrence. — The baddeleyite is an accessory constituent of
a gneissoid corundum-syenite containing microcline-microper-
thite, biotite, and corundum, with subordinate amounts of mus-
covite, sillimanite, and zircon. It occurs in minute crystals
and rounded blebs with a maximum size of about 3""". The
baddeleyite is found in both the feldspar and the corundum,
but is especially abundant on the surface of the corundum, and
often adheres to the feldspar when the corundum crystals are
broken out of the matrix.

* Mineralogical Magazine, vol. z, p. 148, 1898.
f Neues Jahrb. Min., 1892, vol. ii, p. 141.
tMin. Petr. Mitth., vol. xiv, p. 395, 1895.
§ Neues Jahrb. Min., 1898, vol. ii, p. 228.
I BuU. 269, U. S. G. S., p. 133, 1906.



Digitized by VjOOQ IC



A. F. Rogers — Badddeyite from Montana. 56

Crystal Form. — The baddeleyite crystals are square-pris-
matic or tabular-prismatic in habit with only one well-dehned
zone with the forms |100(, |110|, and |010{. Eight cn^stals
were measured on the reflection goniometer with indifferent
results, the angle for (100 : 110) varying from about 44° to 46°.
Finally, one crystal was found which gave the following meas-
urements :









Average


Calc.


100


: no


44** 22', 44° 25', 44° 26'


44° 24'


44° 20'


no


: 010


45° 31', 45° 36', 46° 38'


45° 35'


45° 40'


100


:010


89° 57', 90° 0', 90° O'


89° 69'


90° 0'



The calculated angles established by Blake and Smith* are
given in the last column.

For the angle (100 : 110) Fletcher gives 44° 0' and Hussak,
44° 17i'. The terminal faces are obscure and rounded, but on
one crystal faces of the form |Ill} were identified by the fol-
lowing measurements, (010 : III) = 63° 15' (average of 62° 30'
and 64° 0'), the calculated value according to Blake and Smith
being 64° 18'.

Physical Testa. — The cleavage seems to be in four direc-
tions, parallel to {100|, jllOf, and |010^. The hardness is
about 6i, for it scratches glass faintly and in turn is scratched
by quartz. The luster is submetallic or metallic adamantine.
The color is black, but in fragments it is translucent brownish
red. The double refraction is strong, and the index of refrac-
tion is greater than that of methylene iodide. Some fragments
show several sets of twinning lamellae crossing each other. In
thin-sections the baddeleyite appiears as elongate crystals with
almost parallel extinction and slight pleochroism, the greatest
absorption being parallel to the length of the crystal.

Pyrognostic Tests, — The mineral is practically infusible
when heated before the blowpipe, but examination with a
high-power microscope shows a slight sprouting on the edges.
In the oxygen-gas blowpipef it is fairly easfly fusible to a
black globule. In the closed tube it is unaltered. When
heated in the platinum forceps it glows with a brilliant light,
which recalls the fact that zirconia is the principal constituent
of the filament of the Nernst lamp.

Chemical Tests. — In fine powder the baddeleyite is insoluble
in a sodium metaphosphate bead but is soluble in a borax bead
giving a faint iron test. It is insoluble in aqua regia but is
decomposed by strong sulphuric acid. The hydrochloric acid
solution of the soda fusion turns turmeric paper orange, and
since no green flame was obtained with boracic acid flux this

♦ Mineralogical Magazine, vol. xiv, p. 378, 1907.

tSee Lnquer, Sch. Mines Quart., vol. xxjx, p. 179, 1908.



Digitized by VjOOQ IC



56 A. F. Rogers — Badddeyite from Moiitana.

test is proof that zirconium is present. The acid solution of
the soda fusion with ammonium hydroxide gave a white floc-
culent precipitate which is insoluble in KOH, but soluble in
oxalic acid. The sulphuric acid solution with ammonium
hydroxide gave a precipitate which glows when heated in
platinum forceps before the blowpipe.

Badddeyite ds a Rochformvag Mineral. — Baddeleyite is
one of the rare accessory minerals of igneous rocks. As far as
known, it is confined to rocks low in silica, the jacupirangite of
Brazil having only 88 per cent SiO„ and the present corundum-
syenite about 44 per cent. The deficiency in silica is proba-
bly the reason baddeleyite forms rather than zircon.

Identity with Baddeleyite, — From the presence of zirconium
and the crystal habit one might call the mineral zircon, badde-
leyite, or one of the zirconium-pyroxenes, Iftvenite, wohlerite,
orhiortdahlite. It is distinguished from zircon by luster, hard-
ness, and cleavage ; from altered zircon by the absence of water,
and from the three zirconium pyroxenes mentioned by the
solubility and fusibility. The angles in the prism zone
[100:110:010] agree better with the baddeleyite angles than
with the angles of the other minerals. The sum total of the
characters undoubtedly identities the mineral as baddeleyite.

Stanford nniveraity, California, September, 1911.



Digitized by VjOOQ IC



Che^niatry and I^hyncs, 67



SCIENTIFIC INTELLIGENCE.



I. Chemistry and Physios.

1. Uranium Hexafluoride, — Since no hexavalent halogen com-
pound of uranium has been previously prepared, the preparation
of the fluoride XJF, by Rupp and Heinzblmann is of considerable
interest. It appears that a supposed previous preparation of this
compound by Ditte was not founded on fact. Three methods
were worked out for the preparation of the compound : 1. The
action of fluorine upon uranium pentachloride ; 2. The action of
anhydrous hydrofluoric acid upon uranium pentachloride ; 3. The
action of fluorine upon metallic uranium or uranium carbide.
According to the first two methods the product consisted of a
mixture of two fluorides, according to the equations

2UCK-f 5F, = UF,4-UF,-f 5C1, and
2U CI. + 5H,F, = UF, + UF, + lOHCl.

The higher fluoride, being very volatile, is easily separated from
the other one, but when prepared according to the second method
it was separated only with considerable difficulty from the excess
of anhydrous hydrofluoric acid. The action of pure fluorine
upon metallic uranium gave, as Moissan had shown, a product
consisting chiefly of the tetrafluoride, but it was found that when
a little chlorine was mixed with the fluorine, complete conversion
into hexafluoride took place. The chlorine appears to act as a
catalytic agent in this case.

Uranium hexafluoride is a very volatile, pale yellowish, crystal-
line solid which boils at 65° C. at atmospheric pressure. Its
melting-point, 69*6° C, lies above its boiling-point, so that a pres-
sure of about two atmospheres is required to melt it. The sub-
stance is extremely reactive ; it is very sensitive toward moisture;
it reacts with hydrogen even when cold ; the chief product of its
reduction appears to be uranium tetrafluoride. Its reaction with
sulphur is particularly interesting, as it seems to give a new gase-
ous fluoride of sulphur.— Ze/^cAr. ayiorgan, Chem.y Ixxii, 63.

H. L. w.

2. The Atomic Weight of JSxtra- terrestrial Iron, — Since, as
far as is known, atomic weight determinations have always in the
past been carried out with material of terrestrial origin, it seemed
worth while to Baxter and Thoryaldson to make such a deter-
mination on an element of meteoric origin. Iron was selected as
the subject of investigation, and for the purpose a piece of a 63 lb.
meteoric iron found in 1903 near Cumpas, Mexico, was used.
After an elaborate series of purification operations, ferrous bro-
mide was prepared and analyzed according to a method that had
been previously employed by the authors in connection with ter-
restrial iron. The results indicated no appreciable difference



Digitized by VjOOQ IC



58 Seientijic Intelligence,

between the atomic weights of the two kinds of iron, for the
average of eight determinations on the meteoric material was
55-832, while the outcome of the results of the other work was
65*838. — Jour, Amer, Chem. Soc, xxxiii, No. 3. h. l. w.

3. The Compounds of Ammonia aiid Water, — The question of
the existence of the compound NH^OH in solutions of NH, in
water ^as attracted much attention in the past, but while the
theory of the existence of the hydroxide has been a plausible one,
it has been opposed by several investigators. Smits and Postma
have now shown from freezing-point determinations of mixtures
of NH,and H,0 that the compounds NH,.H,0 and 2NH,.H,0,
corresponding to NH^OH and {NHJ,0, certainly exist, and that
these are the only compounds formed. They melt at about
— 77° C. and — 78° C. respectively, or a littte below the melting
point of pure NH„ while there are eutectics between the two
compounds and between each of them and a constituent. The
existence of these compound?, although it gives no proof of their
chemical constitution, undoubtedly strengthens the hydroxy! and
oxide theory, for it shows that NH^ can be regarded, as combin-
ing with water in the same way that an alkali metal does, for
example, to produce KOH and K^O. It may be added that
Ruppert has obtained the same results by freezing-point deter-
minations as those that have been mentioned, but published them
somewhat later. — Zeitschr. anorgan, Chem,y Ixxi, 250. h. l. w.

4. The Quantitative Determination of Fluorine as Calcium
Fluoride. — The usual method for carrying out this determina-
tion, following Berzelius, is to precipitate the calcium fluoride
together with calcium carbonate, and, after filtering and igniting
the mixture, to dissolve out the calcium carbonate with dilute
acetic acid and finally collect and weigh the fluoride. Stabck
and Thobin have made a modification of this method, which con-
sists in precipitating calcium fluoride in the presence of an exactly
known amount of an oxalate in a solution containing about two
cubic centimeters of free acetic acid. The amount of oxalate
should be about the same as that of the fluoride. The precipita-
tion is made in the hot solution by the addition of calcium chlo-
ride solution. The mixture, which filters as well as calcium
oxalate alone, is collected on asbestos, dried at 210° C, and
weighed. By deducting the known amount of calcium oxalate,
the \veiffht of fluoride is found. From the results obtained by
the authors it appears that the method gives very satisfactory
results, and it is evidently a great improvement upon the old
method. — Zeitschr, analyt, Chem,y li, 14. h. l. w.

5. The Volumetric Determination of Antimony in Alloys. —
Dr. G. S. Jamibson of the Sheflield Laboratory has applied
Andrews' iodate titration to this determination. T*he alloy, such
as hard lead or solder, is dissolved in concentrated sulphuric acid
by heating to boiling, the residue is treated with a mixture of
equal volumes of strong hydrochloric acid and water, the lead
sulphate is filtered off and washed with hydrochloric of the same



Digitized by VjOOQ IC



Chemistry and Physics. 59

strength, and the filtrate is titrated directly in a bottle in the
presence of a little chloroform with potassium iodate solution
until the color of iodine, which is produced in the chloroform, at
last disappears. The reaction in the presence of the strong
hydrochloric acid proceeds according to the equation

2SbCl,+KIO, + 6HCl = 2SbCl. + KCl + ICU3H,0. .
Since tin, copper, and iron do not interfere with the method, it
is convenient in many cases, and it appears to give very accurate
'results. — Jour. Indust. and Eng. Chem.^ iii. No. 4. h. l. w.

6. J%e Radio-actimty of the DiXrkheim Mineral Waters. —
Since it had been found by Ebler in an examination of the " Max-
quelle " water of Ddrkheim that a certain amount of radio-activ-
ity appeared to follow the alkali-metals, and since on this account
it was suspected that a sixth alkali metal with radio-active prop-
erties might exist in this water, Ebler and Fellneb, in connec-
tion with an elaborate investigation of the radio activity of these
waters and their products, have made a further study of the
question of the presence of a new alkali metal. The result, how-
ever, is disappointing from a chemical point of view, for they
found that potassium salts prepared from the water of this spring
showed no more radio-activity than ordinary potassium salts. ^
Zeitschr, anorg. Chem.^ Ixxii, 233. . h. i*. w.

7. Bidletin of the Bureau of Standards.— The third number
of volume VII, recently issued, opens with a paper by L. W.
Austin, giving in detail the results of Some quantitative experi-
ments in long-distance radiotelegraphy. These tests were carried
on two years since under the auspices of the Navy Department
between two cruisers and the wireless station at Brant Rock.
Measurements were made up to a distance of one thousand miles,
their chief object being the determination of the law governing
variations of strength of signal with the distance. The results
arrived at, briefly stated, are as follows :

" (a) Over salt water the electrical waves decrease in intensity
in proportion to the distance as found by Duddell and Taylor. In
addition they are subject to an absorption which varies with the
wave length and which may be expressed mathematically by the

term e . The complete expression for the received current

is then

This is true in general for day transmission. The absorption at
night is entirely irregular, varying from zero to the day value,
but is on an average much less during the winter than in summer.
Variations also appear to occur during the daytime, but these are
probablyin general small.

" {h) The received antenna currents between two stations with
salt water between are proportional to the product of the heights
of the sending and receiving antennas and inversely proportional



Digitized by VjOOQ IC



60 Scientific Intelligence.

to the wave length, provided the antenna resistances remain con-
stant. These experiments were carried on with flat-top antenna
heights of from 30 to 80 feet and wave lengths from approxi-
mately 1500 to 4000 meters. ^

'' (c) Taking account of the influence of antenna height and
wave lengthy the above equation ma}' be extended and a general
day transmission formula written as follows :

0-0015 d

where the currents are given in amperes and all lengths in kilo-
meters. This formula has been tested for sending currents from
7 to 30 amperes, antenna heights 37 to 130 feet, wave lengths
from 300 to 3750 meters, and distances up to 1000 nautipal miles.

" In regard to the value of the day absorption it is only possi-
ble to say that the above expression is satisfied within the limits
of error of the observations. It is quite possible that when
observations are made at distances of 2000 to 3000 miles, the
value of the absorption coeflicient will have to be corrected by
10 or even 20 per cent, as this amount of error could exist with-
out discovery at the distances covered in these experiments. It
is also possible that the square-root law relating the absorption
with the wave length is only an approximation."

The other papers in the same number include one by C. E.
Waters on the behavior of high-boiling mineral oils on heating in
the air ; another by J. R. Cain on determination of vanadium in
vanadium and chrome-vanadium steels; by £. Buckingham and
J. H. Dellinger on the constant C, of Planck's equation for the
intensity of radiation ; by T. T. Fitch and C. J. Huber on Ameri-
can voltmeters and ammeters ; by P. 6. Agnew on the current
transformer ; and by H. S. Carhart on thermodynamics of con-
centration cells.

CiRCULAK No. 24 (pp. 41) gives a summary of the published
work of the Bureau of Standards, which is remarkable in extent,
considering the short time since the Bureau was established.
Seven volumes of the Bulletin have appeared, the first in 1904-6 ;
in all 174 papers are enumerated, some of these now in press ;
these are here classified and their contents briefly indicated. A
series of thirty-one circulars of the Bureau is also enumerated ;
these give the regulations under which the tests and comparisons
of the Bureau are made, with a general statement as to its work.

8. Annual Tables of Constanta arid Numerical JData^ Chemi-
caly Physical aiid TechnologicaL — The first volume of the Annual
Tables of Constants and Numerical Data, Chemical, Physical
and Technological, compiled and published by an international
commission, appointed by the 7th International Congress of
Applied Chemistry (see Science, August 4, 1911, p. 158), is now
open to subscription. Subscription blanks, the terms of sub-
scription and descriptive leaflets maj' be obtained from any one



Digitized by VjOOQ IC



Chemistry and Physics. 61

of the three American Commissioners : Dr. G. N. Lewis, the
Massachusetts Institute of Technology, Boston, Mass. ; Professor
6. F. Hull, Dartmouth College, Hanover, N. H., and Professor
J. Stieglitz, the University of Chicago, Chicago, III. After Jan.
15th, 1912 the price of the volume is likely to be increased.

9. The Sun; by Chables G. Abbot. !Pp. xv, 448 ; 26 plates,
72 figures. New York, 1911 (D. Appleton & Company). — A nota-
ble work which every student of physics and of astronomy will
find of absorbing interest. It is the first considerable book bear-
ing its title since that of Professor C. A. Young, so usefully and
widely known, published just thirty years earlier ; and it is pleas-
ing to find that its frontispiece, like that of its predecessor, is a
reproduction of S. P. Langley's famous drawing of a typical sun-
spot, unfortunately somewhat marred, however, by a partial fill-
ing of the nucleus of the spot. This is to be regretted because too
many contemporary readers will get an inadequate impression of
the merit of the wonderful original drawing.

It is impossible in a brief review to do more than indicate in
the most cursory manner some of the important advances in astro-
physics which are here admirably set forth. The first five chap-
ters cover much the same ground as Young's treatise, with
amplified and modernized discussion of methods for determining
the radiation of the sun, together with excellent reproductions of



Online LibraryJohn Elihu HallThe American journal of science → online text (page 6 of 61)