John Almon.

The American journal of science and arts online

. (page 61 of 102)
Online LibraryJohn AlmonThe American journal of science and arts → online text (page 61 of 102)
Font size
QR-code for this ebook

associated ; they occur not only as more or less well defined de-
posits of varying extent, but they must originally have pervaded
every sedimentary rock from the lowest Silurian to the latest ma-
rine formation. Subsequently they have found access to many
formations of a later date and different origin by the ascent of
percolating waters. Most of these rocks contain, more or less
abundantly, compounds of magnesia, particularly carbonates,*
and have thus uninterruptedly yielded Quantities of that sub-
stance to solutions, wherever such poweriul disintegrating agen-
cies as change of temperature, access of moiisture and free car*
bonic acid have been at work ; the destruction of these rocks is
a mere matter of time. Many sedimentary rocks may have been
exposed to these influences temporarily, perhaps in a less indu-
rated state, and consequently have presented conditions most
&vorable for transformation and subsequent extraction ; circum-
stances which may sometimes account for the total or partial ab-
straction of sulpfiate of lime firom a number of sedimentary
rocks of marine origin.f

Belying here, more on the duration than on the intensity of the
action, I believe that the final results of reactions like those con-
sidered in the foregoing discussion, must in common with other
causes^ have exerted an important bearing on the composition of
the marine waters, during the development of our present sur-
&ce-configuration ; and they may in particular instances have as-
sumed proportions sufficiently extensive to account even for more
localized accumulations of sulphate of soda, in connection with

* ReseATcfaes of Ebelman, confirmed bj T. S. HoDt, have prored that magDena
and lime are first abstracted, bj means of carbonic acid, from eren basaltic and sim-
ilar rocks and that in the case of labradorite, the rsmoral of these two alkaline
earths was even more complete than that of the alkalies, (this Journal, II, tttit, 181,
March, 1866); a series of anises of waters from along the N. T. Central railroad
b^ Dr. 0. F. Chandler (New Tork, 1866), as well as some of my own analyses fur-
nish a direct proof of the fact, that the carbonate of magnesia enters in an unusual
proportion into the composition of a large number of our spring waters.

f Mineral waters issuing from rocks In the city of Oswego (gray sandstone) along
the Oswego rirer contain large percentages of chlorid of sodium, besides the chlo-
ride of magnesium and calcium ; they are free from sulphate of lime. AnalysiB
gires, ID 1000 parts, NaCl 6*622, MgOl 0*140, CaCi 0*8162, etc.

Digitized by


C. A. Ooessmann on the Chemistry of Brines. 87

chlorid of sodium as well as with other saline compoands. The
whole reaction of earbonate of magnesia, in the presence of car*
bonio acid gas, upon brines like ours, resembles somewhat, so
fiur as some of the final results are concerned, that of carbonate
of soda provided the latter does not exceed in chemical equiv-
alents the chlorid of calcium and sulphate of lime present, sul-
phate of soda, and carbonate of lime being formed in both in-
stances ; they differ only in one essential point ; in case of the
addition of carbonate of soda the chlorid of calcium is simply
eliminated (only traces of carbonate of magnesia being produced)
and its place supplied by a corresponding amount of chlorid of
sodium, while in the case of the addition of carbonate of mag-
nesia the chlorid of calcium is replaced by a corresponding
Quantity of chlorid of magnesium. The following statement
emonstrates the changes due to the addition of an amount of
carbonate of magnesia or of carbonate of soda chemically equal
to the amount of chlorid of calcium present, since, as soon as
that point is passed both carbonates produce sulphates of sodsi
as long as sulphate of lime remains in excess.

Sulphate of lime,
Chlorid of calcium, •

^ magnesium,

** potassium, .
Bromid of magnesium,
Chlorid of sodium, .
Water, .
Carbonate of lime, .

1 represents the original brine ; 2 represents this brine treat-
ed with the necessary amount of carbonate of magnesia; 3 rep-
resents this brine (1) treated with a corresponding amount of
carbonate of soda; 4 represents the original brine simply treat-
ed with an additional amount of sulphate of soda. We can
scarcely imagine the variety of brines and mineral waters which
m^ result in the course of time from this or similar brines, under
di£^rent geological conditions, particularly if we consider the
consequences, which must result from a mere difference in con-
centration before or after such changes as we have discussed in
these pages have been fully or partly accomplished.

Slim indeed are the chances of learning anything more definite
about the original composition of the saline mass of the oceanic
waters of the Silurian age from our mineral waters and brines ;
since the extensive exposure of our Silurian rocks renders them
subject to important changes by disinte^tion, and causes them
to react decidedly upon percolating saline solutions.

Syracnse, Uudk U, U67.










• • • •

• • • •

• • • •





















• • • •



• • • •

Digitized by


88 O. C, Marsh on a new genus cf fossil S^miges.

Abt. XL — Notice of a new Oeims offossSL Sponges from the Lower
saurian; by Pro£ O. 0. Mabsh, of Yale College.

Within the last few years several specimens of a remarkable
fossil hare been foand in limestone of lower Silurian age in
Franklin county, Kentucky. Although generally recognized as
Am/orphozoa by paleontolodsts who have seen them, and eyi-
dently new to science, no description of them appears to have
been published hitherto, except a brief notice oy Prof D. D,
Owen, who proposed for them the name of Scyphia digitata*

The Cabinet of Yale College contains a yery perfect specimen
of this fossil, recently obtained from Prof. Hovey of Wabash
College; and the writer had previously an opportunity of exam-
ining, in the collection of Mr. Sydney S. L^on, of Jeffersonyille,
Indiana, a large and fine specimen, tne original from which casts
baye since been supplied to the Yale and other museums, by
Prof Ward of Eochester.

In general appearance, and nearly all important characters,
these two specimens are yery similar. The form is that of a short
yase, or cup. with a row of arms extending outward and down-
ward from tne lateral surface. These arms are hollow, and open
directly into the main or central cayity, which is apparently
larger than in any other known sponge, recent or fossil.

The specimen in the Yale collection is about six inches in
diameter, and three in height It has nine arms of nearly eoual
size and length, all with openings at their extremities. The
base has a small protuberance near the center, but no other indi-
cation of attachment. The other specimen has eleyen arms, and
is nearly eleyen inches in diameter, by three and a half in height
Both specimens are externally silicined. The main orifice, or
mouth, is oyal in form, and in the smaller specimen is situated
at the end of a short neck, about one and one-half inches aboye
the bases of the arms.

As these specimens are without doubt generically distinct
from any yet described, the name Brachiospongia is proposed for
the genus they represent : and, since there is apparently more than
one species included under the name digitata, the triyial demgna-
tion of JRoemerana may be added, in honor of Prof. Ferdinand
Boemer, of BreslauUniyersity, whose inyestigations haye thrown
so much light upon paleozoic sponges. The specimen in the
Yale Cabinet may be regarded as typical of this species. Should
the form represented by the large specimen already noticed proye
a distinct species, this may appropriately be named Brachiospon"
gia Lyonii, from Mr. S. S. Lyon, Wie discoyerer of these interest-
ing fossils. A full description of these specimens, with illustra-
tions, will appear in an early number of this Journal.

Now Haten, Ct, Hay 26th, 1867.

* Second Report on the Geology of Ky., p. 111.

Digitized by


/. D. Dana an Crystalline form and Chemical constitution. M

Abt. XTI. — Orysiallogenic and OrysinUographic Contributions; by
James D. Dana. No. IV,* On a connection between OrystaUine
form and Chemical constitution, with some inferences therefrom.

Among oxyds, the protoxyds, like the metallic elements, are
characteristically isometricf in crystallization. The sesquioxyda
are as characteristically hexagonal, this being the form of the
sesquioxyds of iron, aluminum, and chromium. The deutoxyds
are typically tetragonal, as seen in the deutoxyd of tin (tin ore)
and of titanium (rutile and anatase). There are other forms
among protoxyds, sesquioxyds and deutoxyds; for example,
ZnO is hexagonal ; TiO' in brookite, and MnO* in pyrolusite
are orthorhombic ; but these cases, as the following observations
make apparent, may be regarded as a consequence of polymerism
— a principle that has been recognized by others as underlying

Regarding the atom of oxygen as double in its fondamental

* The preceding papers in this series by the writer are not numbered. They
are : I, On the Formation of Compound or Twin crystals, toI. xzx, 275, 296, 1886;
II. On certain laws of Cohesive Attraction (as illustrated by crystals); II, it, 864, ▼,
100, 1847, 1848; HI, On the Homoemorphism of mineral species of the trimetric
[and other] systems, II, xviii, 86, 131, 1864, with the antecedent papers in xvii, 86,
210, 480.

t I propose to employ in the forthcoming edition of my Mineralogy the terms
Itametne, Tetragonal (having a sonare base), and Orthorhomhie (erect on a rhombic
ba^e), in place respectively of Mofiometric^ Dimetric^ and IHmetrie Monomeitie
describes a line better than a cube ; the hexagonal prism is as much dimetric as the
square prism ; and the oblique prisms are as truly trimetrie as the right rhombio.
It is very desirable that the technical terms of science should be uniform over
the worldi, as far as possible, and that authors should be willing to yield their owd
usage for the sake of uniformity. The terms adopted appear to lie the best that
have been proposed, and have nlready extensive use in Europe. Iwmetric Is Haus-
manors term; MragofuU and hexagonal t with ThomhiCi are employed by Naumann.
Mohs's terms pyramidal fur the tetragonal system, and prismatic for the ortho-
rhombic, are exceedingly bad, as there are pyramids among isometric, orthorhombic
and hexngonal fonns, aa well as the tetragonal ; and prisms in all the systems ex-
cepting the ibometric.

There is additional reason, for our proposed change, in the natural relations of tiie
systems of crystallization. For the similarity in the names monometric, dimetric,
trimetric (the latter two the monodimetritehen and trimetriechen of Hausmann) un-
plies a fundamental relation in the forms ; while the true cla^ificatioo is aa foUo^ :
(1) j8ometric, including the isometric system, peculiar in the absence of double re-
fraction or polarization ; (2) Isodiametric (from laofj equal, and diameter), including
the tetragonal and hexagonal forms (alike named f^om the shape of the oase), char-
acterized by equsl transverse nxos or diameters, and uniaxial polarization; and (8)
Aniwmetrie (from mi^o^y unequal, etc.), including the remaining .systems, and dis-
tinct in having the axes or diameters all unequal, and biaxial polntizatiou.

Jfonoclinie, Diclinic, Trielinic (from Kaumaxui) I would retain, us they express
admirably the relations of the systems. CliHorhomhie is often used for the mono-
clinic system, and is well enough. But clinorhomboidal for the triclitiic would not
be desirable, as the French commonly use the word rhomboidal where others use
rhombic; and the diclinic could have no corresponding name, unless it be oUncrtct-
angular, which would be very objcctionnble.

Am. Joub. Sol— Second Seuies, Vol. XLIV, No. 130.— July, 1807.

Digitized by


90 /. D. Dana on a connection between

nature, the number of atoms of oxygen (or the negative ele-
ment) in the protoxyds is 2 ; in the sesquioxyds 6, or a multiple
of 8 ; in the deutoxyds 4.

It appears from a survey of all hexagonal and tetragonal com-
pounds to be a general fact, that the number of atoms of tbe
negative element is 3, or a multiple of 8, in the former; and 2,
4, or a multiple of 4, in the latter; and that, consequently, the
hexagonal and tetragonal systems are based on these namberB^
respectively, their symmetry being a consequence of it

1. Tetragonal species, and the number 4. — Among unisQicaJtes—
the siUcates which have the ratio 1 : 1 between the oxygen of the
bases and silica (SiO'), and the number of atoms of oxygen 4, or
its multiple — tetragonal species are common ; while none occur
among tne bisilicaies, in which the ratio is 1 : 2, and the number
of atoms of oxygen is 8, or its multiple. There are none abo
among the anhydrous carbonates, which likewise have the oxy-
gen ratio 1 : 2. But among these bisilicates and carbonates
there are examples of hexagonal species. The compounds OaW
(scheelite), Phw (scheeletine), Ph&o (wulfenite), Y^P (xenotirae)
are tetragonal, the last having 8 of oxygen (or 16 if doubled)
and the others 4. Matlockite (PbOl+PbO) is tetragonal, while
PbI+2PbO is hexagonal, and PbCl+2PbO is orthorhombic
Cerasine (PbCl+]?bC) is tetragonal; and the number of atoms
of the negative elements, O, CI, is 4. Hausmannite is tetra-
gonal, and with the usual formula ftnltn has 40. Yet the fo^
mula is better written AdsiSd, for this corresponds with itB dose
relation in form to the BO' or deutoxyd group, while tLaSbt is a
formula of the isometric spinel group. Similarly, the tetragonal
species chalcopyrite has the formula 2(€u, Pe)S+FeS*. Braun-
ite, taking the most recent formula for it, that of Bammelsberg,
(Mn, Si)'0", is apparently an exception. Its composition, as Rsm^
melsberg shows, corresponds to sStn+An-f-di^ and this fonnula
has 120, which is a multiple of 8, and satisfies the principle
under illustration. But the true arrangement of the consUta*
ents makes it not a sesquioxyd, as above, but a deutoxyd like
hausmannite, which it approaches in its tetragonal form ; for the
formula may be aStn^fin+flnSi, which is equivalent to 2 of haus-
mannite and 1 of a silicate analogous to the tetragonal species
zircon (ZrSi) * The deutoxyd of manganese, MnO* (pyrolasite)^
is orthorhombic, and approximately isomorphous with ortho-
rhombic TiO* (brookite), the former having for the angles of

* Hausmaimiie approaches more closelv the aoataM form of TiO* than the ni-
tile form, the angle oetween O and the plane made l-t in anataae beii^ 119^ fS'i
and 0: 1 (which mi^ht as well be l-t) in hansmannite being 121^ 8'. firamiitc is
much nearer cassitenie, ratile, and xiroon, the corresponding angles for Oonn ]>fr>
amidal plane in these four species being, respectively, 136^ 26', 186^ 26', 1S1^49',
ini^ 60'. Thus the anatase and rutile form of TiO' are eererallj represented hf
hausmannite and braunite.

Digitized by


Crystalline form and Chemical constitution. 91

the prism /, and the domes 1-% 1-2, respectively, 98^ 40', 104^
22', lO?"" 54' ; and the latter for the correspondiDg angles 98'' 16',
96"" 46', 99'' 60'. MnO' in the tetragonal state is unknown
except when it is in combination with 2MdO, as in haosmannite.
The protozyd of manganese, MuO, it may be remarked, is iso-
metncy like MgO, it having been obtained artificially in octa-
hedrons and cabo-octahedrons by Deville (C. R, liii).

Among artificial compounds, there are the following tetragonal
species aU conforming to the principle stated :

KF+HF; fte8+4aq; SiB + 7aq; ]5riS + 6aq; iriSe + 7aq; ^Be+7aq;
(t.»)»f ; (Am,fi)»f ; (fc,fi)3l«; ('^; (Ja^aaq; (Am+fl)B"+8«q;
IiO,!^; KCl+0ua+2aq; Am01+0u01+2aq; Ag5+2NH», Ag0r+2NH3,
sUc(aeeticaddH4De+12ftq; Cale+Calc+Saq, fij!Lc+281c+2aq, Aglc+
t8Xfi+.2fliq ; (iOa+f £[)'T (tartaric add, ooDUmlng 50)+9bT-f 7aq.

Omitting a few complex organic compounds, these are all the
tetragonal species in the two volumes of Aammelsberg's Crys-
tallographic Chemistry excepting AgQi, Hg'Ci. Other examples
might be mentioned, but the above are fully sufficient

The correspondence with the law for tetragonal species is so
general that we may reasonably believe that the apparent excep-
tk)Ds, where the composition and crystallization are correctly
given, may be brought into conformity to it by an application
of one or the other of the foUowingpnnciples.

0. The tmneipje of polymerism,— ^g' CI is Hg*Cl* in the new
system or chemistry ; and if the whole is doubled, it becomes
Bg*Cl*, which is probably the true formula of this species in
the tetragonal state, the only crystalline state yet known.

b. Part of the ingredients may be only accessory, or subordinate

to a dominant part which determines the crystallization. — Water is

I commonly admitted to be present in this way in most of the

I aanponnds in which it occurs; although essential to the species,

it is subordinate, crystallogenically at least, to the rest. Water

IB now believed to oe not the only substance that may play the

part of indifferentism in compounds, and many formulas have

of late been written by chemists admitting this. Apophyllite

ii a tetragonal species consisting of ft+2Si+2£[. Making the

I water basic, there is still no conformity to the type of either the

\ nnisilicates or bisilicates, the oxygen ratio for the bases and silica

I being 8:4. If half the water be regarded as basic, and the

I formula be written (ift. ^O'Si+^cSi, it is made to consist of a dom-

I inant part which is a unisilicate analogous to the tetragonal spe-

I des meionite, mellilite, etc., and% bisilicate which is a kind of

opal or waterdass, well known to be a " colloid," or uncrystal-

lizable, and wnich therefore might well have no effect toward

modifying the crystallization as determined by the other part.

Digitized by


92 /. D. Dana on a connection between

2. Hexagonal species^ and (he number 8. — Hezasonal species
have been stated to occur among the sesquioxyds (as Fe'O',
A1«0», Cr«0'); the bisilicates (as in beryl, eudialyte, dioptasc,
pjrosmalite, cbabazite, gmelinite) ; and the carbonates (in cal-
oite, and the allied species) ; in which compoanda the number of
atoms of oxygen is 8, or a multiple of 8. Other examples are:

Pyimrgyrite and pronstite, 8AgS-f (8b, Ab)*S*, in which the number of atomi of
aalirfrar is 6 ; gibbette £lft' ; alnnite £S+8Sl3+6aq ; apatite sCa'P+Caa;
ooqnimbite l^eB'+9aq; Al«Cl»+12aq; AgS+8aq ; ilS»-f 27aq; SrOS«0*+4«l.
and the corresponding salt of lime, and of lead; tft; J^aA ; (6a+fis)9; AgO,
010''-Hfi; 8Na01 + IrCl»+24aq; KCl+2Mg01+12aq; B4gCl+PtCl>-M«q.

The exceptions to the principle are to be accounted for ia the
same manner as those under the tetragonal system. Along side
of the hexagonal sesquioxyds, Fe'O', A1«0', Cr*0', there is
the hexagonal protoxyd ZnO, similar in angle. Applying the

Erinciple of polymerism and writing the formula Zn'6*, it then
as, liice the sesquioxyds, 8 of 0. This view of the protoxyd
is abundantly illustrated and sustained among the silicates. For
the constitution of the larger part of them (garnet, scapolite,
epidote, etc.) is based on the mutual replacement of 1 of sesqui-
oxyds (B*0'), and 8 of protoxyds (8B0) ; and this mutual re-
placement signifies isomorphism of B'O' and B^O'. Again,
graphite, or hexagonal carbon, has been shown to have its atomic
weight nearly three times as great as that of ordinary carbon;
and it is altogether probable, therefore, that in this hexagonal
state carbon is 0', in accordance with the principle in view.
Hexflgonally crystallized water, on the same grouno, is not H0|
but U >0 '. ZnS occurs both in isometric and hexagonal forms ;
and while the former may be simply ZnS, the latter should be
Zn'S^; and so for the hexagonjQ sulphid of Fe, Ni, Cd, we
should have Fe'S' (troilite, i)yrrhotine) ; Ni»S* (millerite);
Cd»S' (greenockite) ; and similarly Ni»As» (copper nickel);
Ni»Sb» (breithauptite).

8. Itometric system. — The number of atoms of the negative
element in isometric species appears to be either 1, 2, 8, 4, or a
multiple of 8 or 4; and this aiversity accords with the twofold
nature of a cube ; that is, (1) an equiaxial square prism, and (S)
(if a diagonal be made vertical) a rhombohedron of 90^ ; for it
has this double relation to other forms. Accordingly, isometrio
forms oceur among protoxyds, protosulphids, protochlorids, eta ;
also deutoxyds; also anisilicates ; in leucite, analcime ; also in—

ftg^B*, or boradte; Is; Sb; jTaOw fefir; Jfafir; Jfid + Saq; OnCl+aaq;
^Br+6aq; AmGl+SnCl>; ECy+AgCj; SffaSi^erS'+Qaq ; the alnina, wlMdK
UvB 160. bettdes 240 in the waUr; ]Sra2[c+282je; JiTaW+WW.

Important chemical and crystallographic conclusions flow fixxa

Digitized by VjOOQIC

CiystalUne form and Chemical constitution. 93

the principle which haA been explained, if it is su9tained, as we
believe, by the facts. A few only are briefly touched upon.

1. It follows that the hexagonal state of the elements may be
one corresponding to SB, or SnR ; that while zinc in the isometric
state if such exists (about which there is doubt) is Zn ; in the
hexagonal it may be Zn', the same state in which it exists in
hexagonal oxyd of zinc. So also Pd, As, Sb, may represent the
isometric state of the elements palladium, arsenic, antimony ;
but Pd', As', Sb', the hexagonal; and so for other cases.

2. The oxyd of copper, CuO, which may also be written
€uO', is dimorphous, it occurring both in isometric and ortho-
rhombic forms ; and the orthorhombic form is closely isomorph-
ous with TiO' in brookite — /: /and /: i in the oxyd of copper
being respectively 99° 39' and 126° 29', and in brookite 99* 50'
and 126"^ 15'. This relation to TiO' shows that the ortho-
rhombic state of the cupric oxyd should have the formula
€uO', or that of a deutoxyd, and the isometric alone that of
CuO. And it indicates further that the element copper may
exist theoretically, if not actually, in two corresponding poly-
merous states.

3. As long since illustrated by Laurent, the protoxyds RO,
sesquioxyds K^O', deutoxjjds RO^, and other grades of oxyds
RO', RO* (and the same in corresponding chlorids, sulphids,
etc.), in which 1 part of oxygen balances, in its affinity, 1, f, \^
etc., parts of the basic element (as is seen on dividing hj the
number of atoms of oxygen so as to reduce the oxygen m all
the above formulas to 10), may be viewed as containing the
basic element in as many different states as there are grades of
the above compounds. For convenience these states may be
designated by using the Greek letters as follows :

« , i RO R203 RO^ R0» RO*

Formulas ^^^ ^^^ ^^ ^^ ^^^

States of jR R| R^ R^ R^

Basic element ]^j^ ^R yR dR eR

or the alpha, beta, gamma, delta and epsthn states. It is observed
that 8R0=R'0'; 8(|?R0)=R«0'; 2(rR0)=R0«; 8(yR0)=:
fRO' ; 3(^R0)=R0'; and so on: in other words, the one
molecule R'O' corresponds to ^ree of ^RO ; and in 3(?R0) there
are as many atoms of the basic element /^ as of O.

Now, if a sesquioxyd occurs in isometric crystals, as supposed
to be true of Fe'O' (but reasonably doubted), that sesquioxyd
is not Fe^O', but may be FejO. This is but the converse of the
conclusion, stated above, that if a protoxyd occurs in hexagonal
crystals it is not then RO, but may be R'O'. So in other cases :
if oxyd of tin had an isometric as well as a tetragonal form, the
former in the crystalline state should be Sn|0, andonly the latter

Digitized by


94 /. A Dana an a cannectum between

SdO'. a metal in the different states B, Bf BI9 b>^ acooid-
ioglj, the same isomorphic power; and so also, 2B, 2R}, 2B};
and SB, 8B}, 8B|. Hence nnder the principle explained-—

BO, R|0, BxO ahoold be alike Umnetrk in crystallisation.
2(B0), 2(R|0), 2(B^0) may be Uiragonal <" «*

3(R0), 8(R|0), 8(B^0) may be hacagonal ** •«

Qnartz, which is hexagonal silica, should, according to the
above, be 8(SilO), or else 6(SiiO), and not 2(SiiO)=xSiO'. SiO'
is hence unknown in the c^stalline state ; and if ever obtained
ci^stallized it will in all probabilitj have one of the forms of

Online LibraryJohn AlmonThe American journal of science and arts → online text (page 61 of 102)