786.7 A9lo cop 1
Audsley
Orpan-stops and their artistic
registration,,.
kansas city ||Sl public library
=Si *
Books will be issued only
on presentation of library card.
Please report lost cards and
change of residence promptly.
Card holders are responsible for
all books, records, films, pictures
or other library materials
checked out on their cards.
QCT1
? 1318
WAR 14 1979
i 1) 1981
3 1148 00076 3458
CHURCH OF SAINT-OUKN, UOUtiN
ORGAN -STOPS
AND THEIR
ARTISTIC REGISTRATION
NAMES, FORMS, CONSTRUCTION,
TONALITIES, AND OFFICES IN
SCIENTIFIC COMBINATION * *
BY
GEORGE ASHDOWN^UDSLEY, LL.D.
ECCLESIASTICAL AND ORGAN ARCHITECT
AUTHOR OF "TEE ART OF ORGAN-BUILDING," "THE ORGAN OF THE
TWENTIETH CENTURY," NUMEROUS ARTICLES ON ORGAN MATTERS
AND ACOUSTICS, AND AUTHOR AND JOINT AUTHOR OF TWENTY-THREE
WORKS ON ARCHITECTURE, ART, AND INDUSTRY
ILLUSTRATED
NEW YORK
THE H. W. GRAY CO.
AGENTS FOR NOVELLO & CO., LTD.
COPYRIGHT, 1921, BY THE IT. W, GRAY Co,
COPYRIGHT RENEWED, 1949
FOREWORD
An attempt has been made in the present work to furnish the
organist, and especially the organ student, with a work of ready ref-
erence respecting the numerous Stops which have been and now
are introduced in the Organ: giving, so far as is practicable in a
necessarily brief and condensed form, their various names in dif-
ferent languages, peculiarities of formation, tonal characteristics,
and value and office in scientific and artistic combination and
artistic registration.
It is hoped that the work will be accepted as a text-book in
Organ Schools and Conservatories of Music, leading toward, and
lending help to, a branch of study of the greatest value and impor-
tance to the organ student; but one which, in too many quarters,
has been seriously neglected. That a thorough knowledge of the
tonal forces of the Organ and their varied powers in scientific and
artistic combination and registration, for the production of special
and expressive qualities of both compound, unimitative organ, and
imitative orchestral tones, is essential to the accomplished organist,
admits of no question. Accordingly, it is highly desirable that the
student should leave the organ school with, at least, a foundation
laid for that thorough knowledge. It is with the earnest desire
to contribute effectively to the laying of that foundation that the
present work is placed at the command of both teachers and pupils.
In another direction, and one of great importance at the present
time, this work, if properly understood, will prove of considerable
value; namely, in giving reliable advice and assistance in the prepa-
ration of stop appointments and apportionments for new Organs.
It will lead away, if read aright, from the present systemless, in-
sufficient, and largely retrograde prevailing method of stop appoint-
ment, toward a logical, scientific, and definite artistic system, in
FOREWORD
which compound tone production in all its desirable Conns can be
carried out in accordance with the natural laws of sound.
Care has been taken to render correctly all stop-names in the
different languages in which they originated and are employed
to-day: reference to which will prevent the use of the incorrect
names, either wrongly spelt or improperly compounded of words in
different languages, which are so commonly found on English and
American organ-builders' draw-stop knobs or tablets. With such a
Glossary as is provided in the present work, there need be no mis-
takes made or incorrect renderings of stop-names perpetrated in new
organ consoles,
Certain stop-names introduced by organ-builders have been
omitted from the Glossary on account of their absurd or meaning-
less character. These are names which have no relation to
anything connected with either the formation or tonality of the
pipes forming the stops. All such meaningless names should be
condemned by every organist and lover of the Monarch of all
Instruments.
GEORGE ASHDOWN AUDSLKY,
BLOOMFIKLD, NEW JERSEY,
JANUARY, 1921.
PLATES
CHURCH OF SAINT-OUEN, ROUEN . . Frontispiece
PAGE
PLATE I. BASSOON; BASSOON; BELL GAMBA; CLARINET;
COR ANGLAIS 4
PLATE II CORNO DI BASSETTO; DIAPASON; DOLCAN;
FLITTE A PAVILLON; FLUTE HARMONIQUE ... 78
PLATE III GEMSHORN; HORN; KERAULOPHONE; OBOE;
ORCHESTRAL OBOE H^
PLATE IV THE HASKELL LABIAL TUBA MIRABILIS . 266
ORGAN-STOPS
THEIR ARTISTIC REGISTRATION
*' Without the least hyperbole it may be said, that c&teris paribus, the man
who is capable of being affected by sweet sounds, is a being more perfectly
organized,, than he who is insensible to, or offended by them. "
DR. CHARLES BURNEY.
In Organ-stop Registration, as in artistic orchestration, "there
is no royal road to learning. 71 Natural musical taste and apprecia-
tion of the beautiful in musical sounds may go far in the education
of the organist ; but earnest studies along scientific and artistic lines
are in all cases necessary for the complete command of tonal color-
ing, by means of registration and the knowledge of the tonal values
of the multitudinous and varied voices of the Organ. As Dr.
Burney truly says: "The science of musical sounds, though it may
have been deprecated, as appealing only to the ear, affording nothing
more than a momentary and fugitive delight, may be with justice
considered as the art that unites corporal with intellectual pleasure
by a species of enjoyment which gratifies sense without weakening
reason; and which, therefore, the Great may cultivate without de-
basement and the Good enjoy without depravation."*
Let the organ student realize, once for all, if he is to become an
artist, that haphazard methods of registration must be shunned,
and dependence placed on knowledge acquired by study, observa-
tion, and experience. The earnest study of artistic registration and
the tonal value of organ-stops, singly and in combination, should
accompany all lessons on, and the practice of, the technical branch
of organ playing: but how seldom, in organ schools, is a student's
attention specially directed to the all-important matter of tonal
coloring.
* " A General History of Music, " by Dr. Charles Burney, London, 1776.
I
2 ORGAN-STOPS
Excellence and precision of manual and pedal technic are, of
course, imperative in the satisfactory rendition of an organ composi-
tion; but technic is not the only, or, perhaps, the most important,
factor. It may be said to be the skeleton which has to bo clothed
with the flesh and nervous power of beautiful and expressive sounds
alone secured by scientific combination and artistic registration
of appropriate and expressive tonal elements. The most consum-
mate technical skill is altogether insttfficient in the presence of a
careless and inappropriate registration to produce a truly artistic
and expressive rendering of an organ composition.
With the acquisition of manual and pedal technic should go a
serious study of the tonal forces of the Organ, their combination,
and effective registration; for it is appropriate tonal coloring which
gives the spirit and expressiveness to the music, which no single
performer save the organist is capable of producing, and for which
no instrument save the Organ can furnish the necessary tonal ele-
ments music which can only be surpassed by the united forces of
the grand orchestra under the control of an accomplished conductor.
The organist is as the conductor; the many and diverse tonal forces
of the Organ are his instrumentalists : it rests with him to marshal
those forces, in ever-changing groups, so as to produce the artistic
and life-giving effects his music demands; and for the interpretation
of his most refined conceptions. Again, the organist is as the painter
before his canvas, brush in hand; the stops of his Organ are the
colors of many tints, hues, and shades, spread for his ready use on a
serviceable palette; to be combined, at will, in endless variety as the
spirit of the tone-picture inspires him. What a wonderful world of
tone the organist can live in if he only realizes his birthright his
citizenship in the land of beautiful sound. The organist stands
supreme in the musical world the master of the most stupendous,
the most wonderful musical instrument ever conceived by the mind
and fabricated by the hand of man. Think of it, ye Organists,
tod rise to the level of your birthright !
Before going deeper into our subject, we may here give a few
more pertinent words from the able pen of England's great musical
historian. Alluding to tonal matters and the Organ of his day, he
remarks: "Of Musical Tones the most grateful to the ear are such
as are produced by the vocal organs. And next to singing, the most
pleasing kinds are those which approach the nearest to vocal ; such as
can be sustained, swelled, and diminished, at pleasure. Of these,
first in rank are such as the most excellent performers produce from
the Violin, Flute, and Hautbois. If it were to be asked what in-
THEIR ARTISTIC REGISTRATION 3
strument is capable of affording the GREATEST EFFECTS ? I should
answer, the ORGAN, which can not only imitate a number of other
instruments, but is so comprehensive as to possess the power of a
numerous orchestra. It is, however, very remote from perfection,
as it wants expression, and a more perfect intonation." Much of
what is said in the concluding sentence is true of the Organ of to-
day, greatly superior as the best examples are to the Organ known to
Dr. Burney. The Organ still wants full powers of expression com-
bined with tonal flexibility throughout all its divisions. Accordingly,
while much has been done of late years in certain directions, still
more has to be achieved by the organ-designer and organ-builder
before the Organ can be pronounced an instrument for the true
artist and inspired virtuoso. Certainly, what has been done during
the last seventy years in the direction of imitative orchestral-toned
stops, was hardly foreshadowed in Burney's day. The Swell Organ
known to him was usually of short compass and very indifferently
_stop-apportioned. Writing in 1771, this distinguished author
remarks :
"It is very extraordinary that the swell which has been introduced into the
English organ more than fifty years, and which is so capable of expression and of
pleasing effects that it may well be said to be the greatest and most important
improvement that ever was made on any keyed instrument, should be utterly
unknown in Italy ; and now I am on the subject, I must observe that most of the
organs I have met with on the Continent seem to be inferior to ours by Father
Smith, Byfield, or Snetzler, in everything but size! As the churches there are
very often immense so are the organs; the tone indeed is somewhat softened and
refined by space and distance; but when heard near, it is intolerably coarse and
noisy ; and though the number of stops in these large instruments is very great,
they afford but little variety, being for the most part duplicates in unisons and
octaves to each other, such as the great and small I2ths, flutes, and I5ths; hence
our organs, not only the touch and tone, but the imitative stops are greatly supe-
rior to those of any other organs I have met with."*
Alluding to the first swell introduced in a German Organ, which
is understood to have been in the Organ, built by Hildebrand, in
1764, for the Church of St. Michael, Hamburg, Burney says: "A
swell has been attempted in this instrument, but with little effect;
only three stops have been put into it, and the power of crescendo
dimimiendo is so small with them, that if I had not been told there
was a swell, I should not have discovered it."f
We have introduced the subject of the swell at this early point
* " Present State of Music in France and Italy," 1771, P- 375-
f" Present State of Music in Germany, the Netherlands, and United Provinces," 1775-
Vol. II., p. 103-
4 ORGAN-STOPS
of our brief essay because it has a direct and very important bearing
on the matter of artistic registration. If one examines the stop
appointments of the Organs of Burncy's days, or of all the old Organs
in which no swell was introduced; and, indeed, of those, of later
times, which possessed only a single flexible and expressive division,
one cannot avoid being impressed with the very limited opportuni-
ties they present for varied registration artistic registration being
practically out of the question. So few were the useful combina-
tions of the stops in the older Organs that they were commonly
commanded by foot-levers, called "combination pedals/' Artistic
registration, as it is understood to-day, was practically unknown in
connection with the old Organs : and, it must be acknowledged that
it still continues undesirably difficult in too many of the modern
Organs tonally schemed on old-fashioned lines. Stops properly
chosen and apportioned among the different divisions of the Organ,
under the principles of class grouping, tonal contrast, absolute
flexibility regarding strength of voice, and compound expression, as
long advocated and first practically introduced by us in the Organ,
are the only efficient means by which scientific tonal combination
and truly artistic registration can be carried out to the extreme
extent, now necessary in the proper and effective rendering of the
works of the great and distinguished composers of organ music, and
especially the modern transcriptions of orchestral scores.
The system of stop-apportionment and control which we have
devised and strongly advocate, is widely different from the old-
fashioned and seemingly purposeless method which has so long ob-
tained in organ-building, and may be briefly described here. It
comprises the grouping of stops of different tonalities in the several
manual divisions of an Organ, according to their special offices in
the complete tonal scheme of the instrument, and for the ready
production, without recourse to undesirable coupling, of the numer-
ous and very varied organ and orchestral tonal combinations and
effects, absolutely called for in modern artistic organ playing. To
secure what is essential, each division has a distinct general and
special tonality, contrasting with that of every other division; un-
necessary duplication of stops of the same tonality being thereby
avoided. Each division, having its own office to fulfil, is given
special powers of flexibility and expression; those devoted to the
stops representing the "wood-wind" and "brass-wind" forces of
the grand orchestra being divided into two tonally contrasting por-
tions, each of which is given independent powers of flexibility and
expression. The division devoted to the foundation stops and those
THEIR ARTISTIC REGISTRATION 5
properly belonging to the Organ, commonly designated Great Organ,
is also divided into two portions, only one of which, comprising
special registrational, harmonic-corroborating, and lingual stops, is
rendered flexible and expressive; although both subdivisions may
be so treated if deemed desirable. The Solo, String, and Ancillary
Organs are rendered flexible and expressive, but do not require to
be divided.*
On a first and hasty consideration, the introduction of the new
principles of divisional and subdivisional tonal contrast and flexibil-
ity and expression may not seem to the organist, who has only
known the present unsystematic and haphazard method of stop-
apportionment (to be observed in every modern organ specification) ,
to be of any great scientific or artistic importance or value. But let
him devote sufficient time to the study of those principles, with an
open and unprejudiced mind, and he will find that for the conven-
ient and certain production of refined and beautiful tonal combina-
tions and orchestral effects, and for facility in scientific and artistic
registration, the combined principles form the greatest and most
valuable improvement in the tonal branch of organ-building ever
instituted in modern days, since the time we placed in the year
1870 three independent tonal subdivisions on one manual clavier;
two of which are flexible and expressive, being inclosed in separate
swell-boxes. This unique step may or may not be considered a small
beginning; but, as Burney pertinently remarks: "The feeble be-
ginnings of whatever becomes great or eminent, are interesting to
mankind. To Artists, therefore, and to real lovers of art, nothing
relative to the object of their employment or pleasure is indifferent."
Let us consider briefly what is possible to be done with a single
clavier, tonally appointed in the manner alluded to. Say that it
commands sixteen stops of varied and artistically chosen tonalities
and pitches; eight of which are inclosed in, say, swell-box A and
eight in swell-box B ; the stops being divided so as to provide con-
trasting voices and be practically independent. It also being ar-
ranged that either one or both the subdivisions shall be available on
the clavier at the pleasure of the performer. Now, for the considera-
tion of registration, we shall confine our remarks to the element of
flexibility, neglecting for the present the all-important element of
expression, and suppose both the subdivisions to be commanded by
the clavier. It can be readily realized that by simply opening either
* For full particulars, see Chapter XI., on the Co ncert-room Organ, in our work, "The Organ
the Twentieth Century."
6 ORGAN-STOPS
swell, more or less, the voices of the inclosed stops can be graduated
to any desirable strength of intonation an absolutely invaluable
property in refined and artistic registration, multiplying the utility
and combinational effects of the stops at least tenfold. Now suppose
a registration of certain stops of both subdivisions is essayed, the
voices of which are in sharp contrast, the character of the compound
tone produced can be altered, at will, without changing a stop,
either on one side or the other; and the relative strengths of its con-
trasting elements adjusted through the flexibility imparted to the
voices of the inclosed stops, by the simple opening or closing of the
shutters of the allied swells, commanded by the expression levers.
By the simple touching of these levers any desirable intensity of
tone and effects of light and shade can be imparted to the music
being performed: and this without removing the hands from the
clavier, or having any resort to undesirable coupling, which would
unavoidably cripple the independence of another clavier.
What we have attempted to describe may not appear, on first
impressions, as in any way remarkable to the organist acquainted
only with the one-ply stop-apportionments of the Organs of to-day.
But let us inform him of the fact that the multitudinous and prac-
tically inexhaustible tonal effects, colorings, and refined nuances,
easy of production, under our system of divisional and subdivision^
stop-apportionment and compound flexibility, on this single clavier,
could not be produced on the claviers of the largest Concert-room
Organ ever constructed up to this year of grace (1920), And let
him try to realize what could be done on an Organ in which three or
four of its manual divisions are equipped in the compound manner
described above. We foolishly essayed the task, and our mind was
quickly baffled in the attempt.
Up to this point we have treated only of the stationary com-
pound tones, produced by registration, under the simple graduation
of strength, effected by the property of flexibility, which merely
called for a set or temporarily fixed adjustment of the shutters of
the swells. We have now to consider the operations of the allied
property of compound expression, which can be exercised in four
different ways. i. Expression can be given simultaneously to the
tones of both subdivisions; by operating the expression levers of the
swells (located close together) at the same time, 2. Expression
can be given to the tones of cither subdivision, the tones of the other
subdivision remaining unaffected at the fixed degree of strength
desired; by adjusting one expression lever at any set point, and
operating the other as required. 3. Expression can be given to the
THEIR ARTISTIC REGISTRATION 7
tones of the subdivisions alternately; by operating the expression
levers alternately. 4. Expression can be given to the tones of the
subdivisions simultaneously in contrary directions, one series under-
going a crescendo while the other series is undergoing a decrescendo;
by operating the expression levers, in contrary motion, by both feet
(sans pedale).
The principles of compound flexibility and expression would be
of comparatively little value in the absence of the correlative system
of classified stop-apportionment and tonal contrast we have in-
troduced. There are two grand divisions of organ-stops. First,
that division embracing those stops essentially of a solo character
which imperatively demand the fullest possible powers of expression ;
but which are practically independent of combination or registra-
tion, save for the purpose of accentuation or coloration of tone,
Among such stops are those which represent important orchestral
instruments such as the Violin, Violoncello, Flute, Oboe, Clarinet,
Trumpet, Horn, and Trombone. Secondly, that much larger divi-
sion composed, for the most part, of organ-toned and unimitative
stops, which are largely, and to a considerable extent exclusively,
employed in combination and registration, for the production of
compound tones of great variety and pronounced colorings. For
the full development of these purposes, the system of classified
divisional and subdivisional stop-apportionment, under complete
control as regards flexibility and expression, we advocate, is abso-
lutely necessary from the simple fact that such essential artistic
conditions cannot be met tinder the aimless and heterogeneous
method of stop-appointment followed, practically without varia-
tion, in the organ-building world to-day. If this view of the matter
is questioned, let anyone, capable of forming an open-minded opin-
ion of a scientific and artistic caliber, take a hundred or more of the
so-called specifications of the more important and representative
modern Organs and carefully consider their lists of stops and judge
for himself.*
There are two classes of labial stops which are of great value in
scientific combination and artistic registration; but which have been
very unwisely neglected of late years, partly on account of a want of
a full knowledge of their great offices in tone-production ; but chiefly
on the part of those organ-builders who desire to evade the trouble
and uncertainty attending their proper formation, artistic voicing,
* It may be helpful to compare any of their tonal schemes with that given for the Concert-
room Organ in "The Organ of the Twentieth Century.' '
8 ORGAN-STOPS
and scientific regulation. We allude to the harmonic-corroborating
mutation and the compound stops; the latter commonly designated
MIXTURES, which, as has been wisely said of other things, "when
they're good they're very good, but when they're bad they're
horrid." They are usually bad! It is quite easy to understand, in
this inartistic and dollar-worshiping epoch, the objection of the
organ-builder to the introduction of the necessary amount of har-
monic-corroborating work in a large Organ. A MIXTURE, of five or
more ranks, requiring three hundred and five or many more pipes,
properly composed, scaled, and artistically voiced, and, what is of
equal importance, scientifically graduated in tone in strict accord-
ance with the natural laws of compound musical sounds, is a problem
very few organ-builders will care to solve. It cannot be satisfac-
torily essayed in the hurry and noise of the ordinary factory voicing-
room. On simple trade grounds, such troublesome and labor-de-
manding stops are to be omitted in all possible cases; and an ex-
amination of the great majority of modern organ stop-appointments
clearly show how systematically they are omitted. Alas for Art !
We have before us as we write the lists of the stops of two large
and important Organs, which may be accepted as representing the
latest and highest achievements in organ-designing, in the direction
of stop-appointment. The largest presents upwards of two hundred
and seventy complete and independent speaking stops; and in this
immense number there are only five MIXTURES, comprising in all
only twenty ranks of pipes. The other list presents one hundred and
one speaking stops (including derived and borrowed stops) ; in which
there are three MIXTURES, comprising, in all, only eight ranks of
pipes.
The most effective criticism we can pass on the insufficient
harmonic appointments detailed above, is to give the contrasting
appointment of our scheme for the Concert-room Organ, set forth in
"The Organ of the Twentieth Century." In this there are two
hundred and twenty-three complete and eight derived speaking
stops; among which are apportioned fourteen MIXTURES, compris-
ing, in all, sixty-four ranks of pipes. Great as this number may
appear, by comparison, we are prepared to prove, on both scientific
and artistic grounds, that for such an Organ there is not a rank too
many; indeed, had we not introduced the Ancillary Harmonic Organ
a larger number of compound harmonic-corroborating stops would
have been necessary to meet the demands of artistic and effective
registration and tone coloration.
A great injury was done to the tonal structure of the Organ by
THEIR ARTISTIC REGISTRATION 9
the false idea, introduced by self-interested parties, that the har-
monic-corroborating stops, simple and compound, could be dis-
pensed with, provided certain stops were introduced, the voices of
which were naturally rich in harmonic upper partials, such as
heavily-blown, small-scaled, string-toned stops ; but a greater mistake
was never made. Such highly-colored penetrating tonalities are of
very little use in artistic registration, generally objectionable, and
absolutely valueless in the creation of delicately colored qualities
of tone. Such qualities of compound tone are only possible by the
use of softly- voiced stops associated with such stops as the scientifi-
cally formed and tonally graduated DULCIANA CORNET or HAR-
MONIA ^ETHERIA.
In artistic registration, two distinct classes of musical tones
obtain those due to the Harmony of Analogy and those due to the
Harmony of Contrast. The former class is produced by the regis-
tration of stops of the same family or of very closely allied families,
differing only slightly in strength of voice, or between which a very
close affinity exists. The latter class, and much the more effective
and important, is produced by the skillful registration of stops of
widely different families, of contrasting tonalities, and of pitches
far apart. In both classes harmonic-corroborating stops are valu-
able, and, if properly chosen, enrich without altering to any unde-
sirable degree their characteristic tonalities. Light and shade can
be imparted to a compound tone without changing its dominant
character; while, on the other hand, very refined colorings can be
imparted by introducing compound harmonic timbre-creating stops.
Organists will, however, have to wait patiently for these; for organ-
builders are at present trying to get rid of, rather than contemplat-
ing the introduction of new and more valuable, compound harmonic-
corroborating stops: that must be evident to everyone who takes
the trouble to critically examine the lists of stops in the so-called
specifications of Organs being turned out to-day.
The beauty that perfectly constructed and scientifically bal-
anced compound harmonic-corroborating stops impart to the vari-
ous tonal groupings characteristic of the Organ is hardly known
to the organists of the present time. This is greatly to be regretted ;
and the knowledge will be very difficult to obtain until some thor-
oughly scientific artist in compound harmonic tone production
furnishes organists with the means of acquiring it.
In scheming the stop-apportionment of every division and sub-
division of an Organ, the special value and office of every stop, in
solo effects and in artistic, combination, should be studiously con-
io ORGAN-STOPS
sidered. There should be no combinational stops inserted which
would prove of little use. It should be realized that each division
and subdivision of an Organ should be devised to occupy a distinct
place, and fulfil a special office, in the tonal economy of the entire
instrument. As great a degree of tonal independence should be
aimed at as general conditions permit of; thereby preventing any
undesirable resort to coupling of the claviers. When any two
claviers are coupled, one or the other is for the time being crippled
tonally, unless under some exceptional conditions. Such coupling
should seldom have to be resorted to, unless necessary for the mass-
ing of different and contrasting tones and the production of full
effects. The more perfectly an Organ is stop-apportioned in its
several divisions, the less will the resort to coupling be required,
especially in artistic registration for music of an orchestral or sym-
phonic character.
There never has been a more serious blow given to the science
and art of organ tonal-appointment than that given when Leonard
Dryvers, organ-builder, of Wesscl-Loo-Louvoin, sprang on the
organ-building world his unscientific, inartistic, and pernicious
system of stop-appointment, tinder the appellation "I/Orgue
Simplifie." A system founded on trade expedients, find devoid of a
single scientific or artistic clement in its favor. Had it been con-
fined to its originator and his interested supporter Couwenbergh
little injury might have accrued. Unfortunately it was espoused
by the late Robert Hope-Jones, and is still advocated by some of his
followers. On the other hand, and very fortunately for the true art
of organ-building, the system (if system it can be called) is con-
demned by every thoughtful and accomplished organist and lead-
ing organ-builder of the present hour. Prom a purely artistic and
common-sense point of view we raise our voice in protest against a
system of tonal appointment at variance with all the laws and
canons of musical sound.
The Organ, tonally appointed on the Dryvers system, described
by H. V. Couwenbergh, of Averbode, in his pamphlet published in
1887, comprises only the following stops: MONTRK, 16 FT,; BOUR-
DON, 16 FT. ("dont le dessus est ouvert en FLOTE HARMONIQUE");
VroLON, 16 FT.; and BOMBARDE, 16 FT. From these four stops, of
extended compass, are formed no fewer than thirty-one stops of
i6ft, 8ft, 4ft, and 2 ft pitch: the MONTHE furnishing seven stops,
all of the same quality and strength of tone; the BOURDON furnish-
ing ten stops of somewhat irregular quality and strength of tone; the
VIOLON furnishing seven stops of the same quality and strength of
THEIR ARTISTIC REGISTRATION u
tone; and the BOMBARDE furnishing also seven stops of the same
quality and strength of tone : only four qualities of tone, and no
mutation and high-pitched harmonic-corroborating ranks, in an
instrument presenting after a fashion thirty-one speaking stops.
Can anyone gifted with musical knowledge and refined taste imagine
such a tonal monstrosity anywhere save in a noisy merry-go-round
at a seaside resort or a country fair? To conceive it an instrument
for the performance of true and refined organ music is impossible :
to speak of its stops being capable of scientific combination and
artistic registration would be laughable.* Yet the fact remains
that instruments we will not call them Organs have been con-
structed on such a system in this country : but we are not aware of
any examples having been perpetrated either in England or France
We sincerely trust that such an outrage on the noble and scientific
art of organ-building will not be encouraged.
Only slightly less objectionable and destructive of tonal powers
is the present growing system of borrowing stops from certain divi-
sions of an Organ to make up deficiencies in other divisions. This
method has the effect of making a poor and badly schemed stop-
apportionment look impressive on paper and on the draw-stops of
the Organ; and that is all that can be said in its favor. This un-
desirable and inartistic practice is carried to the extreme in the
formation of the Pedal Organ from manual stops; and is necessarily
destructive of clavier independence, effective tonal grouping, and
artistic registration. To show to what an extent this objectionable
practice is now being carried, we may state that lying before us as
we write is an organ-builder's Specification for an Organ, showing
an ostensible list of sixty speaking stops. The Pedal Organ of
twelve stops is entirely borrowed from the manual divisions, with
the exception of twenty-four pipes which strictly belong to it.
Comment is unnecessary.
Borrowing between manual divisions, if limited in extent and
done with well-considered scientific and artistic ends in view, and
without having the effect of crippling the independence of the con-
tributing divisions, may be productive of valuable aid in compound
tone creation and artistic registration; and, accordingly, may well
receive serious attention in scheming the stop-apportionments of
the tonal divisions of Organs of moderate dimensions. But unless
such borrowing is done with great judgment, care, and a full know-
*For further particulars, respecting "I/Orgue Simplified" and of a larger example of the
application of the "nouveau systSme d'orgues." see "The Art of Organ-Building," Vol. II..
PP 13-18.
12 ORGAN-STOPS
ledge of tone production, the result may be largely destructive of the
musical qualities of the instrument, as past experience has amply
proved.
The great value of Mutation harmonic-corroborating stops in
registration, for the creation of special compound tonalities and
orchestral coloring, must not be overlooked. We desire to impress
this important matter on organists generally, and especially on
those who may have the opportunity of scheming the tonal appoint-
ments of Organs. The importance of such stops is little understood
or realized in this country at the present time; simply because they
so seldom appear in the stop-appointments of the Organs that have
been turned out of our workshops.
For proofs of this statement, we may refer to the two large
organ schemes, already commented upon as seriously deficient in
compound harmonic-corroborating stops. They display a still
greater disregard of the important and tonally essential Mutation
stops. In the larger scheme, comprising upwards of two hundred
and seventy speaking stops, there are only three fifth-sounding
stops, one belonging to the 32 ft. harmonic series; one belonging
to the 16 ft. series; and one belonging to the 8 ft. scries. In the
lesser scheme, comprising, ostensibly, one hundred and one speak-
ing stops, there is only one Mutation stop, fifth-sounding, and
belonging to the 16 ft. harmonic series. Not a single one belonging
to the all-important 8 ft. series! We have no desire to be severe in
our comments: but to us it seems very astonishing that the cles^gn-
ers of these Organs intended to represent the latest advance on
old-fashioned tonal appointment should have so seriously over-
looked the important scientific and artistic offices of the Mutatio-i
stops. And the question naturally arises : Did they know, or simply
ignore, the necessity, on both scientific and artistic grounds, of
introducing third-sounding Mutation stops?
Again, by way of comparison and furnishing an example, we
may refer to the scheme for the Concert-room Organ given in our
work. In the two hundred and thirty-one speaking stops it pre-
sents, are included twenty-five Mutation stops, divided as follows :
Of the 64 ft. harmonic series there is one fifth-sounding labial stop;
of the 32 ft. series there are two labial and one lingual fifth-sound-
ing stops; of the 16 ft. series there are seven labial and two lingual
fifth-sounding stops; and of the 8 ft. series there are six fifth-sound-
ing and six third-sounding Mutation stops. The interested student
should note the introduction of the three lingual Mutation stops.
Again we say, large as the number is, there is not a stop too many
THEIR ARTISTIC REGISTRATION ij>
to furnish an accomplished performer with adequate means for
scientific and artistic registration and orchestral-tone production.
It is to be regretted that no definite rules of universal applica-
tion can be formulated to direct the organist in the all-important
art of registration; because, on the one hand, there are no univer-
sally adopted standards of tone in the stops bearing similar names,
produced by different makers, at different times, and under differ-
ent voicing conditions; and because, on the other hand, Organs
differ very widely in their stop appointments. So much so, that
what may be highly satisfactory and artistic registration on one
Organ, may be the reverse on another instrument.
Composers and writers of organ music render great assistance
to performers by indicating what to them is desirable and effective
registration for their works: but it must be remembered that such
registration has been formulated from the tonal material furnished
by some special Organ, at the command of the composer, or domi-
nant in his mind. It is, accordingly, questionable to what extent
such specific registration will be possible or satisfactory on other
Organs of different stop-appointments or stop-tonalities. Marked
differences in both directions are almost certain to obtain. It would
seem, under such conditions, that special artistic registration, to
properly render the composer's intentions, would be absolutely
called for on each Organ on which his works are to be performed.
We may close our brief remarks on this introductory branch of
our general subject with a few words, which may be of interest to
the organ student desirous of acquiring skill in the all-important art
of registration. In the first place, it is very desirable that the
student should form a clear and comprehensive concept of the tonal
structure of the Organ in its fullest scientific and artistic develop-
ment. This can only be properly arrived at by becoming familiar
with the special tonalities, combinational properties, and color-
values of all the stops which form that structure; and which are to
be found, or are available for introduction, in the Organ of to-day:
all of which are more or less fully commented upon in the following
pages, under their accepted names. It must be borne in mind,
however, that the names given to the stops found in modern Organs
do not invariably, or, indeed commonly, indicate their correct or
most desirable tonalities. This is simply due to the fact that there
are no universally recognized standards of tone in organ-pipes:
each organ-builder or voicer invariably adopting his own pipe-scales
and system of voicing; thereby producing his own peculiar or pre-
ferred quality of tone in each individual stop or family of stops.
I 4 ORGAN-STOPS
This diversity of tonal quality and color-value is very noticeable
in even the principal organ-toned stop the DIAPASON -the voice
of which varies, often widely, in Organs by different builders. In the
voices of both the labial and lingual imitative stops the difference of
tonality is generally very marked. In this class, one seldom hears
the same quality of tone in stops of the same name in any two
Organs.
Such being the case, it is very important that the student should
become thoroughly acquainted with the special tonalities and color-
values of all the stops which are comprised in the appointment of
the Organ on which he studies and performs: and that he should
follow up the knowledge so gained by a study of the tonal effects
produced by the combination of the stops of different colorings,
strengths of voice, and pitches; carefully noting their mixing quali-
ties and their harmonies of analogy and contrast, until he becomes
thoroughly familiar with the tonal resources of his instrument. This
somewhat exacting study, requiring time, thoughtful observation,
the exercise of memory, and not a little scientific and artistic cul-
ture, goes to prove, as we have already said, " there is no royal road
to learning."
TONALTIY OF ORGAN -STOPS
The stops which are at the disposal of the designers of modern
Organs are very numerous, as the present Glossary plainly shows;
and their tonalities display a diversity almost as great as that of
their names. Certain stops have distinct and pronounced voices,
which are either peculiar to them, or which belong to families formed
of members having difference of scale, slightly modified voicing, and
especially varied pitch. Others, less pronounced in their voices, are
more or less derivatives from special or more assertive stops,
Further, stops are either unimitative in their voices and strictly
belong to the Organ, their tones being unproducible by any other
instrument; or they are, in their voices, more or less closely imita-
tive of the tones produced by the string, wood-wind, and brass-
wind instruments of the orchestra. Under these conditions, two
grand divisions obtain in the tonal forces of the Organ ; one formed of
stops strictly belonging to the Organ proper, and the other of stops
which are imitative and orchestral in their tonalities. In th e modern
Organ, and especially in the Concert-room Organ, both divisions of
the tonal forces must be adequately represented, if the instrument is
to be sufficient for the artistic rendition of all classes of organ music,
THEIR ARTISTIC REGISTRATION 15
A careful study of the tonal forces which properly group them-
selves under the two grand divisions has led us to classify them into
eight subdivisions, each of which has a distinctive general tonality:
and this system of tonal grouping and subdivision will prove a valu-
able aid to the organ student in the study of registration and the
production of tone-color. The system may be presented in the
following form:
FIRST DIVISION SECOND DIVISION
Organ-Tone Orchestral-Tone
Unimitative Quality Imitative Quality
1. PURE ORGAN-TONE. i. ORCHESTRAL STRING-TONE.
2. FREE ORGAN-TONE. 2. ORCHESTRAL FLUTE-TONE.
3. FLUTE ORGAN-TONE. 3. ORCHESTRAL REED-TONE.
4. VIOL ORGAN-TONE. 4. ORCHESTRAL BRASS-TONE.
The eight different qualities of tone thus set forth may be
accepted as completely covering the tonal forces of the Organ of to-
day in its highest development; and the several classes of -stops which
yield them may, with advantage, and necessarily briefly, be alluded
to here.
It may, in the first place be pointed out that all the stops of the
Organ, so far as their formation is concerned, are of two kinds,
labial and lingual; the former being constructed of either metal or
wood pipes open, covered, or half -covered provided with mouths
which form their sound-producing portions, hence the term labial*
The lingual stops are also constructed of metal and wood pipes, the
resonators of which are either open or partially closed; their sounds
being produced by tongues, or languettes, which vibrate against
openings in reeds, or echalotes, hence the term lingual. Such stops
are confined to the production of the imitative orchestral reed- and
brass-tones. All other varieties of tone are produced by labial stops.
PURE ORGAN TONE. This is the foundation tone of the Organ
proper. It is sui generis, peculiar to the Organ, and cannot be imi-
tated by any other musical instrument. It is yielded in its purity
by the true English DIAPASON, 8 FT., formed of full-scaled open
*The objectionable term "flue" has been hitherto used by English and American organ-
parlance: but as the term flue pipe is strictly appropriate in connection with a stove or heating
apparatus, it certainly should be abandoned, along with other misnomers in organ-building
language, in favor of the proper and expressive term labial, A labial pipe is literally a mouthed
pipe; just as a lingual pipe is literally a iongued pipe.
i6 ORGAN-STOPS
metal pipes, unslotted, and voiced on a copious supply of wind of
moderate pressure. The tone is full, round, and dignified, singularly
free from harmonic upper partial tones; the absence of which leaves
a simple or pure tone, which, when used alone, is somewhat cloying
and unsatisfying to the ear familiar with the complex tonality of
orchestral sounds. Nevertheless, the DIAPASON'S tone has alway k s
been, and always will be, the unique and special glory of the Organ
a foundation on which to build compound tones of surpassing
grandeur and beauty.
In addition to the fundamental DIAPASON, 8 FT,, which strictly
belongs to the manual department, and especially to the Great
Organ, there is the DIAPASON, 16 FT., which furnishes the true
foundation tone of the Pedal Organ. This is, in its correct form, an
open metal stop, of large scale, constructed similar to the manual
DIAPASON, and, like it, yielding pure organ-tone. Under the name
DOUBLE DIAPASON, 16 FT., a similar stop, but of much smaller scale
and less powerful voice, is introduced in the manual department,
and, properly, in the First or Great Organ. This important stop
also yields pure organ-tone.
The early organ-builders found that unison stops alone, however
massed, produced tones of too monotonous and unsatisfying a
character; and to obtain relief and desirable variety they resorted to
the formation of ranks of pipes yielding higher and concordant tones,
This proceeding led to the early and the mediaeval Organs becoming
great MIXTURES of several unison-, octave-, fifth-, and third-sound-
ing ranks. The original introducers of this system of stop-appoint-
ment had no clear knowledge of the science of compound musical
sounds or their natural harmonic constituents ; they simply realized
the tonal value of the concordant ranks, and were content with their
introduction, unscientifically voiced as they were, and the musical
effects they produced. Since their introduction, such octave and
mutation harmonic-corroborating stops have never been omitted
from the appointment of the Organ. While the old European organ-
builders certainly introduced high-pitched harmonic-corroborating
stops and MIXTURES to excess, and voiced them unscientifically and
undesirably loud;* it does not follow that their formation and
introduction in the Organs of to-day should be neglected, in the
manner in which they are, by the unscientific organ-designers and
* An example may be given of this excess. In the Organ in the Monastic Church, at WeJn-
garten, Swabia, built by Gabler, of Ravemburg, in 1750, there are sixty-six speaking stops,
having 6,666 pipes, and of this number there are ten compound harmonic-corroborating stops,
having a total of 9$ ranks and 4,797 pipes. While these numbers are given on authority, the
number of pipes are probably not Quite accurate.
THEIR ARTISTIC REGISTRATION 17
labor-saving organ-builders of the present time; with a serious loss
to the musical resources and beauty of the modern Organ; and a
very undesirable narrowing of the opportunity for a wide exercise
of artistic and expressive registration, and the production of varied
tonal colorings which are unproducible in their absence.
The harmonic-corroborating stops, alluded to, directly asso-
ciated witr^the DIAPASONS, are, like them, formed of open metal
pipes, yielding pure organ-tones of graduated degrees of assertive-
ness, becoming perceptibly softer in tone as they rise in pitch, in
accordance with the natural laws of compound musical sounds.
These stops, accordingly, require to be properly scaled with refer-
ence to the scale of the fundamental DIAPASON, 8 FT., and voiced so
as to combine with, and be absorbed in, the prime tone; brightening
and richly coloring it without asserting undue prominence individ-
ually. It has been the neglect of the teachings of the natural laws
of compound musical sounds, either through ignorance or careless-
ness, that has rendered the single and compound harmonic-cor-
roborating stops in so many Organs unsatisfactory and of little
value in combination and artistic registration. It is, accordingly,
desirable that organists who take any interest in the Organ, beyond
merely playing it, should see that in instruments, over the tonal
appointment of which they have any control, a sufficient proportion of
properly formed and voiced harmonic-corroborating stops be
provided.
The more important associated harmonic-corroborating stops
are the OCTAVE, 4 FT., TWELFTH, 2% FT., FIFTEENTH, 2 FT., SEVEN-
TEENTH, i% FT., NINETEENTH, i J^ FT., and TWENTY-SECOND, i FT.
These stops are introduced in separate ranks of full compass, voiced
to yield pure organ-tone, and introduce and corroborate the first,
second, third, fourth, fifth, and seventh upper partial tones of the
foundation unisons, represented by the DIAPASON, 8 FT. To com-
plete the harmonic series it is desirable to add the FLAT TWENTY-
FIRST i l /7 FT., corroborating the sixth upper partial tone. This stop
is almost unknown, in a complete form, in English Organs, al-
though it has been used in certain breaks of rare MIXTURES. Under
the name SEPTIME, it appears as a separate and complete stop in
certain French Organs. In the Grand Organ in the Cathedral of
Notre-Dame, Paris, constructed by Cavaill6-Coll, there is a SEP-
TlfeME, i l /7 FT., in the Grand-Choeur, belonging to the foundation-
work; a SEPTI^ME, 2 2 /7 FT., in the Bombardes, belonging to the 16
ft. harmonic series; and a SEPTIEME, 4 4 /7 FT., in the P&lale, be-
longing to the 32 ft. harmonic series. We have not found an instance
1 8 ORGAN-vSTOPS
of the SEPTIEME being introduced in a French MIXTURE. The
higher-pitched harmonic-corroborating stops, owing to the smallness
of their pipes, cannot be carried throughout the compass of the
manual clavier without a break; accordingly, they appear only in
broken ranks of MIXTURES. These stops are the TWENTY-FOURTH,
4/5 FT., TWENTY-SIXTH, % FT., TWENTY-NINTH, }/> FT,, THIRTY-
THIRD, y$ FT., and THIRTY-SIXTH, Y^ FT., corroborating the ninth,
eleventh, fifteenth, twenty-third, and thirty-first upper partial tones
of the foundation unisons.
There are, in addition to the stops already named, certain others,
properly pure organ-toned, which belong to the series founded on
the DIAPASON, 16 FT., of the Pedal Organ. These are the OCTAVE,
8 FT., which is practically a DIAPASON, the TWELFTH, 51/3 FT,, and
the SUPER-OCTAVE, 4 FT., introducing and corroborating the first,
second, and third upper partial tones of the 16 ft. harmonic series.
Harmonic-corroborating stops of higher pitch are introduced in the
MIXTURES sometimes inserted in the Pedal Organ,
The stops, complete and incomplete, enumerated above are all
that practically belong to the foundation-work of the Organ ; and all
should be present in a completely appointed instrument. They are
necessary for the production of the entire range of pure organ-tones;
and are essential in effective tonal coloration and artistic registration.
FREE ORGAN-TONE. The few stops which may be properly
classed as yielding free organ-tone are strictly of mediate or transi-
tional character; that is, in addition to a foundation of pure organ-
tone they encroach, more or less, as it were, upon tones of some dis-
tinctive character; producing voices of different shades of tonal
coloring, never pronounced, and on this account of extreme value in
artistic registration. While the compound voices of such stops
cannot be classed as belonging to any special tonality, for they differ
greatly, they are invaluable in building-up and delicately tinting
other distinctive voices, by means of which the artist organist is
painting his musical picture.
Between pure organ-tone, as previously defined, and what must
be recognized as free or impure organ-tone, there is a very fine line
of demarkation. Immediately on this line stands the beautiful
English DULCIANA, 8 FT,, the tone of which is only distinguishable
from pure organ-tone by its delicate, silvery, singing quality, due
to the presence of almost imperceptible harmonic over-tones, Like
the foundation DIAPASON, the DULCIANA is properly attended by its
family of harmonic-corroborating stops, forming very lovely com-
THEIR ARTISTIC REGISTRATION 19
pound tones, the delight of the tonal colorist. The most useful of
these stops are the OCTAVE DULCIANA, 4 FT. ; DULCIANA TWELFTH,
2% FT.; SUPER-OCTAVE DULCIANA, 2 FT.; and DULCIANA CORNET.
There are several other stops yielding free organ-tones of a more
pronounced mediant character, in which pure organ-tone is more
absorbed and a richer coloring imparted. Among these may be
named the DOLCAN, yielding a voice of a refined and somewhat
plaintive character; and the KERAULOPHONE, yielding a richer tone,
which, in good examples, approaches a horn-like quality, extremely
valuable in the production of rich and quiet colorings. This fine
stop is rarely introduced in Organs built to-day. Still more pro-
nounced in tone are the stops known as the HORN DIAPASON and
BELL DIAPASON, the voices of which are rich colorings on grounds of
pure organ-tone. Certain stops yielding what must be classed as
free organ-tone have their voices comparatively rich in harmonic
upper partial tones. Prominent among these stops is the GEMS-
HORN, the pipes of which, Helmholtz says, have the property of ren-
dering some higher partial tones comparatively stronger than the
lower; hence its peculiar value in registration in which brightness
is desired, without a cutting quality which is so frequently destruc-
tive of refinement and repose. Perhaps such terms as these are
rarely used in speaking of organ registration, but they will be under-
stood by the artist organist and the musician, and their meaning
should be grasped by the organ student.
FLUTE ORGAN-TONE. By the term flute organ-tone is signified
that wide and varied range of flute-like tones, produced by several
varieties of organ-pipes, which are not strictly imitative of the clear
and penetrating voices of the Flutes of the orchestra. Some of these
unimitative tones differ widely from those of the orchestral Flutes,
while others are so closely allied as to be almost undistinguishable
from them. It is in this great range of flute-like voices that the
organist finds a richly spread palette for the production of innumer-
able beautiful tonal colors. The organ-toned FLUTES are not only
widely varied in strength and character of voice, but they, of all the
stops of the Organ, lend themselves most readily and efficiently to
effective registration, combining perfectly with the stops of every
other tonality, both labial and lingual.
The stops yielding flute organ-tone may be divided into four-
families, each of which produces a characteristic tonality. These
are formed, respectively, of open, covered, half-covered, and har-
monic pipes. The variation of their voices is mainly due to the
20 ORGAN-STOPS
presence of different groups of harmonic upper partiuls and their
relations to the prime tones. As all the stops just alluded to are
described in the Glossary, it is unnecessary to go into particulars of
their^special voices in this brief essay; but the names of a few of the
representative ones may be mentioned. Of the open family, may be
named the TIBIA PLENA, HOIILFLOTK, SiTran/rrK, and WALD-
FLOTE; of the covered family, the TIBIA CLAUSA, DOPPKLFLOTK, and
LIEBLICHGEDECKT; of the half-covered family, the ROURKLOTK and
FLUTE A CHEMINE; and of the harmonic family, the French FuOrii
HARMONIQUE.
So numerous and varied are the voices of the unirnitative flute-
toned stops of the modern Organ, and so valuable are they in artistic
registration with the voices of all other tonalities, that the organist
should make a careful study of them; acquainting himself with all
their coloring properties and their peculiar mixing and separating
qualities. It must be observed that certain flute-tones hold them-
selves distinct and remain assertive; while others become absorbed
and completely lose their individuality in the compound tones.
VIOL ORGAN-TONE. Although the stops yielding unimitative
string-tone, which, for the sake of distinction, we have designated
viol organ-tone, are comparatively few in number, they occupy an im-
portant place in the stop-appointment of the Organ, and a prominent
place on the color palette of the organist skilled in artistic registration.
Standing on the border-line between pure organ-tone and viol
organ-tone is the refined and beautiful voice of the true SALICIONAL,
The most important stop of this tonality is the VIOLIN DIAPASON
or GEIGENPRINCIPAL. The tone of this fine stop varies in different
examples, according to the taste of the voicer; but it should be dic-
tated by the position it holds in the tonal scheme of the Organ, and
how it can be of the greatest value in effective registration. While
in this stop the string tonality is, properly, not more than the pure
organ-tone, there are others, under different names, in which the
string-tone is the more assertive; and these are very desirable, in
Organs of large dimensions, for registration with labial and lingual
stops of powerful intonation. In the Concert-room Organ, they are
required for the building-up of the necessary volumes of string-tone.
There are certain other stops which may be classed as yielding
viol organ-tone, but which may claim to be imitative, in so much
that they are accepted as reproducing the tones of the obsolete
Viola da Gamba and Viola d'Amore. But when their tones are
compared with those of the modern stops which imitate so closely
THEIR ARTISTIC REGISTRATION 21
the assertive and complex tones of the Violin, Viola, and Violon-
cello, their timid voices seem to fall back into what may be correctly
considered viol organ-tone. When these desirable stops appear in
an Organ they deserve the careful attention of the organist.
ORCHESTRAL STRING-TONE. This is the first and most essential
of all the orchestral imitative tones of the modern Organ; and may
almost be said to be the foundation of the tonal structure of the
true Concert-room Organ; just as the string forces are the founda-
tion of the grand orchestra. Obvious as this must have been, one
would think, to every thoughtful musician and accomplished or-
ganist, it is truly remarkable that such a self-evident fact remained
unacknowledged and unacted upon in the organ-building world,
cramped by old-fashioned tradition, until we instituted a separate
and complete string division, comprising eighteen ranks of string-
toned pipes, including full harmonic elements, and endowed with
full and special powers of expression and flexibility, in our scheme
of the Concert-room Organ installed in the Festival Hall of the
Louisiana Purchase Exposition, St. Louis, Mo., 1904.* We feel very
proud of having been the first in the history of the art of organ-
building to realize the necessity for, and to carry into effect, so im-
portant and art-serving a step in advance in tonal appointment ; now
becoming tardily acknowledged by organ-builders and organists as
essential in every Organ suitable for recital purposes, and which has
any pretension toward completeness.
The stops are comparatively few which strictly fall under the
class now under consideration, being primarily the VIOLIN, VIOLA,
VIOLONCELLO, and CONTRABASSO, which also pass under other equiv-
alent names. These stops are, however, attended by certain deriv-
atives which are required in the Organ for the production of well-
known orchestral effects, and for purposes of artistic registration.
The derivatives are those which represent the orchestral instru-
ments played under characteristic conditions; namely, con sordini
and vibrato. It is, perhaps, unnecessary to remark that the great
value of the orchestral string stops lies in the closeness and beauty
of their imitative voices and their perfect regulation; but such
artistic perfections are rarely the products of the competitive and
hurried organ-building of to-day. The cry is for quick production,
not artistic and painstaking work; and no class of stops suffers more
*See Specification of this Organ given on pages 503-8; and our scheme for the Ancillary
String Organ, of twenty-three ranks of pipes, in the Concert-room Organ on page 323. of our
work, "The Organ of the Twentieth Century/' New York, 1919.
22 ORGAN-STOPS
from this than that which forms the orchestral string-tone forces of
the Concert-room Organ.
Notwithstanding that the foundation orchestral string-toned
stops are in themselves very rich in harmonic upper partials, so
much so as to practically require no additions in purely solo work:
yet for the production of full orchestral effects, it is necessary to
have for registration string organ-toned harmonic-creating and cor-
roborating ranks of pipes, belonging to both the 8 ft. and 1 6 ft. series,
and including an effective OCTAVE VIOL or VIOLETTA, 4 FT. These
afford great opportunities for tonal coloring and registration of a
high order, leading to the production of volumes of brilliant and
expressive tone, hardly dreamt of in the Organ of to-day, and only
to be surpassed by the full string forces of the grand orchestra.
When will the distinguished and respect-commanding musician
arise to teach the thoughtless organist and organ-builder what the
Concert-room Organ of the twentieth century must be?
ORCHESTRAL FLUTE-TONE. Little need be said respecting the
tonality of the stops yielding orchestral flute-tone. They are only
three in number; namely, the ORCHESTRAL FLUTE or FLAUTO
TRA VERSO, 8 FT., the PLAUTO TRA VERSO, 4 FT, and the PICCOLO,
2 FT. ; and their voices successfully imitate, especially in the finer
examples, those of the Flutes and Piccolo of the orchestra, and ably
represent them in the tonal appointment of the Organ. Their
voices are refined and penetrating having few assertive upper
partials except the octaves, naturally subordinated; and they are of
extreme value in both solo-work and artistic registration. Seeing
the great importance that has been given to the Flute by the great
composers, who frequently gave it the leading wind part in their
scores, there can be no excuse for the insertion of an indifferent
ORCHESTRAL FLUTE in any important Organ. In registration with
the orchestral string-toned stops the ORCHESTRAL FLUTES produce
beautiful qualities of tone. In general registration, however, they
do not stand preeminent, for there are several of the flute organ-
toned stops that are equal, if not superior, in imparting effective
coloring.
ORCHESTRAL REED-TONE, The important instruments of the
grand orchestra which are more or less closely imitated by the stops
of the Organ yielding orchestral reed-tone are the Oboe, Clarinet,
Corno di Bassetto, Cor Anglais, Bassoon, and Saxophone. The
stops bearing these names, or their equivalents in other languages,
THEIR ARTISTIC REGISTRATION 23
are all of unison pitch on the manuals; but the DOUBLE BASSOON
or CONTRAFAGOTTO, 1 6 FT., a stop of great value, is also placed on
the manuals in important instruments.
The OBOE appears in two forms, the one commonly introduced
not being strongly imitative in its tonality; indeed, it has been
found very difficult to imitate the "small acid-sweet voice, having a
pastoral character, full of tenderness, 1 ' as Berlioz cleverly de-
scribes it. Certain attempts have been made to imitate this charac-
teristic voice, and results of a fairly satisfactory nature have been
obtained from stops of special formation, named ORCHESTRAL
OBOES. The Oboe of the orchestra is essentially a melodial instru-
ment, and the ORCHESTRAL OBOE of the Organ must also be ac-
cepted as a melodial or solo stop, which is comparatively of little
value in registration save for the production of very unusual effects
of a pathetic character.
Much more closely imitative is the CLARINET, 8 FT. ; indeed,
when made by a master-hand, it may be pronounced the best rep-
resentative of an orchestral reed instrument to be found in the
Organ. It is especially satisfactory when associated with a unison
covered stop, such as a soft-toned DOPPELFLOTE or a LIEBLICH-
GEDECKT, 8 FT. These voices impart the full and somewhat hollow
tonality of the orchestral instrument in its best register. The tube
of the orchestral Clarinet is of the nature of a covered pipe, pro-
ducing, in addition to the prime tone, the second, fourth, sixth, and
higher even upper partial tones. Such being the case, it can be
realized that the tone of the lingual CLARINET derives considerable
richness and increase of character by combination with the tone of
a covered stop, which has in its voice the same progression of har-
monic upper partials. This affords a simple lesson in scientific
combination. The peculiar quality of the tone of the CLARINET
renders it a valuable addition to the voices of almost all the labial
stops of medium power, producing compound tones of rich and
varied colorings: accordingly, in registration, the stop deserves
greater attention than seems to have been hitherto bestowed
upon it.
More effective and fuller in tone-creation is the CORNO DI BAS-
SETTO, 8 FT., which belongs to the CLARINET family. The orchestral
instrument from which it derives its name, and of which it is the
organ representative, is practically a Tenor Clarinet having the
compass from FF to c 3 . The voice of the Corno di Bassetto is fuller
and more reedy than that of the Alto Clarinet; and, accordingly,
the tone of CORNO DI B ASSEtTO is bolder and richer than that of the
24 ORGAN-STOPS
CLARINET, 8 FT. Properly made and voiced by a master-hand, the
CORNO DI BASSETTO is of greater value than the CLARINET in effec-
tive registration. Unfortunately it is a stop rarely found even in
large Organs. While it is properly of 8 ft. pitch, the insertion of a
CORNO DI BASSETTO, 16 FT. is greatly to be desired in the expressive
"wood-wind" division of the Concert-room Organ; where it would,
in combination with the unison CORNO DI BASSETTO and other
wood-wind stops of the Organ, be productive of compound tones
absolutely unknown in the old-fashioned tonal-appointments of the
Organs of to-day.
The Bassoon of the orchestra is a double-reed instrument like
the Oboe, of which it is considered to furnish the proper bass. Its
representative in the Organ, the BASSOON or FAGOTTO, 8 FT., fails
like the OBOE, in yielding a perfectly satisfactory imitative voice.
Nevertheless, the tone of a really good example is extremely valu-
able in artistic registration and tonal coloration; so much so, that
no Concert-room Organ can be considered complete in which the
stop does not appear in the manual department. Its proper place
is in the "wood-wind" division of the instrument.* The orchestral
Bassoon has a conical bore of about eight feet in length, regularly
increasing in diameter from -3% inch at the reed end to I % inches at
the open end. This form of tube generates or favors the production
of the uneven upper partials (natural to an open organ-pipe), and,
accordingly, yields a compound tone essentially different from
that of the Clarinet, while it partakes of that of the Oboe. The
voice of the BASSOON, 8 FT., while it closely resembles that of the
orchestral instrument, has the advantage of being more uniform in
color throughout its compass, and in being entirely free from the
grotesqueness which characterizes certain of its lower notes.
The tonal value of the Double Bassoon or Contrafagotto has
long been recognized by the great composers; Beethoven, Haydn,
Mozart, and Mendelssohn having introduced it in their great works.
As Dr. W. H, Stone remarks: "In all cases it forms a grand bass
to the reed band, completing the 1 6-foot octave with the six lowest
notes wanting on three-stringed Double Basses. J> These facts are
enough to show the value of, and the necessity for, the DOUBLE
BASSOON or CONTRAFAGOTTO, 16 FT., which represents the orchestral
instrument in a properly-appointed Organ. Its presence in an im-
portant Church Organ is highly desirable; while in a Concert-room
Organ it is imperative. The value of its voice in tonal coloration,
*See "The Organ of the Twentieth Century, " Pages 307 and 505,
THEIR ARTISTIC REGISTRATION 25
and in the registration of grave tones is so great, that in the Con-
cert-room Organ it may, with advantage, be introduced in more
than one manual division; preferably in the expressive subdivision
of th, First or Great Organ, and in the Organ containing the stops
representing the wood-wind instruments of the orchestra. In the
latter it may be of the more assertive variety, distinguished by the
name FAGOTTONE, 16 FT. It is never desirable to have two stops
of precisely similar tones in an Organ, however large it may be.
Variety of tone is an essential element in artistic registration.
When properly made and artistically voiced, the DOUBLE
BASSOON yields a tone of the same quality as that of the BASSOON,
8 FT., but of rather a greater body. Accordingly, if the tone of the
middle c r (4 ft.) pipe of the former is carefully compared with that
of the tenor C (4 ft.) of the unison stop, it will be observed that
while they are of the same pitch and character, they are different in
tonal value, the unison being somewhat lighter and brighter the
voicing having properly generated one or two higher harmonic
upper partials. The true orchestral relationship is thereby
established.
Judging from what one observes in the unscientific and inartistic
competitive organ-building of to-day, organ-builders certainly will,
and many organists probably will, consider such refinements as
have been advocated in the preceding remarks quite unnecessary.
Certainly they will call for scientific and artistic culture and a high
sense of duty on the part of the former; and for a keen appreciation
of tonal values on the part of the latter. But the obvious absence
of such tonal refinements only goes to prove that the culture alluded
to is much to be desired on both sides.
It is unnecessary to comment at length respecting the tone of
the COR ANGLAIS, 8 FT., which in its present state imitates as closely
as seems practicable that of the orchestral Cor Anglais; which is in
reality an Alto Oboe, bearing about the same relation to the ordi-
nary Oboe as the Corno di Bassetto does to the Clarinet. Regarding
the orchestral instrument, Berlioz remarks: "Its quality of tone,
less piercing, more veiled, and deeper than that of the Oboe, does
not so well as the latter lend itself to the gaiety of rustic strains. . . .
It is a melancholy, dreamy, and rather noble voice." The organ
COR ANGLAIS, in its best form, yields a tone which has very valuable
mixing and coloring properties; and these, combined with its some-
what subdued or " veiled " voice, renders the stop highly suitable
for the chief accompanimental divisions of both the Church and
Concert-room Organs. In such a position it will prove of the maxi-
26 ORGAN-STOPS
mum service in artistic registration. While it would seem to belong
to the "wood-wind" division, the kindred nature of its voice to that
of the OBOE renders its presence there unnecessary. In an accom-
panimental division it is most desirable to have a considerable range
of different tonalities, so that by skillful registration expressive
accompaniments of all tones and colors may be formed.
The latest addition to the orchestral reed-tone forces of the
Organ is the SAXOPHONE, 8 FT., a fine stop of the CLARINET family.
The Saxophones of the orchestra are single-reed instruments, having
wooden mouthpieces, similar to those of the Clarinets, attached to
conical tubes of brass. They are of several keys, covering a compass
of from BBB to c 3 , one octave short of the present full manual
compass of the Organ. The lingual SAXOPHONES made up to the
present time have only been moderately successful in their imitative
tones, while otherwise fine stops. The most satisfactory imitative
stop hitherto produced, so far as our knowledge extends, is a labial
one formed of wood. The character and value of its voice may be
in some way realized from the following description by Berlioz of
the tones of the orchestral Saxophones, which he states as possess-
ing, "Most rare and precious qualities. Soft and penetrating in the
higher part, full and rich in the lower part, their medium has some-
thing profoundly expressive. It is, in short, a quality sui generis,
presenting vague analogies with the sounds of the Violoncello, of the
Clarinet, and Corno Inglese, and invested with a brazen tinge
which imparts a quite peculiar accent." Dr. Stone comments on the
peculiar Violoncello quality of the tone ; he says : " It reproduces on a
magnified scale something of the Violoncello quality, and gives great
sustaining power to the full chorus of brass instruments, by intro-
ducing a mass of harmonic overtones," Prom these particulars it
can be readily realized how valuable such a compound tone must
prove in the Organ, and, at the same time, how difficult its produc-
tion must prove from a single lingual stop. The prominence of the
Violoncello quality clearly points to the labial stop formation in
which a string-tone can, with skillful voicing, be produced accom-
panied by a pronounced reedy quality.
ORCHESTRAL BRASS-TONE. The lingual stops of the Organ,
which produce tones imitative, more or less closely, of those of the
brass wind-instruments of the orchestra and band, are not numerous
but of very great importance. To them are added brass-toned stops
which belong exclusively to the Organ, and which have tonal powers
and compasses beyond those possible in instruments blown by the
THEIR ARTISTIC REGISTRATION 27
human breath; and these impart an element of impressiveness and
grandeur, when properly used, which alone is sufficient to stamp the
Organ the Monarch of all Instruments.
The brass instruments of the orchestra which have representa-
tives in the Organ are the Trumpet, Horn, Trombone, and Ophi-
cleide. Of these the Horn is the instrument of which a satisfactory
tonal equivalent is most desirable in the Organ; but which it has
been found, up to the present time, practically impossible to produce.
Berlioz calls the Horn a " noble and melancholy instrument," and its
peculiar tonality supports that description. Although its tones are
produced from a large brass instrument, sounded by the lips and a
cupped mouthpiece, they have properly no trace of clang and brassi-
ness. The Horn has two series of sotinds, known as the open and
closed; and it is the imitation of the latter, also called the "hand
notes, " which the pipe-voicer has found the crux, in the production
of the organ HORN, 8 FT, Certain painstaking organ-builders have
prodticed stops which, though not strictly imitative in all respects,
are fine tonally, and extremely valuable in refined registration with
string-toned stops. The HORN properly belongs to the "brass-
wind" division of the true Concert-room Organ; where it is useful
as a solo and combinational stop. Every exertion should be made
by the organ-designer to obtain a satisfactory HORN; for it is ab-
solutely essential in the artistic rendition of orchestral scores.
The brass wind-instrument of the orchestra most commonly
represented in Organs of all classes is the Trumpet, and its normal
tone is more satisfactorily imitated by a lingual stop than that of
any other brass instrument. The desirable tone for the ORCHESTRAL
TRUMPET, 8 FT., is that of the Trumpet played mezzo forte, yielding
a bright silvery voice with just sufficient brassiness to give it true
character. This tone under expression is valuable in solo effects
and in general registration, mixing perfectly with tones of all classes.
The stop to be preferred for the Solo Organ is the more powerful
HARMONIC TRUMPET, which represents the orchestral Trumpet
played fortissimo. This powerful and penetrating stop must be
placed under perfect control. A TRUMPET should never be placed
en chamade, as in the Organ in the Church of Saint-Ouen, Rouen,
and in several large Spanish Organs. To imagine any orchestral
instrument, or its organ representative, played without expression,
could only be possible in the brain of a musical ignoramus.
The Trombone of the orchestra is a member of the Trumpet
family, and, indeed, it may be properly considered as furnishing the
bass to the orchestral Slide Trumpet. Its tone is, accordingly, an
28 ORGAN-STOPS
enlargement of that of the Trumpet, capable, on account of the
important" size of the instrument which produces it, of great power,
brilliancy, and brazen clang. The three Trombones alto, tenor,
and bass along with the Trumpet, form the only complete enhar-
monic wind quartet in the orchestra. These instruments are per-
fectly true in their intonation ; in this respect rivaling the stringed
and bowed instruments.
The organ representatives the TROMBONE, 8 FT., and CONTRA
TROMBONE, 16 FT., are fine and valuable stops when voiced by
master-hands; but they cannot be considered perfectly satisfactory
tonal equivalents of the orchestral instruments. There is great
difficulty in producing the characteristic brazen clang in lingual
pipes, without objectionable coarseness and reedy clatter; but,
perhaps, this is not to be regretted, for the more desirable tones of
the orchestral Trombones are those in which this extreme clang is
absent. The TROMBONES of the Organ are of great importance,
especially in the Concert-room Organ, and their tonalities and
strengths of voice should in all cases be dictated by the places they
occupy and the offices they have to fulfill in the tonal economy of the
Organ. Although the use of the TROMBONES is somewhat limited in
artistic registration, their employment is of supreme value in the
proper rendition of certain classes of orchestral music, and in
the production of impressive effects and in full organ passages. The
proper position of the TROMBONES in the properly stop^apportioned
Concert-room Organ is in the expressive division chiefly devoted to the
representatives of the "brass-wind" forces of the grand orchestra.*
The Ophicleide of the orchestra is a large key instrument of the
Bugle family, being practically the bass correlative of the Key or
Kent Bugle. The usual Ophicleide is of 8 ft. pitch, extending only
one semitone below the compass of the Violoncello: but as a " Con-
trabass Ophicleide " has been used in the orchestra (at the Musical
Festival in Westminster Abbey and the Birmingham Festival, both
in 1834), there is good authority for a representative of this im-
pressive instrument, of 16 ft. pitch, finding a place in the Concert-
room Organ of the twentieth century. The tone of the Ophicleide is
broad and dignified, so much so, that Wagner found it necessary to
fill its place in the orchestra by the Bass and Contra Tubas. Men-
delssohn introduces the Ophicleide freely in the scores of his impor-
tant works, notably in the " Elijah" and the "Midsummer Night's
Dream" music.
*See "The Organ of the Twentieth Century, Pages 311-14 and 506.
THEIR ARTISTIC REGISTRATION 29
As the Ophicleide has for many years disappeared from the
orchestra, it is practically impossible for voicers of to-day to realize
the tone they are called upon to reproduce in the organ representa-
tive. Some idea may, however, be gathered from the tonality of the
substitutes Wagner adopted. The Tuba is a large bass instrument of
the Saxhorn family, the tone of which is smooth and rich when
produced by a master. It is such a tone, not too assertive, and
entirely free from brassiness, that is required in the OPHICLEIDE
employed in the performance of orchestral scores. The OPHICLEIDE,
8 FT., of the ordinary modern Organ is a widely different class of
stop, yielding a powerful and strident tone. It is voiced on wind of
high pressure, as in the representative example in the Solo of the
Organ in St. George's Hall, Liverpool, which speaks on wind of 22
inches: and as this dominating stop has very foolishly been planted
. on an uninclosed wind-chest, and is, accordingly, devoid of powers
of flexibility and expression, it is valueless in artistic registration.
Of the tones of the band brass instruments the Cornopean,
Euphonium, and Tuba and their organ representatives; and of
those of the several brass-toned stops which are practically unimi-
tative, and strictly belong to the Organ, it is unnecessary to enlarge
here: all useful particulars will be found, under their respective
names, in the following Glossary.
DO MATERIALS AFFECT THE TONE OF ORGAN-
PIPES?
A question of no little importance has frequently been asked, but
never, so far as we are aware, has it been satisfactorily answered.
The question is : Does the material of which an organ-pipe is con-
structed affect the quality of its tone ? We have, by long observation
of the behavior of organ-pipes, and some special experimenta-
tion, given the question careful and thoughtful consideration; yet
we are far from being prepared to be dogmatic upon it. The only
answer would seem to be this paradoxical one: It does and it does
not.
We know that the old German, Dutch, and French organ-
builders were very careful about the metals and woods they used
for their pipes; and that their finest qualities are very rarely em-
ployed in modern organ-building : but whether they were selected on
account of their value in tone production, or on the score of their
durability only, it is impossible to decide. We do know, on the one
hand, that their durability has been remarkable; and, on the other
30 ORGAN-STOPS
hand, that the tones of the pipes formed of them are very beautiful.
Take for example the Organ in the Cathedral of Haarlem, built in
1738, All its displayed pipe-work including the Pedal SUB-PRINCI-
PAL, 32 FT., is of pure Cornish tin; while all the rest of the metal
pipes in the instrument are of an alloy of equal parts, by weight, of
Cornish tin and pure lead. All these pipes one hundred and
eighty-two years old are as good as the day they were made. For
extreme purity and beauty of tone we have only to go to the Kilber-
mann Organs in the Cathedral and churches of Strasbourg, the
pipes of which are formed of similar high-class materials. We
know, from examination of existing pipes of wood and metal, that
the English organ-builders of the seventeenth century used Cornish
tin and fine woods in their fabrication. Bernard Smifli seems to
have invariably used wainscot for his STOPPED DIAPASONS and
GEDACKTS, as specified for his Temple Church Organ (1688),*
Examples of these which have been preserved to our time are char-
acterized by great beauty of tone. Even in quite recent times we
find, in the construction of the Organs in the Cathedral and St.
Pauls-Kirche, Schwerin, the organ-builders Ladegast and Priese
using oak, maple, pear-tree, and mahogany in the formation of their
pipe-work. While it is quite reasonable to suppose that these cele-
brated builders used these special woods on account of their dur-
ability, and to favor perfect workmanship, it is still more reasonable
to suppose that they were employed for a still higher reason, namely,
that they were found to be conducive to the perfection of voicing
and the production of beautiful tones. Where do we find such fine
woods used in the pipes made for our organs to-day?
It is well known that the column of air within a pipe, while
speaking, is in a state of extreme pulsation or tremor, especially so
in pipes blown by winds of high pressures ; and it is obvious that this
tremor must be conveyed to, and properly resisted by, the walls of
the pipe, be it of metal or wood. By careful observation we have
found that the maximum effect of this pulsation is on the sides of a
pipe, not on its front in which the mouth is formed, or back, This
would point to the necessity of specially fortifying the sides of wood
* Dr. Burney, speaking of this master, says: " I have been assured by Snetzler, and by the
immediate descendants of those who have conversed with Father Smith, and seen him work,
that he was so particularly careful in the choice of his wood, as never to use any that had the*
least knot or flaw in it ; and so tender of his reputation, as never to waste his time in trying to
mend a bad pipe, cither of wood or metal, so that when he came to voice a pipe, if it had any
radical defect, he instantly threw it away, and made another, This, in a great measure, account**
for the equality and sweetness of his stops, as well as the soundness of his pipes to this
day,"- "History of Music."
THEIR ARTISTIC REGISTRATION 31
pipes, and imparting a sufficient power of resistance to the walls of
metal pipes. The former requirement would be met by the use of a
hard wood of sufficient thickness : and for perfection in voicing, and
the production of pure and beautiful tones, a close-grained hard
wood should be used for the mouth fronts of pipes of moderate size,
and for the mouths of large pipes. In the case of a metal pipe, a
metal or alloy of a firm body and sufficient thickness to withstand
the pulsation of the internal column of air is absolutely necessary
for the production of a satisfactory tone. Tin has been found to be
suitable in all respects, being much firmer and more resisting than
the usual alloys, cast in sheets in the ordinary way which leaves
them soft and undesirably open in structure; but tin is much too
expensive for general use; and the same may be said "of the high-
grade alloys of tin and lead. What has long been wanted is a pipe-
metal having all the good qualities of tin, without its prohibitive
cost; and this has at last appeared in the Hoyt Two-ply Pipe-metal
a product, not of the open casting-table, but of the rolling-mill.
The body of this compound pipe-metal is, in its perfected form, an
alloy of pure lead, tin, and copper, on which is laid a substantial
layer of pure tin. The combination is rolled, under great pressure,
into sheets, perfectly uniform in thickness, remarkably tough, and
of great resisting properties. These sheets are furnished in twenty-
two standard thicknesses, regularly graduated from 0.015 to 0.120
of an inch. Both surfaces are perfectly smooth and polished, giving
the interior of a pipe the surface most conducive to the production of
refined intonation. From severe tests we have made of its several
properties, we have no hesitation in saying it is superior to the
ordinary cast spotted-metal so long used in high-class pipe-work.
It has everything in its favor, including a moderate price.*
Beyond what has already been said regarding the importance of
using fine hard woods and firm and tremor-resisting metals in pipe
formation, little seems to have been decided respecting their more
direct influence on tone-production: and the deeper one goes into
the question the more complex it seems to become. We must admit
that we have heard tones in every way satisfactory yielded by pipes
formed of nearly all classes of pipe metal, from the deadest alloys
of lead and antimony to the richest alloys of tin and lead, and even
tin itself, and have been compelled to acknowledge that, after all
is said, it would appear that the skill of the voicer was the dominat-
ing factor. As we have said elsewhere: although in general practice
* For further particulars and Table of standard thicknesses and weights of the Hoyt Two-
ply Pipe-metal, see " The Organ of the Twentieth Century," pp. 345, 346.
32 ORGAN-STOPS
certain classes of pipes appear to be most satisfactory in tone when
formed of tin or high-class spotted-metal, their tonal character and
excellence depends chiefly on their correct scaling, proper thickness
of material, perfection of formation, the pressures of wind used, and,
above all, on their skillful and artistic voicing. It is largely on the
score of strength, rigidity, and durability that high-class materials
are essential in organ-pipe construction.
GLOSSARY OF ORGAN-STOPS
A
ACUTA. Vox ACUTA (from Lat. acutus sharp). Ger,,
AKUTA, SCHARF. Dtch., SCHERP. A compound harmonic-corrob-
orating stop, composed of three or more ranks of open metal labial
pipes, preferably of small scales. All the ranks are high-pitched and
voiced to yield bright and penetrating tones hence the name of the
stop. When in its best form it has a third-sounding rank which adds
greatly to its acute tonality. As the pipes forming the stop are
necessarily very small, the ranks will have to break three or four
times in the manual compass of five octaves. The following are
examples of the composition of stops of three and four ranks:-
ACUTA III. RANKS,
CC to B 22 24 26.
C 1 tO b 1 17 19 22.
c 2 to b 3 15 17 19-
C3 tO C< 12 15 17.
ACUTA IV. RANKS.
CC to BB 24 26 29 33.
C to B 22 24 26 29.
C 1 tO b 2 19 22 24 26.
C3 tO C4 12 15 17 19.
In compound stops of the CORNET class, in which third-sounding
ranks are introduced, it is desirable to subdue such ranks so as to be
less assertive than the octave- and fifth-sounding ranks; but the same
practice should not be followed in voicing and regulating the ACUTA,
because the sharpness given by the third-sounding rank is an impor-
tant element in its characteristic tonality. The standard rule which
dictates that all compound harmonic-corroborating stops must be
gradually softened in tone as they rise in pitch has to be observed in
the regulation-of the ACUTA.
3 33
34 ORGAN-STOPS
The ACUTA is only required in Concert-room Organs ot" the first
magnitude, and even in them it has been seldom introduced. One
of four ranks is in the Great of the Organ in the Cincinnati Music
Hall; but in it there is no third-sounding rank.
COMBINATION AND REGISTRATION. As the ACUTA, when prop-
erly made, is strictly a member of the fundamental unison har-
monic structure based on the DIAPASON, 8 FT., it properly enters
into numerous combinations either directly with that stop in full
harmonic sequence; or with it and other stops in varied registrations,
its presence being desired on account of its special brightness and
life-giving character. The ACUTA is specially valuable in registra-
tion with full-toned lingual stops such as are properly inserted in
the First or Great Organ. With the TRUMPET alone or with the
TRUMPET and CLARION it is valuable, imparting great brilliancy and
a singular orchestral coloring to the brass-tones, by powerfully cor-
roborating the higher harmonics present to some extent in the voices
of the lingual stops.
. Lat.,^EoLiNA. Ger., A.OLINE. Fr., UOUNE. Ital.,
EOLINA. The name employed by different organ-builders to desig-
nate extremely soft-toned stops both lingual and labial. Seidel de-
scribes it as a lingual stop voiced in imitation of the ^Eolian Harp.
He adds: " The bodies of the pipes are very small and of a narrow
measure. . . . The stop cannot be used by itself, but only in
combination with some soft 8ft. covered or open stop of narrow
measure, such as the GAMBA." This definition is supported by
Hamel, who remarks:
"C'est un jeu d'anches libres qui, ainsi que son nom 1'indique, doit imiter le
munnure de la harpe aeolienne, et qui, par consequent, doit avoir une intonation
extremement tendre et adricnne. Le corps dcs tuyaux qui sonnent quelquefois le
seize pieds, sont tres-petits et d'un diapason tres-dtroit. On trouvc ce jeu dispose*
avec des huit pieds dans le nouvel orgue (le Sainte-Marie & Wisrnar,"*
The ^EOLINE of the old builders was evidently a free-reed stop
furnished with small resonators. Topfer describes it as a free-reed
stop of 1 6 ft, and 8 ft, pitch, either like the PHYSHARMONICA or fur-
nished with small conical tubes or resonators* Walcker has placed
on the Second Manual of his Organ in Riga Cathedral an ^ELODICON,
1 6 FT., a lingual stop of the ^EOLINE class; and on the Fourth Man-
ual an ^EOLINE, SFT., a labial stop of tin, with the bass octave of
wood, as in the Third Manual of his Organ in St. Petri-Kirche in
Hamburg. Speaking of the ^EOLINE, Carl Locher, of Berne, re-
* ** Manuel Complet du Facteur d'Orgues."
THEIR ARTISTIC REGISTRATION 35
marks: "It is of a soft string-toned character, occurring in Ger-
many and Switzerland on almost all large and small Organs as an
8-ft. solo stop. It is considered the most delicate of all string-toned
stops."*
FORMATION. The MOLINE, properly of eight feet pitch, in its most approved
modern form may be classed in the SALICIONAL family, of which it forms the
softest-toned member. If voiced to yield a more decided string-tone, it may be
considered an ECHO VIOLA DA GAMBA. The stop is formed of small-scale cylindri-
cal pipes, preferably of tin or Hoyt two-ply, hard-rolled, pipe-metal. The mouths,
which are about one-fifth the circumference of the pipes, are cut low and sharp,
furnished with ears, and voiced with either some form of j rein harmonique or har-
monic-bridge, and on a wind between iX inches and 2^ inches. The scale may
vary according to the class of Organ in which the stop is placed, and its position
in the Organ ; but a suitable one, which may be accepted as normal, in the ratio of
1 : 2 -S l 9> g* ves the CC pipe a diameter of 2.51 inches; the C pipe a diameter of 1.54
iuches; and the middle c 1 pipe a diameter of 0.94 inches. Lochersays: "The
MOLINA was originally a metal stop throughout, but as the art of intonation in
modern organ-building is capable of making the transition from metal to wood
quite imperceptible, it is permissible to construct the lower notes of wood in this
and other stops." As before mentioned, Walcker, of Ludwigsburg, has used wood
basses for this stop. A certain English organ-builder has used a bass octave of
stopped wood pipes, and even gone farther by grooving the ^EOLINE to a soft
covered stop. None of these money-saving devices should be followed.
COMBINATION AND REGISTRATION. The extreme softness of the
voice of the ^OLINE renders its value in combination and registra-
tion, comparatively speaking, very limited; yet it has a delicate
timbre-creating property which is worthy of the organist's study.
For instance, it combines well with such a stop as the FLAUTO
D'AMORE, 8 FT. , creating a compound tone of peculiar charm. It also
combines in an effective manner with the MELODIA, HARMONICA,
and other flute-toned stops, provided they are not too loudly voiced
to be tonally affected. In simple registration it will be found of
value as a solo stop ; its delicate singing string-tone rendering it very
fascinating to the lover of refined organ music. The ^OLINE is
frequently drawn with the Voix CELESTE.
JiLODICON, Grk. The name originally used to designate a
keyboard, free-reed instrument and the precursor of the Harmonium.
As an organ-stop it is a variant of the free-reed ^EOLINE. An ex-
ample, under the name ALODICON, 16 FT., by Walcker, of Ludwig-
burg, occupies a place in the Second Manual Division of the Organ
in the Cathedral of Riga, where it is a soft-toned lingual stop. As '
* "Aeoline 16' und 8', ein zartes Rohrwerk von schdnwirkendem, sanftem, sauselndem Tone f
mit freischwingenden Zungen und Kurzem Schalltrichter." "Die Orgel," P. Zimmer.
36 ORGAN-STOPS
such its presence, in this age of noisy, high-pressure stops, would,
indeed, be most welcome.
^QUALPRINZIPAL, Ger. The term which has been used by
German organ-builders to designate the principal manual unison
stop. In early times of the art in Germany the simple term ^EQUAL
or AQUAL was deemed sufficient. The term, signifying unison, was
sometimes applied to other stops,
AMOROSA, Lat. The name that has been used by Steinmcyer
and other German organ-builders to designate a wood FLUTE, 8 FT.,
of small scale and soft and pleasing tone, resembling that of the
PLAUTO D'AMORE. The extended term Vox AMOROSA has occa-
sionally been used.
ANGENEHMGEDECKT, Ger. The term which has occa-
sionally been used by German organ-builders, instead of the usual
term LIEBLICHGEDECKT, to designate a small-scaled covered stop
yielding a refined tone. It is formed from the word angenehm
pleasant.
ANTHROPOGLOvSSA, Grk. The name that has been given by
old German organ-builders to the lingual stop yielding a tone some-
what resembling the human voice; now designated Vox HUMANA.
APFELREGAL, Ger. Eng., APPLE-REGAL. An obsolete lin-
gual stop which must now be classed among the curiosities of Ger-
man organ-building. The stop received its peculiar name from the
shape of its resonators; which were formed of very short cylindrical
portions surmounted by apple-shaped heads, perforated with numer-
ous small holes for the emission of wind and sound the latter nec-
essarily muffled. Seidel tells us that the cylindrical portion of the
largest resonator was only about 4 inches long. The stop was also
called KNOPFREGAL; and was made of both 8 ft. and 4 ft. tone.
See REGAL,
ASSAT. The term occasionally met with in the stop-lists of old
Organs. It is obviously a corruption of the proper term NASAT.
B
BAJONCILLO, Span. Port., BAIXONILIKX A lingual stop
found in Spanish and Portuguese Organs; the tones of which re-
semble those of the orchestral Bassoon, which, however, they do not
THEIR ARTISTIC REGISTRATION 37
imitate very closely. The stop is of 8 ft, pitch. BAJON or BAIXO
being the stop of 16 ft. pitch.
BARDUEN. The name used by Praetorius to designate a
covered stop in all essentials similar to that now known as the
BOURDON. He gives the stop as of 8 ft. pitch.
BARDONE, Ital The term that has occasionally been em-
ployed to designate the stop commonly known as the BOURDON.
There was an old bass stringed instrument called Viola di Bardone,
which may have suggested the term. It is now rendered, more in
keeping with modern nomenclature, BORDONE.
BAREM, Ger. The name given to a covered wood stop of 16 ft.
and 8 ft. pitch, voiced to yield a pure and soft tone. The term is de-
rived from the old German word baren to sing. The stop is prac-
tically identical with that commonly known as the STILLGEDECKT.
BARPFEIFE, Ger. Dtch., BAARPYP. An old lingual stop, the
resonators of which assume very peculiar forms. The most common
seems to be that of two cones joined together at their bases, and both
truncated ; the lower one soldered to the reed-block and the upper
one open at top, the opening being comparatively small so as to
subdue the tone. The illustration given by Seidel shows these
double cones surmounted by a third truncated cone, forming a bell
to the resonator. The tone of the stop is described by Wolfram* as
of a soft growling character. The stop was made of both 16 ft. and
8 ft. pitch. A BAARPYP, 8 FT., is to be found in the Echo of the
Organ in the Cathedral of St. Bavon, Haarlem, and in several Organs
in Holland and Germany. See REGAL.
BARYTON, (Grk. jtopuirovoq deep-toned). Ital., BARITONO.
Span., VARITONO. The name given by certain organ-builders to a
lingual stop of 8 ft. pitch, properly yielding a singularly rich and full
tone of medium strength when voiced on wind of moderate pressure.
Under the name BARITONE, 8 FT., a stop of this class exists in the
Solo of the Roosevelt Organ in the Cathedral of the Incarnation,
Garden City, Long Island. Another example, of 16 ft., obtains in
the Organ in the Royal Albert Hall, London, built by Willis in 1871.
FORMATION. For the production of the desirable tone, the reeds should be the
"closed " variety, and the tongues somewhat light and finely curved so as to avoid
all brassiness. The resonators to be inverted conical in form, and shaded for fine
regulation.
* ' ' Anleitung zur Kenntniss, Beurtheilung und Erhaltung der Orgeln." J. Christian Wolfram,
Gotha, 1815.
3 8 ORGAN-STOPS
COMBINATION AND REGISTRATION. A lingual stop of this class,
having a refined and sympathetic voice, should be placed in one of
the accompanimental manual divisions of every important Organ.
Its value in artistic registration could not well be overrated. Its
rich and full tone would combine with those of the flute-toned and
string-toned stops, producing numerous beautiful lights and shades
of tonal coloring, not possible with the usual and assertive lingual
stops. The day is coming, we feel sure, when the value of soft
normal or unimitative lingual stops will be realized in refined and
artistic registration.
. BASSET-HORN. FT., COR DE BASSET. Ger., BASSETHORN.
ItaL, CORNO DI BASSETTO. A lingual stop of 8 ft. pitch, voiced to
yield a tone resembling that of the orchestral instrument of the same
name. See CORNO DI BASSETTO. Locher says the German BASSET-
HORN, 8 FT., and the SERPENT, 16 FT., "measured on the same foun-
dation, are smooth-toned, free-rcecl, Pedal Organ stops, as a rule
without resonators, like the PHYSHARMONICA. They represent the
smooth reed-tone on the Zweitcs Pedal of the Ulm Cathedral Organ. "
BASS FLUTE. Ger., BASSFLOTE. An open labial stop, of
8 ft. pitch, formed commonly of wood but sometimes of metal,
voiced to yield a powerful unimitative flute-tone. This stop prop-
erly belongs to the Pedal Organ, where it is frequently derived
from the DIAPASON, 16 FT., by means of tin extra octave of pipes and
an octave coupler. For the purpose of combination and artistic
registration it is very desirable that it should be an independent stop
of distinctive strength and character of tone; and for this purpose it
should be made of metal, The German name PLOTENBASS is some-
times given to this independent stop.
BASSONELL. Described by Wolfram as a lingual stop of 8 ft.
and 4 ft. pitch, made of metal. Its tone was, in all probability, that
of a soft Bassoon character. This is seemingly the stop referred to
by Hamel under the Italian name BASSANKLLO- ' ' BASSANELLI. Ce
sont des instruments & vent du si&cle dernier [xviii* 5 ] ; ils resemblent
beaucoup au chalumeau. Dans Torgue, ils ont &t& imit^s par des
jeux d'anches particuliers de huit et de quatre pieds."
BASSOON. Ital, FAGOTTO, Pr., BASSON. Ger., FAGOTT.
Span., BAJON. A small-scaled lingual stop, voiced to imitate as
closely as practicable the tone of the orchestral instrument of the
same name. As a manual stop of 8 ft. pitch, it is sometimes, and
correctly, associated with the OBOE or HAUTBOY; being labeled
THEIR ARTISTIC REGISTRATION 39
OBOE & FAGOTTO or HAUTBOY & BASSOON. It is preferable that, in
the modern Organ, the BASSOON, 8 FT., should be introduced as a
complete and independent stop. This is very desirable, to facilitate
artistic registration. As a stop of 16 ft. pitch, the BASSOON fre-
quently appears in the Pedal Organ, and occasionally in a manual
division. When of this grave pitch it is more expressively called
DOUBLE BASSOON or CONTRAFAGOTTO. A BASSON, 16 FT., appears
in the pedal department of the Grand Organ in the Church of Saint-
Sulpice, Paris ; and a CONTRA-FAGOTTO, 16 FT., is placed in the Solo
of the Organ in St. George's Hall, Liverpool. It seems, howev&rr-ta
have been left to Walcker, of Ludwigsburg, to realize the true value
of this important stop. In his original Organ in the Cathedral of
Ulm were the following stops: A CONTRA-FAGOTT, 16 FT., and a
FAGOTT-DISCANT, 16 FT., in the Hauptwerk (First Clavier); a FA-
GOTT, 8 FT., on the Z^gites Clavier; a CONTRA-FAGOTT, 16 FT., a
SECUND-FAGOTT, 16 FT., and a FAGOTT, 8 FT., on the Yiertes Clavier;
and a FAGOTTBASS, 16 FT., on the Erstes Pedal. In all, seven im-
portant stops of the BASSOON family in an Organ of 113 speaking
stops. What an opportunity here for unique registration !
FORMATION. The BASSOON, when in its best form as a striking reed, has
resonators of the inverted conical form and small scale. The manual stop of 8 ft.
pitch may be constructed entirely of metal, or the resonators of its bass octave
may be made of some suitable hard wood, so as not to require undesirable thick-
ness. The same remark applies to the two lower octaves of the DOUBLE BASSOON,
1 6 FT. The resonators of the Pedal Organ stop may be of wood throughout. For
fine regulation the resonators should be shaded, or both shaded and slotted. Satis-
factory tonal results have been obtained from metal resonators, closed at top and
slotted or perforated in some manner near their closed ends. The wooden resona-
tors are square in transverse section, their lower ends being rounded externally
so as to fit into metal socket pieces soldered to the reed-blocks; and they must be
bored vertically with holes corresponding with the size of the reeds below. These
holes must be coif Js&ued until they open clearly into the interior of the resonators.
T n Fig. I, Plate I, is shown a CC BASSOON pipe of metal, the resonator of which is
closed and slotted. Fig. 2 shows a corresponding pipe with a resonator of wood,
shaded with an adjustable plate of thick pipe-metal. , The illustrations are drawn
to scale. The reeds are of the closed variety, and the tongues are narrow and
specially curved so as to produce the characteristic Bassoon tone.
Continental organ-builders have largely resorted to free-reeds in the formation
of their FAGOTTS, probably finding their style of striking-reed work productive of
too much clang and brassiness, altogether foreign to the tone of the orchestral
Bassoon. They have used for the free-reeds comparatively short resonators,
formed of two truncated cones, soldered together at their bases, the upper one
having a small opening for the emission of the wind and sound. Seidel, however,
gives an illustration of a BASSOON pipe having a slender cylindrical resonator, like
that tx^w sed for the CLARINET.
40 ORGAN-STOPS
COMBINATION AND REGISTRATION. The orchestral Bassoon fur-
nishes the proper bass to the Oboe. Both arc double-reed instru-
ments. The value of a good imitation of its peculiar voice in the
Organ may be realized from what Berlioz says: "The Bassoon is of
the greatest use in the orchestra on numerous occasions. Its sonor-
ousness is not very great, and its quality of tone, absolutely devoid
of brilliancy or nobleness, has a tendency toward the grotesque
which should be always kept in mind when bringing it forward into
prominence. Its low tones form excellent basses to the whole group
of wooden wind instruments. . . . The character of its high notes
is somewhat painful, suffering, even, I may say, miserable,
which may be sometimes introduced into either a slow melody, or
passages of accompaniment, with most surprising effect." Even
when voiced by an artist, the organ BASSOON, 8 FT., is only moder-
ately imitative; and on this account it may be recognized as being
more generally useful than if its voice were absolutely imitative.
An artistically made and voiced BASSOON is a stop of the greatest
value in an Organ, combining with all the medium- and softly-voiced
labial stops, producing dual or fuller tonalities of great beauty. In
combination with the OBOE or CLARINET it produces reed-tones of
singular fullness and charm. These facts go to prove that registra-
tion in which the BASSOON enters deserves the artist organist's
careful study. Unfortunately, the stop is very rarely introduced in
the ordinary Organs of to-day. In the wood-wind division or sub-
division of the Concert-room Organ, its presence, as a complete stop,
is imperative. For remarks on the DOUBLE BASSOON, see CONTRA-
FAGOTTO.
BASSPOSAUNE, Ger, Eng., BASS TROMBONE. The Pedal
Organ lingual stop of 32 ft. pitch, also termed CONTRAPOSAUNK, as
in the Organ in Christ Church, at Hirschberg. It appears in several
important German Organs, termed POSAUNE, 32 FT. ; as in the Organ
in St. Peter's Church, Berlin. It also appears, under the same name,
as a free-reed stop, in the Pedal of the Organ in the Cathedral of
Merseburg, built by Ladegast in 1855; an d a l so as a free-reed stop
in the Pedal of the Organ in the Cathedral of Gothenburg, built by
Marcussen in 1849. See CONTRAPOSAUNE*
BASS TUBA. Ger., BASSTUBA. A powerfully-voiced lingual
stop, the tones of which are supposed to represent in the Organ those
of the Bass Tuba of the modern orchestra as constituted by Wagner
and others. See CONTRA-TUBA.
PLATE I
BASSOON
7
BASSOON BELL GAMBA CLARINET COR ANGLAIS
THEIR ARTISTIC REGISTRATION 41
BASS VIOL. An appropriate name for an organ stop of 8 ft.
pitch, formed of metal and voiced to yield a string-tone less pro-
nounced than that of the ORCHESTRAL VIOLONCELLO, 8 FT. Its tone-
may be accepted as representing in the Organ that of the old Bass
Viol, the largest of the four instruments in the old English "Chest of
Viols." This stop, if voiced to yield a bright, singing string-tone,
would be invaluable in refined and artistic registration. In com-
bination with all the softer stops of pure organ-tone, flute-tone, and
reed-tone, it would impart to their different voices a delicate coloring
without destroying their individuality, adding, at the same time,
volume to their voices.
BAUERFLOTE, BAUERPFEIFE, Ger. Literally Peasant
Flute. A covered stop of small scale, commonly made of wood, and
of 8 ft., 4 ft., 2 ft., and I ft. pitch. The stops of 8 ft. and 4 ft. pitch
have clear fluty tones; while those of 2 ft. and i ft. have whistling
tones resembling those of the human mouth. An example of 8 ft.
pitch, under the name BAUERPFEIFE, appears in the Choir of the
Organ in the Church of St. Jacobi, Hamburg. The stop in its high
pitches has been commonly, and strangely we may add, introduced
in Pedal Organs; one of 2 ft. is to be found in the Bedal of the Organ
in the Church of SS. Peter and Paul, at Goerlitz, and one of I ft.
pitch in the Pedal of the Church of St. Dominico, at Prague. Wol-
fram gives the terms BAUERNFLOTE and BAUERNFLOTENBASS. The
stop has been made with perforated stoppers, and called BAUERN-
ROHRFLOTE,
BEARDED GAMBA. An open metal labial stop of medium
scale, producing an unimitative string-tone. The name is given on
account of the peculiar treatment of the mouths of its pipes. Each
mouth is provided, in addition to its projecting ears, with a project-
ing plate, soldered to the lower lip and attached to the ears. While
this beard exercises an influence on the tone, it must not be con-
founded with the harmonic-bridge or the frein harmonique. The
stop is not of much importance and has fallen into disuse.
BELL DIAPASON. FT., FLUTE A PA VILLON. This stop was
invented in France, and introduced into England at the London Ex-
hibition of 1851. For some time it was much esteemed by English
builders, but for certain reasons it is no longer made. The English
name is not appropriate, while the French name conveys the proper
classification of the stop tonally, as well as the distinguishing feature
in the formation of its pipes, technically termed the pavilion. Ton-
42 ORGAN-STOPS
ally the stop belongs to the flute-work of the Organ, and not to the
foundation- or diapason-work. For further particulars, see FLUTE A
PAVILLON.
BELL GAME A. Ger., GLOCKENGAMBA. A metal labial stop
of 8 ft. pitch ; the pipes of which have their bodies of a conical form,
and of medium scale, surmounted by a slender pavilion or bell, in
the manner shown in Fig. 3, Plate I. The name was used to distin-
guish the stop from the ordinary German GAMBA, which has pipes of
plain cylindrical form. The stop has very rarely been carried below
tenor C in its characteristic form, the bass octave either being
omitted or inserted in cylindrical pipes. As the form of the BELL
GAMBA does not allow of its being tuned at top in the ordinary
manner, its mouth is furnished with large, projecting, flexible ears,
as shown in Fig. 3, which flatten or sharpen the tone as they are bent
toward or away from the mouth. This stop was usually made of tin
or high-class alloy; and when artistically voiced yielded a string-tone
of great delicacy and charm, strongly resembling that of the old
orchestral Viola da Gamba. The trouble attending its construction,
voicing, and perfect regulation has led to its disappearance from the
Organs of to-day. See VIOLA DA GAMBA.
BIFARA, Ger. Lat, TIBIA BIFARIS or BIFARIUS. This term
has been employed by certain old German organ-builders to desig-
nate two different labial stops, both of which were double-toned, and
one positively dual in formation. Seidel, who one must recognise
as an authority on old German organ-building, says: the BIFARA "is
a fine but scarce labial stop in the manual, eight or four feet, open, of
PRINCIPAL scale. Every pipe has, like those of the DOPPLFLOTE,
two mouths, one of which stands a little higher than the other,
whereby a pleasant sort of vibration is caused, similar to that of the
UNDA MARIS. To obtain from this stop a very soft and agreeable
tone, the feet of the pipes are but partly open, so as to admit only
a small amount of wind. The BIFARA is also produced in another
way; namely, by giving one tone to two pipes, one of which is tuned
a little higher than the other. A BIFARA of tin, of two ranks, first
rank a GEDECKT, 8 FT., second rank open pipes of 4 ft., of a soft
string-tone, stands on the Third Manual of the colossal Organ built
by Walcker, of Ludwigsburg, in St. Peter's Church, St. Petersburg/ 1
The BIFARA is extremely interesting as the first step toward the
artistic formation of dual stops, for the production of compound
tones, impossible to be produced from single stops, or even two stops
THEIR ARTISTIC REGISTRATION 43
not specially combined.* For many years we have pleaded for
attention to be given to tone-coloring in this direction, but in this,
as in other matters, our appeal has fallen on deaf ears. Dollars and
don't-careism have blocked the way
BLOCKFLOTE, Ger. Lat, TIBIA VULGARIS. Eng., BLOCK-
FLUTE. The name given to an open metal labial stop of the ordinary
cylindrical form, of very large scale, and usually of 4 ft. pitch. The
tone of this stop was of a normal flute character, which varied in
power in different examples. Respecting the original German ver-
sions of the stop, Seidel remarks: "It is of Flute-work, sometimes
open, sometimes stopped, and now and then made of conical pipes.
It is made of tin or pipe-metal, of 16 ft., 8 ft., 4 ft., and 2 ft. tone.
It imitates the tone of the Flute into which the wind is blown at the
end." In the concluding remark, allusion is evidently made to the
old FlUte & Bee, or Direct Flute. "Father Smith" introduced the
BLOCKFLOTE in his Durham Cathedral Organ (1683), and in that he
built for St. Paul's Cathedral, London (1697).
BOMBARDE, Fr. A lingual stop of 16 ft. pitch and powerful
intonation. The BOMBARDE is commonly found in important French
Organs; for instance in the Organ in Saint-Sulpice, Paris, built by
Cavaill6-Coll, there are four BOMBARDES; and in his Organ in the
Cathedral of Notre-Dame there are also four a CONTRE-BOMBARDE,
32 FT., and three BOMBARDES, 16 FT. In his notable scheme for the
"Orgue Monumental" to be erected in St. Peter's, at Rome, we
find in the Pedal Organ a CONTRE-BOMBARDE, 32 FT.; a BOMBARDE,
16 FT. ; and a QUINTE BOMBARDE, 10% FT. The tonal effect of these
three commanding voices in combination, heard in such a building
as St. Peter's, is beyond one's power to imagine. This important
stop gives a name to a manual division in large French Organs;
namely, "Clavier des Boxnbardes." The resonators of the BOM-
BARDE are of the inverted conical form, constructed of stout metal
and to a large scale.
There was an old lingual stop sometimes called BOMBARD, but
more commonly BOMMER or BARBOMMER, evidently deriving its
name from Bombardo a mediaeval reed instrument of large size and
* " La BIFARA (de bifaris) est done une autre espece de flute a deux tuyaux ouverts, de moy-
enne taille, de huit-pieds, et en etain. Ses deux benches, etroitement pincees, mais dorit 1'une
est plus haute que 1'autre, n'aspirent que peu de vent par un seul pied, et rendent une harmonie
plus ou moins agreable. On la fait encore de deux tuyaux separes, ayant chacun leur pied; 1'un
des deux tuyaux ayant un peu plus hauteur nue 1'autre, et par consequent de rondeur de son."
Regnier.
44 ORGAN-STOPS
coarse intonation, and probably the precursor of the Fagotto. The
old instrument, or some modification of it, was called Pommer, and
led to the formation of a family of six instruments under the name,*
According to Wolfram (1815), the old BOMMER, BOMBARD, or BAR-
BOMMER, was a Pedal Organ stop of 16 ft. and 8 ft. pitch, having
resonators of wood, which were sometimes partly covered to soften
the tone.
BOMBARDON, Pr. Ital., BOMBARDONE. A Pedal Organ
lingual stop, full-toned, and of 32 ft. and 16 ft. pitch. It derives its
appellation from the large brass instrument of the same name, the
powerful and grave tones of which it is designed to imitate so far as
its compass extends. The accepted compass of the Bombardon is
from FPF to d 1 , but five notes lowerto CCC can, with difficulty,
be produced by a skillful performer. The BOMBARDON has been
confounded with the BOMBARDE, but they are properly two distinct
stops tonally. The voice of the BOMBARDON should be between the
voices of the BOMBARDE and the BASSOON, while partaking of the
character of both, hence the value of such a stop in Pedal Organ
registration. A BOMBARDON, 32 FT., exists in the Pedal of the Organ
in the Cathedral of Ulm.
BORDUNALFLOTE, Ger. The name given to an open wood
stop of 8 ft. and 4 ft. pitch, the pipes of which are properly of inverted
conical form, and slightly oblong in transverse section. The pipes
are voiced to yield a smooth fluty-tone, having, owing to the creation
of certain upper partials, a trace of string quality. An example, of
8 ft. pitch, made of pine and pear-tree, by the celebrated organ-
builder, Ladegast, exists on the Second Manual of the Organ in the
Cathedral of Schwerin. Another, by the same builder, exists on the
Second Manual of the Organ in the Nicolaikirche, Leipzig, The
stop deserves consideration for introduction in important Organs;
but, owing to more than ordinary labor being required in its forma-
tion, it is not likely to be favored by organ-builders to-day, The
stop has been frequently, but less correctly labeled PORTUNALFLOTE,
BOURDON, Pr. Ital,, BORDONE. Ger., BORDUN, BRUMMBASS,
BASSBRUMMER. In English and American Organs the BOURDON
may be said to invariably appear in the form of a covered, labial wood
stop, of large scale and 16 ft, pitch. Its characteristic tone has a
somewhat dull droning quality, which is well expressed by the old
* The entire family of Pommers is shown in the Crosby Brown Collection of Musical Instru-
ments in The Metropolitan Museum of Art, New York City.
THEIR ARTISTIC REGISTRATION 45
German synonyms BRUMMBASS and BASSBRUMMER(from brummen
to hum or drone).
The BOURDON, 16 FT., most frequently appears, in English and
American instruments of moderate dimensions, in the Pedal Organ;
while in small Organs it is too frequently the only pedal stop. In
large instruments it is introduced in one or more of the manual divi-
sions, and commonly in the Swell Organ. In the Concert-room
Organ in the Town Hall of Leeds, England, the BOURDON, 16 FT,, is
introduced in the Great, Swell, and Echo Organs. As might be
expected, the manual stops are made of a smaller scale than those
introduced in the Pedal Organ. In German instruments the BOUR-
DON, 1 6 FT., may be said to be a manual stop, although it sometimes
appears in the Pedal Organ. In important instruments it is found
on the manuals of 32 ft. and 16 ft. pitch. In the Organ in the Cathe-
dral of Bremen, built by Schulze in 1850, we find in the Hauptwerk
BORDUNS of 32 ft. and 16 ft., and in the Unterwerk BORDUNS of
32 ft. and 1 6 ft. ; but no stop under the name in the Pedal. It occa-
sionally appears of 8 ft. pitch, as on the First Clavier of the Organ in
St. Peter's Church, Hamburg. In French instruments the BOURDON
is introduced in the manual divisions of 16 ft. and 8 ft. pitch. In the
Organ in the Cathedral of Notre-Dame, Paris, there are no fewer
than five manual BOURDONS two of 16 ft. and three of 8 ft. pitch.
In some rare instances the French organ-builders have applied the
name to a covered metal stop of large scale and of 4 ft. pitch.
FORMATION. In modern English and American Organs the true BOURDON,
1 6 FT., is invariably and properly made of wood; but we are told in Dr. Burney's
"History of Music" that the old English builders sometimes constructed it of
metal. Speaking of Snetzler's Organ in St. Margaret's Church, Lynn Regis, he
says : " One of the metal stops of this instrument, called the Bourdon, is an octave
below the Open Diapason, and has the effect of a double bass in the chorus." In
all probability the quint was somewhat prominent in the tone of this covered metal
stop, imparting to it a similarity to that of the orchestral Double Bass. Both the
French and German organ-builders use metal pipes in the higher octaves of their
manual BOURDONS, especially in those of 8 ft. pitch. In the Great of the celebrated
Haarlem Organ, the BOURDON, 16 FT., is of metal throughout the compass of the
manual.
The pipes of the BOURDON, like those of all quadrangular wood stops, are
formed of four boards, glued together at their edges and at one end to a block, or,
in large pipes, to two cross-pieces of wood, which assume the position and function
of the block. The large scale of the pipes renders the use of solid blocks undesir-
able save in the upper octaves. The block or upper cross-piece should be faced
with hard wood, as its upper edge forms the lower lip of the mouth, which is
cut in the front board of the pipe. In the accompanying illustration, Fig. 4, is
shown the construction of the lower part of a CCC pipe, and the formation of the
mouth according to the German and English methods. The former is shown in
4 6
ORGAN-STOPS
Diagram A, in which will be observed the depression of the upper cross-piece,
and its front edge beveled and cut so as to form the wind- way of the mouth, and
placed level with the upper edge of the cap. In Diagram B is given the correspond-
ing Section of the English mouth, showing the horizontal upper cross-piece, and
FIG. 4
the adjustment of the hollowed cap,- in which is formed the wind- way, slightly
below the lower lip of the mouth. Both these treatments have an influence on the
tonality of the pipes. In Diagram C is given a Front View of the lower portion of
the pipe showing the German mouth.
SCALE. The scale of the BOURDON pedal or manual- is a
matter of considerable importance, as it afTcctvS both the power and
tonality of the stop, and, accordingly, its value in combination and
registration. It should be dictated by the general character of the
Organ, and specially with regard to the tonality of the other stops
of the division in which it is placed. It would seem, from the many
examples we have examined, that this matter has not been fully
considered by the builders of many Organs. Inordinate scales are
not uncommon. The BOURDON, in its normal form, varies greatly
in scale, chiefly because it is inserted in both the pedal and manual
departments of the Organ, and in its different manual divisions,
The most suitable scales for the Pedal Organ stops range from a
width of 5% inches by a depth of 8 inches to a width of 8J^ inches
by a depth of 10 inches internal measurement for the CCC pipe;
all scales having the ratio I : V^, halving on the seventeenth pipe.
The former scale has been adopted by Schutec for the stop in the
Pedal Organ of his fine instrument in St. Peter's Church, Hindley,
England ; the CCC pipe of which has a mouth 4 inches high. The
THEIR ARTISTIC REGISTRATION 47
scale of 6.60 inches by 8.56 inches is ample for all ordinary Church
Organs which have properly apportioned pedal departments : while
the scale of 4.10 inches by 5.56 inches is suitable for a small Church
or a Chamber Organ. Larger scales than the maximum given above
have been used by English builders, but with no advantage in any
direction.
The scales of the manual BOURDON, 16 FT., may range from a
width of 3.76 inches by a depth of 5.10 inches, to a width of 5.10
inches by a depth of 7.20 inches. Schulze's effective BOURDON in
the Great of the Hindley Organ has its largest pipe measuring 37/3
inches in width by 6% inches in depth. This scale is noteworthy on
account of its proportionate depth, which is unusual. This pipe has
its mouth 3 J/g inches in height.
TONALITY. The tones of the BOURDON vary almost as greatly
as its scales. Those which have characterized so many of the stops
made and voiced by the older English builders are the least satis-
factory; due to largeness of scale, lowness of mouth, and undue
thinness of the upper lip. The tones of such stops being tubby and
often unmusical in the extreme. The BOURDONS usually made by
English organ-builders have mouths seldom over half their width in
height, and often less than one-third. Blown by wind of moderate
pressure, these stops yield tones in which the twelfth, or first har-
monic of a stopped pipe, is more or less prominent in combination
with a somewhat dull prime tone, sometimes approaching the tone
of the QUINTATEN. In certain tonal apportionments it may be desir-
able to introduce a BOURDON with, this compound voice. The
BOURDONS of the German organ-builders, usually of moderate scale,
and copiously blown by wind of moderate pressure, have their
mouths cut up very high ; rising from about two-thirds their width
to a height exceeding their width, as exemplified by the stop in the
Swell of the Hindley Organ, labeled LIEBLICHBOURDON, 16 FT.,
the largest pipe of which measures 3^ inches in width by 5 inches in
depth, and has a mouth 3^ inches in height. This beautiful stop
yields almost a pure organ-tone of singular fullness and sweetness
a perfect stop for artistic registration.
The tones yielded by small-scaled and high-mouthed BOURDONS
are fuller and purer than those produced by the large-scaled and low-
mouthed English stops; and on this account, combined with the
comparatively small standing room they require on the wind-chest,
they are most suitable for insertion in the manual divisions of the
Organ, and especially those inclosed in swell-boxes. The tone of the
BOURDON pipe is affected to a considerable extent by the thickness
48 ORGAN-STOPS
and finish of its upper lip. These facts were realized by the late T. C.
Lewis, one of England's most artistic and accomplished organ-
builders. The largest pipe of the fine BOURDON, 16 FT. , in the Great
of his Organ in the Public Halls, Glasgow, measures 5^-3 inches in
width by 6 ^ inches in depth, having a mouth 3 7/8 inches in height,
with a square-cut upper lip -J-J inch in thickness. The tone of the
BOURDON pipe is also affected by the shape given to the upper lip of
the mouth, which may vary from the straight line, as shown in Fig.
4, to an arch approaching a semicircle. To produce the desirable
ground-tone from pipes so voiced, a copious flow of wind, at a moder-
ate pressure, is absolutely necessary.
While the tones yielded by the different types of BOURDON are
somewhat indeterminate and difficult to classify, it is certain the
stop belongs to the flute-work of the Organ. It has been considered
the bass to the so-called STOPPED DIAPASON, 8 FT. ; but it is just as
little allied to the DIAPASON proper as the unison stop, which also
belongs to the unimitative flute-work.
COMBINATION AND REGISTRATION. The BOURDON, when prop-
erly scaled and voiced, is a valuable stop in artistic registration;
combining in a highly satisfactory manner with both labial and lin-
gual stops of almost all pitches. Combined with stops of unison
(8 feet) pitch, it adds gravity without prominently affecting their
characteristic timbres; this is due to the indeterminate or neutral
nature of its tonality, which is, perhaps, its greatest virtue. The
true tone of the BOURDON gives firmness and body to the softer
lingual stops, and a certain fullness and orchestral richness to such
lingual stops as the DOUBLE TRUMPET, CONTRAFAGOTTO, and TROM-
BONE, 1 6 FT.; losing itself in their assertive voices so far as their
individuality is concerned.
The indeterminate character of the BOURDON tones renders it
impracticable to form such a family as one finds in the LIEBLICH-
GEDECKTS, but that is not a matter for regret, for little is gained by
massing neutral tones.
We strongly recommend the organist who has the command of an
Organ which contains a good BOURDON in any of its manual divisions
to carefully study what may be called the Registration of the BOUR-
DON. He will be repaid if only education of the ear is the result,
rendering it sensitive to the perception of compound tones on a 16 ft,
foundation.
BOURDON DOUX, Pr. The name given by French organ-
huilders to a soft-toned covered stop of wood and metal, somewhat
THEIR ARTISTIC REGISTRATION 49
resembling in formation and tone the German LIEBLICHGEDECKT.
This is a very desirable stop in registration, with the more refined
labial and lingual stops, on the 16 ft. base.
BOURDONECHO Ger., BORDONECHO The term which has
been employed to designate a BOURDON of small scale and of a soft
humming tone, differing widely from the covered tone of the LIEB-
LICHGEDECKT. The pipes, made on the BOURDON model, should be
of wood in the three lower octaves. This stop should be voiced on a
wind-pressure not exceeding 3 inches, but preferably lower.
The value of a 16 ft. stop of this class, which would bear about
the same relationship to the BOURDON as the DULCIANA does to the
DIAPASON, would unquestionably be great in a soft-toned manual
division in which an unassertive double voice would prove highly
effective in a wide range of artistic registration, in which an assert-
ive double voice would be inadmissible.
BUCCINA, Lat. The name originally given to an instrument
formed from the horn of an ox, blown by a shepherd to gather his
flock. It is understood to be derived from bucca the cheek puffed
out. Subsequently used by the old organ-builders to designate a
POSAUNE. The term is now obsolete in organ nomenclature.
BUZAIN, Dtch. The name commonly used by the Dutch
organ-builders to designate a stop generally known as the POSAUNE.
It is applied without any qualification to the stop in its different
pitches. In the Pedal Organs of the large instruments in the Cathe-
drals of Haarlem and Rotterdam it appears of both 32 ft. and 16 ft.
pitch.
CAMPANA, Ital Ger., GLOCKLEIN. Lat., TONUS FABRL
The exact nature of the stop to which this name is strictly applicable
has not been decided. That its voice shall resemble the sound of
bells, or the clang of metal against metal, is evidently implied.
Seidel, under the term GLOCKLEIN-TON (Tonus fabri), describes it
as a stop of a large scale and high pitch, the tone of which resembles
the clang of hammers beating on a sonorous anvil; an example of
which is to be found, of 2 ft. pitch, in the Oberwerk of the Organ in
St. Peter's, at Goerlitz. The term has also been applied to a stop
of I ft. or 6 in. pitch, which, on account of the smallness of its pipes
and tonal reasons, consists of similar octaves, repeating at every
50 ORGAN-STOPS
octave throughout the compass of the manual clavier. It was in-
troduced, of I ft. pitch, in the Organ in the Church of St. Paul, at
Rusthale, Kent, built by Bryceson Brothers, in 1876. It was in-
serted in the Swell Organ, in which, in certain registrations, it pro-
duced bell-like tones. There is an idea suggested by the tonal effects
of this probably unique stop well worth consideration, We are of
opinion, after some study of the subject from a scientific point of
view, that some remarkable tonal effects could be produced in regis-
tration, in which a small-scaled and softly-voiced octave-repeat-
ing stop, of i ft. or 6 in. pitch throughout, would be introduced along
with both labial and lingual voices. The capabilities of organ-stops
of a refined nature are by no means exhausted.
CARILLON, Fr. Ger., GLOCKENSPEIL. The term used to
designate a timbre-creating compound stop, formed of two, three, or
four ranks of medium-scaled open metal pipes, the clang-like tones
of which somewhat resemble the sounds of bells. The characteristic
features of the labial CARILLON lie in the scales and voicing of its
pipes, and in its composition presenting octave-, third-, and fifth-
sounding ranks. The following is an example of a three-rank stop,
in which a third-sounding rank appears throughout its compass.
CARILLON III. BANKS.
CC to E 17- 19 22.
F to e x ...,.., 15 17 19.
f 1 to e 3 12 -15- 17.
f a tO C 10- 12- 15.
In the regulation of the stop, the third-sounding rank has to be
kept bright and slightly prominent, as on it largely depends the
clang and bell-like effect. Good examples are to be found in some of
the important Organs constructed by Cavailld-Coll. The CARILLON
in his Organ in the Town Hall of Manchester has a fifth-sounding
rank only from CC to F#, and a fifth-, a third-, and an octave-sound-
ing rank from G to c 4 . On examining this effective stop, we found
the pipes of large scale, having wide and low mouths, and languids
finely and closely nicked. It would appear that Cavaill6~Coll con-
sidered it unnecessary to carry the bell effect below G- There is a
four-rank stop of this class, under the name GLOCKENSPIEL, in the
Echo of the Organ in the Centennial Hall, Sydney, N. S. W. The
CARILLON is to be found in certain old Dutch Organs, as in that in
the Church of St. John the Baptist, at Gouda, built in 1736. The
stop is in the Choir, and is divided and labeled Discant III. ranks
and Bass II. ranks; the third-sounding rank being confined to the
THEIR ARTISTIC REGISTRATION 51
Discant. The stop has been introduced in some German Organs,
but we have not been able to find it in the stop-lists of any impor-
tant modern instruments made by German organ-builders. For
particulars respecting the percussion stops, see GLOCKENSPIEL.
COMBINATION AND REGISTRATION. As the compound labial
CARILLON is both harmonic-corroborating and timbre-creating, it
can readily be understood that it would enter into numerous tonal
combinations with both labial and lingual stops, and produce, in
artistic registration, many very effective tonalities. The effects
produced in registration in which labial stops are combined with a
bell-toned percussion stop are now fairly well known; but the more
legitimate organ effects, produced by the employment of the labial
CARILLON in registration is almost unknown in this country en-
tirely so we think. We have heard them and can speak very strongly
in their favor. In our study of registration for the production of
uncommon tonal effects we struck one, in the Cavaille-Coll Organ
in the Albert Hall, Sheffield, worthy of note. When the combination,
in the Solo Organ, of the FLUTE HARMONIQUE, 8 FT., the DOUBLETTE
2 FT., and the TIERCE, i% FT., was drawn and played staccato, the
most remarkable bell-like effect was produced. This would seem to
point to a type of registration not commonly adopted or understood :
but it might be studied with advantage.
CELESTA. A percussion stop of recent introduction in the
Organ. In its best form it consists of a series of metal plates, of
graduated dimensions, adjusted over resonators; all of which being
accurately tuned produce pure and agreeable tones which combine
well, when in strict accord, with the voices of the softer labial stops.
CfiLESTE, Fr. A stop formed of small-scaled open labial pipes,
preferably of tin, voiced to yield a singing tone, inclining in good
examples to a string quality. It is tuned slightly sharp, so as to
produce when drawn with another stop of similar tonality and of
standard pitch a delicate undulating effect, supposed to imitate a
celestial voice. See Voix CELESTE.
CELESTINA. The name given by English organ-builders to
stops of different forms and tonalities. The name appears to have
been introduced by William Hill, of London, and given to a small-
scaled and softly-toned wood FLUTE of 4 ft. pitch, invented by him.
A stop of this name and pitch, and, doubtless, of the same character,
is introduced in the Choir of Hill's great Organ in the Centennial
Hall, Sydney, N. S. W. A CELESTINA formed of metal pipes, yield-
52 ORGAN-STOPS
ing a louder tone differing from that of the original stop, is to be
found in the Willis Organ in the Royal Albert Hall, London.
COMBINATION AND REGISTRATION. In the present school of
organ-building in this country, with its craze for high wind-pres-
sures, too little attention is paid to stops of the CELESTINA class and
octave pitch : yet anyone who has studied registration must realize
the great value of their voices in giving a certain brightness and life
to unison combinations, without in any way disturbing their pitch.
From a scientific point of view, one must look upon the introduction
of such an octave stop as merely corroborating the first harmonic of
the open unison stop or stops, while it may, at the same time, in-
troduce a new element into the timbre of the tonal combination.
Hence the value of the CELESTINA, 4 FT., in artistic and scientific
registration. See QELESTINA.
CHALUMEAU, Pr. Ger., SCHALMEI. Ital., SCIALUMO. Eng.,
SHAWM. The name given to a soft-toned lingual stop, commonly
of 8 ft. pitch, the voice of which is supposed to imitate that of the
obsolete instrument called the Schalmei or Shawm, the precursor of
the Clarinet. So far as reed instruments are concerned, the term
Chalumeau is applied now only to the low register of the Clarinet.
Describing the old SCHALMEI, Seidel says: "It is a soft, agree-
able reed stop, in imitation of the instrument of the same name used
by the shepherds of Southern Europe. It is of 8 ft. and 4 ft. on the
manuals, and 16 ft. in the Pedal Organ, when it is termed SCHAL-
MEIBASS. The tubes are generally funnel-shaped, and of a larger
scale than those of the TRUMPET. Sometimes they were closed with
the exception of a few sound-holes. The tone of the stop varies as
greatly as the form of its pipes." Examples of the stop are to be
found in several German Organs, but it is no longer in favor. It
appears, of 8 ft. pitch, in the Oberwerk of Ladegast's Organ in the
Cathedral of Merseburg, built in 1855. Stops labeled CHALMEAUX,
of 8 ft. tone, were inserted by Silbermann in the Choir^of his fine
Organs in the Catholic Church and the MarienkircsSeTat Berlin. In
Schulze & Son's notable Organ in the Marienkirche, at Lubeck,
there is a SCHALMEI, 4 FT, , in the Erste Pedal, thus described ; " Von
engerer Mensur und schonem singenden Ton, vortrefflich zur Fuhr-
ung des Cantus firmus geeignet ' ' In the Great of the Organ built by
Elliot & Hill for York Minster there was isTSHAWM, 8 FT, There is, a
SCHALMEI, 16 FT., the resonators of which are cylindrical terminating
in a partly covered bell, in the Organ in the Colston Hall, Bristol,
built by Norman & Beard, The tone of this stop is rich ; partaking,
THEIR ARTISTIC REGISTRATION 53
as might be expected, of the tones of the CLARINET and the COR
ANGLAIS in effective combination. A Pedal Organ stop of this de-
scription would be a valuable addition to a large Organ. Softly-
voiced lingual stops are much called for in Pedal Organs to-day.
They are conspicuous by their absence.
CHAMADE, Fr. The term affixed by French organ-builders
to the name of a lingual stop, to indicate that its resonators are dis-
posed in, or near to, a horizontal position, and so displayed as to
send their sounds perfectly unobstructed to the ear. In the descrip-
tion of the Organ erected by Cavaille-Coll in the grand Church of
Saint-Ouen, at Ruen, we find in the Grand-Orgue, TROMPETTE EN
CHAMADE and CLAIRON EN CHAMADE. These are projected, in fan-
form, from the lateral divisions of the case, and free of the advanced
Positif . See Frontispiece. This treatment is so common in Spain
that it may be considered a characteristic of the important Organs
of that country. It is to be condemned on all musical grounds ; for
the powerful lingual stops so disposed should be inclosed in swell-
boxes, and so endowed with powers of flexibility and expression.
Uncontrollable lingual stops are an abomination from every musical
point of view.
CHIMNEY FLUTE. Fr., FLUTE A CHEMINEE. The names
given by English and French organ-builders to the half-covered
labial stop more appropriately called in German nomenclature
ROHRFLOTE (REED-FLUTE). See ROHRFLOTE.
CHORALBASSET, Ger. Described by Wolfram (1815) as
an open labial stop of 2 ft. pitch introduced in the Pedal Organ.
The stop is also called CHORALBASS. The term CHORAL has been
prefixed to several loud-toned stops to indicate their special suit-
ability for giving out the melody of a Choral. With this meaning,
we meet with such stop-names as CHORALPRINZIPAL, CHORALPRAS-
TANT, and CHORALFLOTE. According to Seidel, the Pedal Organ
CHORALBASSET has been made of large-scaled pipes of 4 ft., 2 ft.,
and i ft. pitch : the manual stops, CHORALPRINZIPAL and CHORAL-
PRASTANT, being of 8 ft. and 4 ft. pitch.
CINQ, Dtch. The name given to a lingual stop, of 2 ft. pitch,
inserted in the Pedals of the Organs in the Cathedrals of St. Bavon,
Haarlem, and St. Lawrence, Rotterdam; and churches in Delft and
Utrecht. The introduction of high-pitched stops is a characteristic
of Dutch Pedal Organs; desirable on account of their brightening
effects, and much to be desired in the Pedal Organs of to-day.
54
ORGAN-STOPS
CLAIRON, Fr. A lingual stop of large scale and 4 ft. pitch,
commonly attending the TROMPETTE wherever it is introduced, as
in the pedal and four manual divisions of the Grand Organ in Notre-
Dame, Paris. See CLARION.
CLAIRON-DOUBLETTE, Fr. Eng., OCTAVE CLARION. A
very uncommon lingual stop, of 2 ft. pitch; an example of which
exists in the Grand-Chceur of the Organ in the Church of Saint-Sul-
pice, Paris. As "this-Jaigh-pitched stop cannot well be carried in
lingual pipes of its true pitch above treble c 2 , the two higher octaves
may either be duplications of the treble octave, or be formed of
large-scaled and loudly-voiced labial pipes. The value of such a
stopis very questionable; so much so that the example alluded to is
the only one we have found in an Organ. Regnier docs not mention
the stop.
CLARABELLA. Ger., OFFJENFLQTE. The name given to an
open wood stop of 8 ft. pitch, which, in its original form, was
invented by J. C. Bishop, of London. It became a great favorite of
B
FIG. 5
the English organ-builders; and during the latter half of the last
century was inserted in the majority of the organs then constructed.
A fine example of the stop exists in the Great of the Bishop Organ
in the Church of St. Mary, Nottingham. The CLARABELLA, in its
THEIR ARTISTIC REGISTRATION 55
original and most desirable form, has of late years been compara-
tively seldom made, a fact to be regretted.
FORMATION. The CLARABELLA was devised by J. C. Bishop to take the place
of the higher octaves of the STOPPED DIAPASON, 8 FT., which he found insufficient
in volume of tone to cope with the increase of power being given to the OPEN DIA-
PASON and other stops of the Great Organ. At first it occupied the treble octaves
from middle c 1 only, but later it was carried down to tenor C, as in the Great of
the Organ in St. Mary's, Nottingham ; the bass being in covered pipes. The stop
has, however, been made with its open pipes throughout, as in the Organ built by
Booth, of Wakefield, for the Brunswick Chapel, Leeds. When properly made and
voiced the CLARABELLA is an extremely valuable stop, and in an important Organ
it should be carried throughout the manual compass in its open pipes.
The CLARABELLA pipe is of the usual quadrangular form and construction, as
shown in the accompanying illustration, Fig. 5. It has the direct, English mouth,
usually about one-quarter its width in height, having the upper lip somewhat thin
and carefully rounded. It requires no ears; and its cap is hollowed for the wind-
way, level on top, and without a beard, as indicated in Section A.. The propor-
tions of the mouth are shown in Front View B. The scale of the stop should vary
according to the volume and the strength of the tone required; but as a rule it has
been made to what may be considered a large scale for an 8 ft. open wood stop.
The following scales, to different ratios, may be accepted as normal:
CLARABELLA SCALE IN INCHES RATIO 1:2.66
PIPES CC C c 1 c 2 c3 C 4
WIDTH 4.26 2.62 1. 60 0.98 0.60 0.37
DEPTH 5.24 3.21 1.97 1.20 0-74 0.45
CLARABELLA SCALE IN INCHES RATIO 1:2.519
PIPES CC C c 1 c 2 cs c<
WIDTH 3.79 2.39 1.50 0.95 0.60 0.38
DEPTH 4.96 3.13 1.97 1.24 0.78 0.49
TONALITY. The tone of the stop as voiced by its inventor was
of a full, round fluty quality, naturally brighter and more penetrat-
ing than that of the STOPPED DIAPASON pipes it displaced; and,
accordingly, it did not join satisfactorily with the tones of the
covered tenor and bass octaves, whose harmonics were different.
That fact is a strong reason for the formation of the stop with open
pipes throughout. The tone of the CLARABELLA, in its best form
and when voiced by a master, is beautiful and invaluable in any
property apportioned Great Organ, and whether the stop is inclosed
or free. Its tone should be midway between that of the DIAPASON
and the full-toned German HOHLFLOTE, differing from the tone of
the wood DIAPASON as made by Schulze. The true tone of the
CLARABELLA is clear, singing, and reposeful; making it valuable in
every direction in which it may be employed.
56 ORGAN-STOPS
COMBINATION AND REGISTRATION, The CLARABELLA can be
used as a solo stop with good effect, especially when strains of a
placid nature are desired; when its liquid tones afford an agreeable
relief to those of a more assertive and cutting character so commonly
resorted to in solo passages. But it is in combination and artistic
registration that the chief value of the stop is found. There are
very few wood stops which enter so satisfactorily into combination
with lingual stops of all classes. When under control, as it should
be in a properly appointed Organ, it is a desirable addition to the
Vox HUMANA and CLARINET. The true CLARABELLA has the de-
sirable property of building up the tone of almost every unison stop
to which it is added without prominently affecting its character. In
registration it may be considered an all-round helper. It may, in
another important direction, be used as a unison base, on which to
build up many beautiful compound tones. For this purpose it may
be combined with the softer octave, mutation, and harmonic-corrob-
orating stops. With such a refined stop as the DULCIANA CORNET,
V. RANKS (q. z;.), it produces a charming compound tone,* practi-
cally unknown in the Organs of to-day, with their high wind-pressures
and MIXTURES of unregulated and screaming voices.
CLARIBEL FLUTE, An open wood stop of 4 ft, pitch, which
in its original form is practically an OCTAVE CLARABELLA, but
slightly more fluty in tone. Willis, in his dislike of wood stops,!
made his CLARIBEL FLUTES, 8 ft., of metal from middle c r , and har-
monic from g 1 . The bass and tenor octaves being of wood pipes.
COMBINATION AND REGISTRATION. The value of octave stops of
varied tonalities has not been fully realized, judging from their
sparing introduction even in important Organs. Too much depend-
ence being placed on octave coupling, always to be avoided when
possible. The CLARIBEL FLUTE, on account of the good mixing and
not too assertive qualities of its tone (when properly voiced on wind
of moderate pressure), is a most serviceable stop in registration in
both its offices of harmonic-corroborator and timbre-creator. Its
best position in the Organ is in the division in which the CLARABELLA
is placed ; for in combination with that stop it produces a beautiful
unimitative tone, upon which string- or reed- tone can be thrown with
fine effect,
* This tone is well known to us by long experience, for, under the names PLAUTO TBDBSCA,
8 FT., and RIPIENO DI CINQUE, both stops were in our Chamber Organ, placed in different swell-
boxes, and, accordingly, subject to compound flexibility and expression,
f So great was his dislike that in all the manual divisions of his large Organ in the Royal
Albert Hall, London, there are only two wood stops, Ho assured us, on asking his reason, that
he obtained better wood-tones from metal pipes, We questioned it.
THEIR ARTISTIC REGISTRATION 57
CLARIN, Span. A lingual stop of the TRUMPET class and usu-
ally of octave pitch. It is labeled in Spanish Organs with several
qualifications as follows : CLARIN REAL, CLARIN DE B ATALLA, CLARIN
CLARO, CLARIN SORDINO, CLARIN PARDO, CLARIN DE BAJOS, and
CLARIN DE Ecos.
CLARINA, Ital. An open metal labial stop of large scale and
2 ft. pitch, voiced to yield a powerful tone. It is introduced in the
Pedal Organ to impart clearness and distinctness to heavy pedal
combinations, and especially when they are used in solo passages
and fugues. A stop of so high a pitch is very rare in a Pedal Organ.
It is desirable in combination. See COMPENSATIONSMIXTUR.
CLARINET. Pr., CLARINETTE. Ger., CLARINETTE. Ital.,
CLARINETTO. The lingual stop formed and voiced to imitate as
closely as possible the tones of the orchestral Clarinets. When made
by master hands, it is one of the most successful imitative stops in
the Organ. It is invariably and correctly of 8 ft. pitch, and has very
frequently in English Organs been commenced at tenor C, probably
because the orchestral Clarinet in A does not go lower than C# ; but
as the Bass Clarinet descends about an octave below this note, the
CLARINET of the Organ should be carried, in all cases, throughout
the manual compass. The term CLARIONET has been commonly
used by English organ-builders, but should never be employed. It
has no relation to the English word "clarion" any more than the
CLARINET has to the CLARION of the Organ.
FORMATION.- The resonator of the CLARINET pipe is cylindrical throughout
save where it is attached, by a short conical piece, to the reed-block : it is entirely
open at top, where it should be furnished with an adjustable slide for fine regula-
tion. The form of the complete pipe is shown in Fig. 6, Plate i . It is to this form
of resonator that is mainly due the imitative voice of the stop, resembling as it
does the cylindrical tube of the orchestral Clarinet. The resonator of the CC pipe,
yielding a note of 8 ft. pitch, instead of being, approximately, the true speaking
length of 8 ft., is only a little more than 4 ft. in length. Both in length and scale
of its resonators the CLARINET differs from all the lingual stops belonging to other
families. The reeds of the stop are of medium scale and of sufficient length to
receive tongues to speak the unison tone; they are properly of the closed variety,
their triangular openings extending directly from the thick discs which cover their
lower ends. The tongues are firm and finely curved so as to get rid of brassiness.
In German and Swiss Organs the CLARINET is almost invariably a free-reed stop,
of a somewhat large scale and generally feeble in tone and far from imitative.
Some examples have inverted conical resonators which in no way favor the pro-
duction of a characteristic Clarinet tone.
TONALITY. The tones of an artistically voiced CLARINET are
extremely pleasing and full of character, and surpass in imitativeness
58 " ORGAN-STOPS
all the other lingual voices of the Organ. From the point of view of
uniformity of tone the organ stop may be considered superior to the
orchestral instrument. The middle register of the stop its most
valuable portion for solo work is as satisfactory as the correspond-
ing portion of the Clarinet ; but in the low register, the tone does not
reach the rich tonality of the chalumeau of the orchestral instrument.
The harmonic upper partial tones of the Clarinet are those proper
to a covered organ-pipe; and these are present in the imitative tones
of the organ CLARINET, secured by the cylindrical form of its res-
onators and, on account of their half-length, their influence on the
motions of the tongues of the reeds. The voicer should do his ut-
most to impart to the bass and tenor of the stop as much of the
chalumeau quality as his skill can accomplish. It is possible that
some modification of, or addition to, the resonators would assist in
this direction.
COMBINATION AND REGISTRATION. Although the CLARINET,
in both its single and dual form, is par excellence a solo stop, it ad-
mits of combination and artistic registration with numerous labial
stops both simple and compound of refined and contrasting
tonalities, producing beautiful qualities of tone. It loses too much
of its peculiar voice or individuality when combined with loud-toned
labial stops; and becomes entirely absorbed when combined with
powerful lingual stops. Yet in suitable cases it never fails to be a
timbre-creator. With both unison and octave flute-toned stops of
the softer varieties it produces compound tones which demonstrate
the value of contrasting voices in registration. In combination with
the string-toned stops it produces tones strong with orchestral
coloring. In these directions the organist will find no little profit in
studying what may be designated the Registration of the CLARINET,
It is only by means of such studies that the student of the Organ
can become a master of the art and science of stop registration.
The mere following of registrations marked by the composer or
transcriber on a piece of organ music, which have been schemed, as
a rule, on some special Organ, and which may or may not be possible
or satisfactory on another Organ, will not conduct the student very
far in acquiring a personal knowledge and mastery of artistic regis-
tration. As we have said elsewhere: In organ-stop registration, as
in artistic orchestration, "there is no royal road to learning."
CLARINET FLUTE. The name which has been given by
certain English organ-builders to a half -covered wood stop, of me-
dium scale, and 8 ft, pitch. It is a species of ROHRFLOTK, differing
THEIR ARTISTIC REGISTRATION 59
from that stop in having longer stopper-handles through which
larger holes are bored; and also in having much lower mouths
usually under one-quarter their width in height and voiced with
thin upper lips and fine and close nicking. This formation and
voicing on a light wind-pressure produces a somewhat reedy tone,
so remote from that of the orchestral Clarinet, or the organ stop so-
called, as to give no support to the inappropriate name CLARINET
FLUTE. Many examples exist in English Organs by different build-
ers, varying in tonality, and commonly labeled CLARIONET FLUTE.
CLARION. Fr., CLAIRON. ItaL, GLARING. A lingual stop of
4 ft. pitch, the reeds and resonators of which are similar in all re-
spects to those of the TRUMPET, to which stop it stands in the rela-
tion of a true Octave, save in regard to its resonators, which should
be somewhat smaller in scale. As an octave stop it should be less
assertive in tone than the TRUMPET, or other unison lingual stop
with which it may be registered, yet its voice should impart richness
and brilliancy to every combination into which it enters. When
artistically voiced and finely regulated it becomes a stop of general
utility in registration, and a good solo voice on special occasions.
More care should be given to the CLARION than is usually bestowed
upon it, and organists should give it more attention. Owing to the
difficulty of constructing lingual pipes of very high pitch, it has been
usual to insert the top octave of the CLARION, 4 FT., in loudly- voiced
labial pipes. French organ-builders have frequently formed the top
octave of unison or TRUMPET pipes, with a satisfactory result.*
CLARION MIXTURE. The name given by Walker & Sons to
a labial stop, introduced by them to take the place of the lingual
CLARION; as in both the Great and Swell of the Organ in Holy
Trinity Church, Chelsea; and in the Swell of the Organ in the
Church of St. John the Divine, Kennington, London. The stop is
* " En France, le CLAIRON a toujours quatre-pieds; en Allemagne, quatre et huit. C'est un
re"gistre d'harmonie mordante et claire, comme 1'indique son nom, et destin6 a donner de la
pointe aux huit-pieds d'anches. Cependant on peut le faire chanter seul, ou melange avec les
fonds; mais dans ce dernier cas il est bon de ne pas le toucher sur les notes le plus hautes, ou il
devient d'une grande aigreur. On fait ce rSgistre detain fin et de forme conique. Comme ces
tuyaux extremes sont fort exigus, on lui donne ordinairement une reprise a la derniere octave;
et pour le toucher, on y joint le PRESTANT et la DOUBLETTE, qui dissimulent la reprise en suivent
la marche ascendante qu'est cense suivre le CLAIRON. L'habitude de mettre deux clairons au
meme clavier expose 1'ensemble des jeux de ce clavier a 6tre criard et mal d'accord. On met un
CLAIRON a chaque clavier complet, m6me a la pe"dale; il forme ainsi le sommet d'une pyramide
harmonique defe jeux d'anches des trois degr6s, quatre, huit et seize pieds. A la pSdale, son
melange est peut-e*tre plus utile que partout ailleurs; il donne aux huit-pieds d'anches, c'est-a-
dire &. la TROMI^ETTE, un vif e"clat, de m&me qu'au melange de TROMPETTS et BOMBARD ES, c'est-a-
' dire aux huit et seize-pieds. " Regnier.
60 ORGAN-STOPS
composed of an OCTAVE, TWELFTH, and FIFTEENTH, of open metal
pipes of full scale, heavily blown, and voiced to yield a powerful
tone. The effect of the stop was found, as might be expected, to be
unsatisfactory. Like too many screaming compound stops to be
found in modern Organs in this country and abroad, the CLARION
MIXTURE was valueless in artistic registration.
CLAVAOLINE, Ger. In the year 1830 Beyer, an organ-builder
of Niirnberg, invented a manual lingual stop, giving it the name
CLAVAOLINE or KLAVAOLINE, the name being rendered in both forms
in old stop-lists. From its name it would seem its inventor consid-
ered it to have some relation to the AOLINE, probably in its tone.
From the best authority we have been able to find, the stop seems
to have been a free-reed something of the nature of the PHYSHAR-
MONIKA (q. v.). The free-reeds appear to have been attached to
small blocks of wood standing directly on the wind-chest, some-
what after the fashion of the reeds of the Harmonium, and devoid
of resonators. The tone under such conditions must be extremely
feeble. The CLAVAOLINE appears in the stop-lists of the Organs in
the Town Church of Fulda, placed in the Echo; and in the Church of
St. Wenzel, at Naumburg, where it appears in the Swell.*
CLEAR FLUTE. Ger., HELLFLOTE. The stop to which this
somewhat indefinite English name was given is said to have been
first introduced by Kirtland & Jardine, of Manchester. It is an
open wood stop of 4 ft. pitch, yielding a full and clear unimitative
flute-tone, but without any particular charm or good mixing quality.
The pipes are of medium scale, the CC, 4 ft., pipe measuring 2%
inches by 3j/ inches internally; their ratio being i : \/8. The chief
characteristics of the CLEAR FLUTE pipe are its inverted mouth,
sloped block carrying the wind-way, and flat cap. It is voiced on a
copious wind supply, necessitating a moderately high mouth. The
name can, with equal propriety, be given to any stop producing a
clear unimitative flute-tone.
CGELESTINA, Lat. An open metal labial stop of 4 ft. pitch,
of small scale, and of a soft singing voice. It may be considered the
Octave of the true English DULCIANA. An example occurs in the
* Alluding to the CLAVAOLINE, Seidel says: " The Organ in St. Wenzel'a, at Naumburg; 52
registers, 3 manuals, 3000 pipes, 7 pairs of bellows. Was first built in 1613 by Joachim Tzchug,
and underwent several repairs, the last one by Beyer, on which occasion several alterations were
made in the arrangement, in order to obtain a tone as full and pompous as possible. Among tbu*
new registers introduced, the TRAVERSFLtJTE, GEMSHORN, H FT., and CLAVXOUNU, 8 KT.. tire
particularly beautiful,"
THEIR ARTISTIC REGISTRATION 61
Choir of the Organ in the Centennial Hall, Sydney, N. S. W. The
stop has been less correctly labeled CELESTINA and CELESTIANA.
COMBINATION AND REGISTRATION. The value of such a stop in
artistic and refined registration cannot well be overrated; and it is
much too seldom that OCTAVES of the nature of the CCELESTINA are
introduced in even large Organs in which some unison stops are
inserted which are, comparatively, of less general use in effective
registration. It is not very generally realized, even among good or-
ganists, that the tonal value and beauty of many a unison voice are
never fully developed until a suitable octave tone is combined with
it. This is according to the natural law of musical sound in respect
to the tones produced by open pipes, to which the octave gives an
effective reinforcement to the first and most important upper partial.
But, on the other hand, when the OCTAVE is associated with a
covered stop, it introduces a foreign element into the unison tone,
which, in its nature, does not "contain the octave in its harmonic
upper partials. Accordingly, a new tonality is created often of a
very beautiful character, and one probably impossible in the absence
of such a stop as the CCELESTINA, 4 FT. These considerations would
lead to the conclusion that a larger number and a greater variety of
OCTAVES should be provided in Concert-room Organs and other
important instruments.
COMPENSATIONSMIXTUR, Ger. Eng., COMPENSATING
MIXTURE. The stop to which the name was originally given is a
Pedal Organ compound stop of a harmonic-corroborating character,
invented by Musickdirecktor Wilke, of Neu-Ruppin, and first in-
stalled in the Organ in the Church of St. Catherine, at Salzwedel.
Its purpose was not only to give to the lower pedal notes the most
distinct speech possible, but to impart to the Pedal Organ through-
out such an even tone that rapid passages could be rendered with the
same roundness and clearness from the lowest to the highest note of
the clavier. According to Seidel, all experts who tested the stop
agreed that it fulfilled its mission. Notwithstanding such approval
the invention did not receive the consideration it deserved by the
conservative and slow-going organ-builders of Germany. Wilke's
stop was of very short compass, extending only to ten notes. It was
composed, according to Seidel, of five ranks, as follows :
I. TIERCE, 3)^ FT. CCC to GGG, eight pipes; the tone of which is gradu-
ally reduced from DDD until almost inaudible at GGG.
II. QUINT, 2% FT. CCC to AAA, ten pipes; the tone of which is reduced
from EEE in like manner.
62 ORGAN-STOPS
III. OCTAVE, 2 FT. CCC to GGG #, nine pipes ; the tone of which is reduced
from DDD in like manner.
IV. QUINT, iH FT. CCC to FFF #, seven pipes; the tone of which is re-
duced from CCC $ in like manner.
V. SIFFLOTE, i FT. CCC to FFF, six pipes; the tone of which is reduced
from CCC ft in like manner.
It is difficult to conceive on what principle Wilke designed the
stop, for it is both insufficient and inartistic. While it may have
ameliorated the lower ten notes of the heavy Pedal Organ of the
Salzwedel instrument, it certainly could do nothing to impart an
"even tone" throughout its compass. The stop, however, suggested
an artistic development which we decided to test practically, with
tonal results that proved eminently satisfactory. We included a
COMPENSATING MIXTURE, VI. RANKS, in our scheme for the Pedal of
the Concert-room Organ installed in the Festival Hall of the Louis-
iana Purchase Exposition, St. Louis, 1904. Again, and for the first
time in a Church Organ outside Germany, we inserted a COMPEN-
SATING MIXTURE, III. RANKS, in the Pedal of the Grand Organ we
designed for the Church of Our Lady of Grace, Hoboken, N. J.*
FORMATION. The following particulars and suggestions are necessarily given
as results of our own studies and practical tests ; for we are not aware of any Bng-
lish or American organ-builder having paid any attention to the stop under con-
sideration. The simple fact that it calls for more than one rank of pipes will deter
organ-builders from advocating its adoption. The stop may be composed of from
three to six ranks of open metal pipes, of the usual DIAPASON form, and, preferably,
of different scales: all the ranks to commence on CCC and to terminate on different
notes, as the judgment of the designer may decide, properly taking into considera-
tion the stop-apportionment of the Pedal Organ. The scale adopted for the first,
the longest, and the lowest pitched rank is to determine the smaller scales for the
shorter and higher-pitched ranks. No rule has been formulated in this matter,
but a regular gradation derived from a complete scale is to be recommended. It is
essential that provision be made for a regular decrease in strength of tone, pipe by
pipe, in every rank, from the CCC pipe to the top pipe, the tone of which is to be
only just clearly audible. To assist in obtaining this diminution of tone, the scales
should decrease so as to place the half diameter on the thirteenth pipe. The follow-
ing is a suggestion for the composition of a stop of three ranks :
PEDAL ORGAN COMPENSATING MIXTURE III. RANKS.
L CCC to G SUPER-OCTAVE, 4 FT. . . 32 Notes.
II. CCC to GG# SUPER-OCTAVE QUINT, 2% FT. 21 M
III. CCC to EE TWENTY-SECOND, 2 FT . , 17 "
Should a stop of four ranks be required for an important Pedal Organ, in which
*For a detailed description of this stop, see our work, "The Organ of the Twentieth Cen-
tury," page 497-
THEIR ARTISTIC REGISTRATION 63
there is a stop of 32 ft. pitch, add to the above, from CCC to D, a TIERCE,
2 7 Notes.
We suggest the following composition for a stop of six ranks, suitable for the
Pedal of a Concert-room Organ of the first magnitude:
PEDAL ORGAN COMPENSATING MIXTURE VI. RANKS.
I. CCC to G SUPER-OCTAVE, 4 FT. . . 32 Notes.
II. CCC to D TIERCE, 3> FT. . . . 27 "
III. CCC tO BB OCTAVE-QUINT, 2% FT. . 24 "
IV. CCC to GG TWENTY-SECOND, 2 FT. . . 20 "
V. CCCtoDDS TWENTY-SIXTH, i J4 FT. . 16 "
VI. CCC to BBB TWENTY-NINTH, i FT. . . 12 "
A five-rank stop may be formed by omitting the fifth rank, which is of the
least importance. It is usually desirable to have both the lowest and highest ranks
octave-sounding.
TONALITY AND REGISTRATION. While the stop is devised as a
tonal compensating agent, for the purpose of giving clearness and
richness of articulation to the lower and somewhat indeterminate
notes of the Pedal Organ unison and double stops, gradually de-
creasing in its effect toward the higher notes where its help is less
required, an office it fulfills in a satisfactory manner it is, at the
same time, a harmonic-corroborating stop of considerable value in
registration; and as such it may take the place of an ordinary MIX-
TURE with advantage to the Pedal Organ. The most desirable voice
for the stop is pure organ-tone, for that will prove most useful in
Pedal Organ registration. As has already been remarked, the tone
of each rank must be uniformly reduced in strength from that of the
first and complete rank; and each rank must be evenly reduced in
strength of tone as it ascends the scale, until at its highest note its
voice is only just clearly audible. The initial strength of tone, or
that of the CCC (4 ft.) pipe of the first rank, must be adjusted
according to the demands of the foundation unison, 16 ft., of the
Pedal Organ, and from it all graduations of tone in its own and the
other ranks have to be regulated. In registration, the COMPEN-
SATING MIXTURE will be found extremely valuable, especially when
well-marked articulation is necessary, as in Solos, Fugues, and pro-
nounced pedal passages.
We introduced, for the first time in the history of organ-building,
a COMPENSATING MIXTURE, V. RANKS, in the String-toned Sub-
division of the Third Organ in the Concert-room Organ installed at
the Louisiana Purchase Exposition. We strongly recommend the
introduction of compound harmonic-corroborating stops of the com-
64 ORGAN-STOPS
pensating class in manual divisions in which there are double stops
likely to make the bass unduly heavy. The subject is a new one, and
we recommend it to the attention of organ and tonal experts.
CONCERT FLUTE. Ger., CONCERTFLOTE. The name occa-
sionally given to an open labial stop, commonly of wood, and of 4 ft.
pitch; the tone of which imitates, as closely as practicable, that of
orchestral Flute. See ORCHESTRAL FLUTE.
CONE GAMBA. The name given to an open metal labial stop
of 8 ft. pitch, yielding an unimitative string-tone; the pipes of which
are conical, being about half the diameter at top of that at the
mouth line; hence the name of the stop. The mouth has a width of
about one-third or two-sevenths the larger circumference of the pipe,
is cut low, and voiced with the harmonic-bridge, held between, pro-
jecting ears. In some examples afrein harmonique takes the place
of the bridge and ears. The pipes are of medium scale and should be
made of high-grade alloy.
The tone of the CONE GAMBA naturally varies in different ex-
amples; but, as a rule, it is, when artistically voiced by a master-
hand, of a singing string quality, delicate, and very beautiful. The
unison stop suggests the introduction of a complete family of 16 ft.,
8 ft., 4 ft., and 2 ft. pitch, which would be of great value in artistic
registration, combining admirably with the family of the LIEBLICH-
GEDECKTS, the other softer flute-toned stops, and the reed-toned
lingual stops. A CONE GAMBA, 16 FT., of full compass, was in-
serted in the Choir of the Organ erected by Hill, in 1863, in the nave
of York Minster. This fine stop was extremely effective in com-
bination. As a timbre-creating and tone-supporting stop the CONE
GAMBA, 8 FT., will be found very useful in artistic registration.
CONTRABASS, Ger. Ital., CONTRABASSO, Pr,, CONTRB-
BASSE. Eng., DOUBLE BASS. The name appropriately used to
designate an open labial stop of 16 ft, pitch, of wood or metal, voiced
to yield a tone imitating, as closely as practicable, that of the Double
Bass of the orchestra. In French Organs the CONTRE-BASSE is in-
variably a Pedal Organ stop, appearing as such in the Organs in the
Cathedrals of Paris, Amiens, and Orleans, and the Churches of Saint-
Eustache, Saint-Sulpice, and la Madeleine, Paris. French organ-
builders and experts do not seem to have realized the possibility- of
introducing so grave an imitative string-toned stop in a manual
division. It does not appear, save in the Pedal, in Cavaill^-CoU's
scheme for the Monumental Organ to be installed in St. Peter's,
THEIR ARTISTIC REGISTRATION 65
Rome. But no French organ-builder ever contemplated the desir-
ability of introducing an independent string-toned division in the
Organ : indeed, no one seems to have ever done so, until we realized
its necessity, and introduced one in the Organ installed in the Fes-
tival Hall of the Louisiana Purchase Exposition.*
FORMATION. The pipes of the CONTRABASS, 16 FT., are in the best examples of
wood, and of what may be considered small scales; although metal pipes can be
used with satisfactory results, especially in the higher octaves of the manual stop.
For the Pedal Organ stop wood is to be preferred, as it aids in producing a tone
fuller in character than is commonly yielded by small-scaled metal pipes. The
best example of the stop which has been directly examined and tested by us is that
labeled VIOLONBASS, 16 FT., in the Pedal of the Organ constructed by Edmund
Schulze for the Church of St, Peter, Hindley, Lancashire, England. In this fine
stop the pipes are square and of the following measurements: CCC, 16 ft., 5^ ins.
by 5 1 A ins. ; width of mouth 5^ ins. ; height of mouth i # ins. CC, 8 ft., 3 J^ ins.
by 3% ins. ; mouth 3% ins. by I in. C, 4 ft., 2 ins. by 2 ins. ; mouth 2 ins. by % in.
The pipes have sunk blocks, the German form of mouth, caps hollowed externally,
no projecting ears, upper lips cut almost to a sharp edge, and furnished with har-
monic bridges of an unusual form.f No wood pipes of imitative string-tone were
known in England prior to their appearance in the Organs installed by Schulze in
the Parish Church of Doncaster and St. Peter's Hindley. They were first made in
this country by Roosevelt, from models furnished by us in 1883. By special
voicing with the harmonic-bridge, covered pipes can be made to yield imitative
string-tone, the prominence of the upper partial the Twelfth contributing
to the tone. The covered stop, owing to its moderate dimensions, is most suitable
for a manual division in all save an Organ of the first magnitude.
TONALITY AND REGISTRATION. In tonality, the voicer of the
CONTRABASS should aim at the imitation of the peculiar tones of the
different stiings of the orchestral instrument, giving greater body to
the tones of the 8 ft. and higher octaves, so as to differentiate them
properly from the corresponding tones of the VIOLONCELLO, 8 FT.
This is important when the stop is associated with a VIOLONCELLO
in a manual string-toned division, as it should be in a Concert-room
Organ. The tones of the VIOLONBASS in the Hindley Organ are so
closely imitative, even to the rasp of the bow on the strings, as to be
deceptive to the ear. That is as it should be, if such an important
voice is to be of full value in the artistic registration of orchestral-
toned stops. The building up of a mass of orchestral string-tone in
the Organ, which we were the first to render possible, is a matter in
registration of the greatest importance in the artistic and proper
rendition of orchestral scores or transcriptions. The student of the
* See "The Organ of the Twentieth Century." Page 505.
t Full details of the mouth are given in "The Art of Organ Building," Vol. II., p. 470;
anfl, in "The Organ of the Twentieth Century," p. 443.
S
66 ORGAN-STOPS
art of registration will soon discover the value of so effective and
colorful a voice as that of the CONTRABASS. When artistically voiced
and perfectly regulated it is a beautiful voice for a bass solo; and
may for this purpose be combined with the CONTRAFAGOTTO, or with
a good Vox HUMANA in sub-octaves. It is to be regretted that so
little attention has been paid to the stop by the organ-builders and
experts of this country. It must, however, appear in the Organ of
the Twentieth Century.
CONTRA-BOURDON, SUB-BOURDON. Ger., UNTERSATZ.
Fr., SOUS-BASSE. The name given to a covered wood stop of 32 ft.
pitch, properly belonging to the stop-apportionment of the Pedal
Organ, but occasionally appearing, in an incomplete form, in the
First or Great Organ. It is occasionally resorted to as a Pedal Organ
stop when either space or funds prevent of the DOUBLE DIAPASON or
SUB-PRINCIPAL, 32 FT., being introduced. As a substitute it is any-
thing but satisfactory tonally. The tone of the stop is that common
to the ordinary BOURDON, 16 FT.-, being carried an octave lower,
where its voice becomes painfully indeterminate, as might be ex-
pected. It appears in Pedal Organs under the following names:
UNTERSATZ in the Organs in the Cathedral of Merseburg and the
Royal Catholic Church, Dresden. GROSSUNTKRSATZ in the Organ
in the Church of Waltershausen, Gotha. SUBBASS in the Organ in
the Church of St. Michael, Hamburg. GRAND BOURDON in the
Pedal of the Organ, by Cavaill6-Coll, in the Church of Saint Vincent
de Paul, Paris. CONTRA-BOURDON in the Organs in the Auditorium,
Chicago, and the Centennial Hall, Sydney, N. S, W.
When the CONTRA-BOURDON, 32 FT., is introduced in the Great
Organ, it commonly is of short compass, not extending below tenor
C. It appears under the following names: UNTERSATZ on the First
Clavier of the Riga Cathedral Organ. BORDUN, 32 FT. , in the Haupt-
werk of Merseburg Cathedral Organ, and in the Second Division of
the First Organ in the instrument in Schwerin Cathedral. SUB-
BOURDON in the Great of the Organs in the Parish Churches of Leeds
and Doncaster, England. On tonal grounds we do not advocate the
introduction of this incomplete stop in any manual division; while
we are of opinion that a pure and soft organ-tone, of 32 ft. pitch,
would be valuable in the First Organ of a large Concert-room Organ.
See DOLCIANO PROFUNDO.
CONTRA-DULCIANA. An open metal labial stop of small
scale, of 32 ft. pitch in the Pedal Organ and 16 ft. pitch when in-
troduced in a manual division, as in the Choir of the Organ in the
THEIR ARTISTIC REGISTRATION 67
Centennial Hall, Sydney, N. S. W. It is also named DOUBLE DUL-
CIANA (g. v.). We have introduced it, in an extended form and for
a double purpose, in the Pedal Organ of our scheme for the Concert-
room Organ of the Twentieth Century.* See DOLCIANO PROFUNDO.
CONTRAFAGOTTO, ItaL Ger., CONTRAFAGOTT. Eng., DOU-
BLE BASSOON. A lingual stop, of 16 ft. pitch, the resonators of
which are of small scale and either of wood or metal. In the former
material they are square and cf inverted pyramidal form : when of
metal they are inverted conical in form. When properly scaled,
voiced, and regulated, its tone closely approaches that of the orches-
tral instrument of the same name. While usually and correctly
placed in the Pedal Organ, where its voice furnishes the true bass to
the manual FAGOTTO and OBOE, its presence is, however, to be
greatly desired in a suitable manual division, where its rich and
quiet tonality would contribute a remarkable coloring to every
effective registration into which it might enter. In our opinion,
there is no lingual stop of 16 ft. pitch so generally useful on the
manual claviers; and we know it to be so important, tonally, that a
series of most effective and refined registrations can be based upon
it. We introduced it, accordingly, in the First Subdivision ex-
pressive of the Third Organ in the instrument installed in the
Festival Hall of the Louisiana Purchase Exposition; and we also
inserted it in the Second Subdivision expressive of the Third
Organ in the Grand Organ in the Church of Our Lady of Grace,
Hoboken, N. J.
CONTRA-GAMBA, ItaL A labial stop, of 16 ft. pitch. Its
pipes are of medium scale, of metal, cylindrical in form, and open.
It is introduced in both manual and pedal departments of the Organ,
but usually, when it is introduced, it appears in a manual division,
as in the Great of the Willis Organ in the Church of St. Margaret,
Liverpool, and in the Swell of the Walker Organ in the Church of
St. Matthew, Northampton, England. In the latter Organ it has a
covered wood bass, dictated by its position in a swell-box. An
example appears in the Great of the Roosevelt Organ in the Cathe-
dral of the Incarnation, Garden City, L. I. When introduced in the
Pedal Organ, it is of unison pitch, and may be labeled, simply,
GAMBA, 16 FT., as in the -Organ in the Centennial Hall, Sydney,
N. S. W. When properly voiced, the CONTRA-GAMBA has a full
string-like tone, but not so pronounced as to be imitative. The
* " The Organ of the Twentieth Century." Pages 291, 297, 317, 319.
68 ORGAN-STOPS
stop is less valuable now than it used to be when string-toned stops
were few and very little developed in comparison to what they are
to-day. VIOLS now take the place of the German GAMBAS, and their
prompter English equivalents.
CONTRA-OBOE, ItaL Eng., CONTRA-HAUTBOY. A very un-
common lingual stop of 16 ft. pitch, the form and tone of which
resemble those of the ordinary OBOE, 8 FT. A fine and note-
worthy example, voiced by George Willis, exists in the Swell of the
Organ in St. George's Hall, Liverpool, labeled CONTRA-HAUTBOY,
1 6 FT. This valuable stop should find a place in the " wood-wind"
division of the Concert-room Organ, where it would be associated
with the unison OBOE and, possibly, with the OCTAVE OBOE, 4 FT.,
forming a complete family.* All would afford valuable voices in
artistic registration. While the CONTRAFAGOTTO may be regarded
as affording the bass to the OBOE, 8 FT., the CONTRA-OBOE would be
found, in many registrations, to produce closer and clearer effects
when combined with the unison stop : but, after all is said, much
depends on the voicing and tonality of the respective stops.
CONTRAPOSAUNE, Ger. A powerfully-voiced lingual stop
of 32 ft. pitch in the Pedal Organ and 16 ft. pitch on the manuals.
It appears under the name in the Pedal of the Organ built by Buch-
holz for the Cathedral of Cronstadt. In the generality of German
Organs it appears as simply POSAUNE, 32 FT., or 16 FT. An example
of the CONTRAPOSAUNE, 16 FT., exists in the Great of the Organ in
the Centennial Hall, Sydney, N. S. W. ; and a CONTRAPOSAUNE,
32 FT., is inserted in the Pedal of the same Organ, and in that of the
Organ in the Parish Church of Doncaster. The pipes of the stops
have open-reeds and broad tongues; their resonators being of the
usual inverted conical form and of large scale, producing tones in-
tended to imitate those of the Bass Trombones of the orchestra
played fortissimo. The resonators are sometimes constructed of
wood, square, and inverted pyramidal in form; but resonators of
thick zinc are to be preferred. Sheet iron may be used in the 32 ft,
octave.
CONTRAPRINZIPAL, Ger. The term which has been used to
designate the open metal labial stop known in English nomenclature
as the OPEN DIAPASON, 32 FT., in the Pedal Organ and of 16 ft. pitch
when inserted in a manual division.
* See "The Organ of the Twentieth Century." Pages 307, 308.
THEIR ARTISTIC REGISTRATION 69
CONTRA-SALICIONAL, DOUBLE SALICIONAL. As the
name implies, this is a SALICIONAL of 16 ft. pitch. In material and
formation its pipes are in all respects similar to those of the manual
unison stop. The stop is valuable in the Pedal Organ, to which it
properly belongs, and, accordingly, its tone should be fuller than
that of the manual stop, especially if it is inclosed and rendered
expressive. See SALICIONAL.
CONTRA-SAXOPHONE. The stop of 16 ft. pitch, the voice
of which furnishes the true bass to the manual SAXOPHONE, 8 FT.
It is, when of imitative quality, a valuable voice in the Pedal of a
Concert-room Organ, where it should be inclosed and rendered ex-
pressive. The stop can be made with either free- or striking-reeds.
On account of the smoothness of the tones of the orchestral Saxo-
phones, we are of opinion that their closest imitation is to be ob-
tained from free-reed pipes ; but the most favorable form of resona-
tor has yet to be discovered. The tone of the CONTRA-SAXOPHONE
should imitate as closely as practicable that of the Double Bass or
Bourdon Saxophone. There is a good free-reed example in the
Pedal of the Chamber Organ in Ham House, Richmond, Surrey,
England. While an almost deceptive imitation of the tones of the
Saxophones have been obtained from labial pipes of wood of 8 ft.
pitch, we are not aware of any attempt having been made to pro-
duce a corresponding labial CONTRA-SAXOPHONE. See SAXOPHONE.
CONTRA-TROMBONE, A lingual stop of 32 ft. pitch in the
Pedal Organ and 16 ft. pitch on the manuals. In formation it is
similar to the CONTRAPOSAUNE, but its resonators are of a smaller
scale. An example of the CONTRA-TROMBONE, 32 FT., exists in the
Pedal of the Walker Organ in York Minster. Another example
exists in the Pedal of the Organ in Leeds Parish Church, where it
speaks effectively on a wind of only 3^ inches an eloquent argu-
ment in support of our plea for low pressures, now so thoughtlessly
neglected by the inartistic lovers of coarse tones and musical noise
in this country. A CONTRA-TROMBONE, 16 FT., occurs in the Great
of the Concert-room Organ in the Town Hall, Leeds.
The tone of the CONTRA-TROMBONE is intended to imitate that
of the Bass Trombone, in G, of the orchestra. The voice of the
organ-stop extends far below that of the orchestral instrument, but
this is a gain of great value which no authority on the Organ will
question. While the tone is similar to that of the CONTRAPOSAUNE,
it should not be so brassy and powerful. The value of the CONTRA-
70 ORGAN-STOPS
TROMBONE, 16 FT., in the "brass-wind" division of a Concert-room
Organ of the first magnitude would, unquestionably, be great, for
not only would it enter with great effect into assertive registration,
but it would prove a most dignified and impressive solo voice.
Under the name of CONTRA-TROMBONE, Messrs. Hill & Son, of
London, have inserted a monster lingual stop, of 64 ft. pitch (if such
a pitch can be recognized as obtaining in the range of audible or de-
terminate musical sounds, a proposition we cannot accept), in the
Pedal of the Organ in the Centennial Hall, Sydney, N. S. W. We
examined this unique stop, and heard it during a recital on the Organ
before it left the factory. While its slow and powerful vibrations
succeeded in shaking us bodily, its noise did not impress our musical
sense. It seemed to support the old saying : ' ' The game is not worth
the candle,"
CONTRA-TUBA. The most powerfully-voiced lingual stop
of 1 6 ft, pitch introduced in the Organ. It is commonly voiced on a
wind-pressure of between fifteen and thirty inches. Its resonators
are of the inverted conical form, and are necessarily made of thick
metal, now usually zinc, so as to withstand the extreme vibration;
and they are of large scale. An example exists in the Solo of the
Organ in the Centennial Hall, Sydney, N. S. W. The tone of the
stop is intended to imitate that of the Bass Tuba of the orchestra,
but is much more powerful. It is, accordingly, singularly grand and
impressive, but of little general use unless it is inclosed and rendered
flexible and expressive, which it should invariably be in all properly
appointed Organs. In the Sydney Organ neither it nor any of the
Solo Organ stops are inclosed and rendered expressive; surely an
anomalous absurdity in a Concert-room Organ, showing a want of
artistic taste and common-sense.
CONTRA-VIOLONE, Ital A labial stop, of 16 ft. pitch, the
pipes of which are open and of medium scale, formed of either metal
or wood, or of wood in the lower and metal in the higher octaves.
The CONTRA- VIOLONE, 16 FT,, is strictly a manual stop; the corre-
sponding stop in the Pedal Organ being properly labeled VIOLONE,
1 6 FT. The tone of the stop should imitate that of orchestral instru-
ment of the same name. While practically identical with the CON-
TRABASS, 1 6 FT. (g. v.), its tone should be smoother and somewhat
less assertive. A very beautiful example of the class, from the
master hand of Edmund Schulze, is to be found in the Great of the
Organ in St. Peter's, Hindley, Lancashire. The stop is formed of
wood pipes from CC to B, and of metal pipes from c 1 to the top note.
THEIR ARTISTIC REGISTRATION 71
The proper place for the CONTRA-VIOLONE, 16 FT., is in the string-
toned division or subdivision of the Concert-room Organ. In regis-
tration for the production of massed string effects of an orchestral
character it is extremely valuable. The CONTRA-VIOLONE, in its
perfect form and tone, is extremely rare; but it can be easily made
now by an artistic voicer of string-toned pipes.
CONTRE-BOMBARDE, Fr. The lingual stop of 32 ft. pitch,
introduced in the Pedal of most of the large Organs in French cathe-
drals and churches. It is to be found in the Organs in the Cathedral
of Notre-Dame and the Churches of Saint-Sulpice and Saint-Eus-
tache, Paris, and in the Royal Church, Saint-Denis, and Saint-Ouen,
Rouen. It was introduced by Cavaille-Coll in the Organ in the
Albert Hall, Sheffield. The stop has resonators of the usual inverted
conical form and of large scale. As common in French lingual stops,
the tone of the CONTRA-BOMBARDE is somewhat dry and hard;
widely different from the rich, round, and velvety tones of the Willis
stops. In its usual exposed position, the use of so grave and powerful
a voice is necessarily very limited, and hardly worth its great cost;
but inclosed in a swell-chamber, its tonal value would be increased
more than tenfold. A crescendo on such a voice would have a stupen-
dous effect one never yet heard on the Organ; but will be when
our scheme for the Concert-room Organ of the Twentieth Century is
carried into effect.*
COR ANGLAIS, Fr. Ital., CORNO INGLESE. Eng., ENGLISH
HORN. A lingual stop of 8 ft. pitch. It has been made in two ways;
namely, with striking- and free-reeds, and probably the most success-
ful examples, tonally, have been of the latter class. The stop, in its
free-reed form, has long been a favorite with the French organ-
builders, being introduced in most of their more important instru-
ments, in either its 8 ft. or 16 ft. pitches. Of the former pitch, it was
inserted by Merklin in the Positif of the Organ in the Cathedral of
Senlis; and in the Clavier de Bombarde of his Grand Organ in the
Church of Saint-Eustache, Paris. It was introduced, in the same
unison pitch, by Cavaille-Coll, in the Clavier du Grand Orgue of the
Organ in the Royal Church of Saint-Denis. Of 16 ft. pitch, it was
inserted by Cavaill6-Coll in the Positif of the Organ in the Church
of Saint-Ouen, Rouen; in the Solo Expressif of his scheme for the
Monumental Organ in St. Peter's, Rome; and in the Recit Expressif
of the Concert-room Organ in the Albert Hall, Sheffield, in which it
* See "The Organ of the Twentieth Century." Pages 273-332.
72 ORGAN-STOPS
is a striking-reed. In his later work, Cavaille-Coll abandoned the
free-reed. The COR ANGLAIS, in both its pitches, has been intro-
duced in a few English Organs, the pipes having, for the most part,
been imported from France. We inserted a CORNO INGLE SE, 8 FT.,
in the Second Organ of the instrument installed in the Festival Hall
of the Louisiana Purchase Exposition. It is a free-reed stop im-
ported from Germany.
FORMATION. The pipes of the COR ANGLAIS arc invariably made of metal;
and, in the most satisfactory examples of the stop, have resonators of a compound
form that has been arrived at as most effective after many experiments. The
form is that of a slender inverted conical tube, surmounted by an expanded, res-
onant chamber formed of two truncated cones soldered together and to the tube,
in the manner shown in the drawing of the complete pipe given in Fig. 7, Plate I.
The scale and relative proportions of the parts forming the resonator have a great
influence not only on the strength but also on the timbre of the tone produced.
The reed and tongues, in the best examples, are similar in form and general treat-
ment to those of the OBOE. The value of the stop is considerable; but the time
and trouble involved in the proper formation of its resonators have largely pre-
vented organ-builders from encouraging its introduction. It should, however,
find a place in every Concert-room Organ of any pretensions.
TONALITY and REGISTRATION. The tone of the COR ANGLAIS of
the Organ is intended to imitate, as closely as possible, that of the
orchestral instrument of the same name, which is, strictly considered,
a Tenor Oboe. The Italians call it the Oboe di Caccia as well as
Corno Inglese; while the Germans use the name Englisches Horn.
Speaking of the instrument, Berlioz remarks: "Its quality of tone,
less piercing, more veiled, and deeper than that of the Oboe, does not
so well as the latter lend itself to the gaiety of rustic strains. Nor
could it give utterance to anguished complainings; accents of keen
grief are almost interdicted to its powers. It is a melancholy,
dreamy, and rather noble voice, of which the sonorousness has
something of vague, of remote, which renders it superior to all
others, in exciting regret, and reviving images and sentiments of the
past, when the composer desires to awaken the echo of tender mem-
ories." These remarks, by so great an authority on orchestration,
ought to enthuse the voicer in the production of the organ COR
ANGLAIS, and inspire the organist who finds a sympathetic one in an
Organ. A good COR ANGLAIS is a beautiful solo stop, while it is
extremely valuable in artistic registration, producing in combina-
tion with softly-voiced stops of contrasting and harmonic-corrobor-
ating qualities compound voices of remarkable tonalities. At the
present time organists and others in this country have little oppor-
tunity of judging the merits of the stop. Many attempts have been
THEIR ARTISTIC REGISTRATION 73
made to produce the characteristic voice of the Cor Anglais by regis-
tration ; for instance, as we are informed, a distinguished organist
succeeded in producing an almost perfect imitation on the Choir of
the Organ in York Minster, by combining the GAMBA, 8 FT., CLARI-
NET, 8 FT. , and GEMSHORN, 4 FT. We venture to think a better imi-
tation would have been the result had an OBOE, 8 FT., been intro-
duced instead of the fuller-toned CLARINET; but the Choir Organ
does not contain one; neither, alas for art, is the Choir Organ
expressive.
COR DE BASSET, Fr. The name employed by French organ-
builders to designate the lingual stop more commonly known under
the Italian name CORNO DI BASSETTO (q. v.).
COR D' HARMONIE, Fr. This term, associated with HAUT-
BOIS, 8 FT., occurs in the stop apportionment of the Positif of the
Cavaill^-Coll Organ in the Royal Church of Saint-Denis. It would
appear, in this connection, to be applied to the bass extending
downward the tone of the HAUTBOIS, which would properly be of
Bassoon quality, and, accordingly would be furnished by BASSOON
pipes from CC to A, almost two octaves.
CORDEDAIN. A labial stop of metal and of 4 ft. pitch, which
is inserted in both the Choir and Echo of Silbermann's Organ in the
Church of St. Thomas, at Strasbourg. The tones of the stops are
described of a bright flute quality, almost imitative.
COR DE NUIT, Fr. Gear., NACHTHORN. Ital., PASTORITA.
An open or covered stop of metal or wood, and of 8 ft., 4 ft., and 2 ft.
pitch. The tone is a combination of flute and soft horn qualities.
See NACHTHORN.
CORMORNE, Fr. The name which has occasionally been
given by French organ-builders to a lingual stop of metal and 8 ft.
pitch. It is probably formed from cor horn, and morne sombre or
mournful. But as the stop is commonly met with in the Cavaille-
Coll Organs labeled CROMORNE, the term may be a corruption of the
German name KRUMMHORN. See CROMORNE.
CORNAMUSA, Ital. Fr., CORNEMEUSE. This name, which
was originally used to designate an instrument of the Bagpipe
family, has been occasionally applied by Italian and French organ-
builders to a labial stop of 16 ft. pitch. Examples are rare, but one
exists, according to Hopkins, on the second manual of the Organ in
74 ORGAN-STOPS
the Church of Santissimo Crocifisso, at Como. It is evidently a
covered stop of wood, and probably derived its name from the drone-
like character of its tone. Examples of the Cornemeuse, in Bagpipe
form, and of eighteenth century date, are to be seen in the Crosby
Brown Collection of Musical Instruments in the Metropolitan
Museum of Art, New York City.
CORNET. A compound labial stop of several ranks of metal
pipes. There are two distinct varieties of the stop, which may be
called the ancient and the modern and which hold different offices
in the tonal economy of the Organ offices varying greatly.
The old and original form comes first in order. It was usually
formed of three, four, or five ranks of large-scaled pipes voiced to
yield powerful and dominating tones.* Its compass seems never to
have descended below tenor C, while it usually, and invariably in
old English Organs, commenced at middle c 1 . The lowest rank,
which was invariably of unison pitch, was, probably on account of
the size of its pipes when the stop descended to tenor C, of covered
pipes. In French examples it was usually a BOURDON, 8 FT. All the
other ranks were of open metal pipes. The stop was much in favor
with the old English organ-builders. In Bernard Smith's Organ
built, in 1684, for the Temple Church, London, were two CORNETS
one of five ranks in the Great and one of three ranks in the ' ' Ecchos. "
In the Organ built by John Harris, of London, in 1 740, for the Parish
Church of Doncaster there were also two CORNETS, of five and three
ranks. All these stops extended from middle c* to the top note.
Sometimes, as the CORNET required considerable space for its ac-
commodation, its pipes were planted on a special wind-chest, ele-
vated above the main wind-chest, and connected thereto by a series
of metal tubes or conveyances one for each note of the CORNET.
When disposed in this manner the stop was designated MOUNTED
CORNET. A MOUNTED CORNET, of 4 ranks, was inserted in the
* "Le CORNET, jeu bruyant, clair et dominateur de 1'harmonie entiere de 1'orgue. Un Cornet
complet dans sa facture a par note cinq tuyaux de grosse taille, dont chacun porte un nom et un
caractere diff6rents, Le premier tuyaux est un Bourdon de huit ; le second un Prestant; le troisi-
eme un Nusard, quinte superieure du Prestant; le quatrieme une Quarts de Nasard, octave du
Prestant; le cinquieme enfin est une Tierce au-dessus de cette octave. Ainsi, Vut du Cornet est &
la fois ut de huit-pieds en Bourdon, ul de quatre en Prestant, sol de trois en Nasard, ut de deux en
Quarts enfin mi d'un-pied cinqseptiemes.
"Ainsi, dans toute sa longueur, le Cornet appuie chacune de ces notes ou de ses marches,
comme on dit, sur cinq notes & la fois; il y a done cinq tuyaux sur marche. Son re'gistre est done
un ensemble de cinq r6gistres, cinq ranges de tuyaux au lieu d'une, comnie le jeux simples.
Mais le Cornet n'a pas toujours toute la longueur du clavier; quoique ces cinq range 1 es soient &
1'unisson du Bourdon, du Prestant, etc,, cependant clles on different par la taille, qui est plus
grosse et 1'harmonie plus forte," Rcnier,
THEIR ARTISTIC REGISTRATION 75
Great of the Organ built by Snetzler, in 1777, for the Parish Church
of Rotherham, England: and we understand, from a pamphlet in
our possession, that the stop still exists in the Organ under that
name. Schulze introduced MOUNTED CORNETS in the Great and
Swell Organs of the instrument he built for Doncaster Parish Church
in 1862. These still remain in the renovated Organ. In Bernard
Smith's Organs the CORNET was never mounted.
In all cases the scales of the pipes of the old CORNET were larger
than those of the corresponding pipes of the major DIAPASON or
PRINCIPAL, and were much more loudly voiced. The compound tone
of the CORNET was, on account of the large scale of its pipes, flutey
in character : and owing to its highest pitched rank being a Tierce,
and uncovered by a fifth- or an octave-sounding rank, the voice of
the stop was necessarily somewhat harsh and cutting.
The old CORNET was chiefly used for playing out the melody of
the Chorals, its powerful voice rendering it highly effective for such
an office.* Being unbroken in its ranks, it was found serviceable in
covering the breaks in the screaming MIXTURES of the time. In
England the old CORNET was frequently used in the rendition of
what were known as Cornet Voluntaries, the chief features of which
were runs and figurative passages on the stop by the right hand.
The best known of such compositions are those by Stanley, Dupuis,
Bluet, and Russell, all of whom wrote between the years 1726 and
1813-
The old, short-compass, and loud-voiced CORNET finds no place
in the modern Organ ; and the name is now applied to stops of widely
different construction and office. At the present time the name is
usually given to a compound harmonic-corroborating stop of full
compass, composed of five or more ranks of small-scaled and high-
pitched pipes, voiced to yield a bright and singing quality of pure
organ-tone. But different qualities of tone may be obtained by
using flute-toned or string-toned pipes of small scales. Such being
* " CORNET, or 'CORNETTO,' is a mixture of a very wide measure, which begins generally at
c 1 or at the G below, and goes through the upper octaves of the manual. It has a strong intona-
tion, and a horn-like tone, which is well adapted for filling out. Sometimes, when hymns are to
be sung with a melody which is not familiar to the congregation, this register will be found very
efficient for the purpose of making the melody prominent, since the right hand plays the melody
upon that manual which contains the CORNET, while the left hand plays the accompaniment
upon some other manual, for which weaker registers are drawn. This MIXTURE has sometimes
five ranks, 8, 4, 3%, 2, and i % feet; sometimes four ranks, 8, 4, 2%, and i % feet; and some-
times three ranks, 4, 2%, and i % feet. In France, the lowest rank of this register is nothing
but a ROHRFLOTE, 8 FT. Wilke deems it best to construct the CORNET with three ranks only,
but so that the lowest of them is a fifth, and next an octave, and the last a third, 5 M ** 4
ft., and 3 > ft., or 2 % ft.. 2 ft., and i % ft. The latter arrangement is better suited for small
Organs, the former for large ones." Seidel.
76 ORGAN-STOPS
the case, it will be seen that the modern CORNET is the antithesis in
every respect of the loud and harsh stop of the old masters. It is
usual and desirable in labeling the CORNET to qualify the name, so as
to convey some idea of its general tonality, accordingly, such terms
as DULCIANA CORNET, DOLCE CORNET, VIOL CORNET, etc., are ap-
propriate. The following is the composition of a five-rank stop,
formed of DULCIANA pipes, which has been tested and found highly
satisfactory in registration :
DULCIANA CORNET V. RANKS.
CC to BB
c to B ....
c r to b 1 ....
c 3 to c< ....
It will be observed that the middle rank is third-sounding
throughout, and that it is properly covered by the fourth and fifth
ranks, while the fifth rank is octave-sounding throughout all its
breaks. It would be difficult to devise a more satisfactory composi-
tion for a stop of this beautiful tonality. In the Swell of the Concert-
room Organ in the Music Hall, Cincinnati, there is a CORNET of six
ranks of small-scaled pipes, the composition of which is here given:
DOLCE CORNET VI. RANKS.
cc to BB . . . 15 19 22 26
C to B . . 12 15 19 22
c i to b 1 . . . 8 12 15 19 2;
c 2 to c 4 . . , i 5 8 ro 12
Artistically voiced and scientifically graduated in strength of
tone throughout its ranks and breaks, such a CORNET would prove
extremely valuable in registration. In this direction, it is to be ob-
served that the composition in the lower two octaves CC to B
points to considerable brilliancy; that in the middle octave c 1 to
b 1 to sufficient richness of tone; while that in the two higher oc-
taves, c 2 to c 4 , which belongs to the 16 ft. harmonic series, points
to great fullness, just where it is most required.
There is another form of modern CORNET which more fully de-
serves the name, and one we strongly recommend for general intro-
duction in all important Organs : we allude to the form in which all
the ranks are carried throughout the compass without a break. Of
this form we introduced two special CORNETS in our scheme for the
Organ installed in the Festival Hall of the Louisiana Purchase Ex-
position. One of four ranks in the Expressive Subdivision of the
THEIR ARTISTIC REGISTRATION
77
First Organ, composed of a SEVENTEENTH, NINETEENTH, SEPTIEME,
and TWENTY-SECOND: and the other, a VIOL CORNET IV. RANKS
(muted), in the string-toned Expressive Subdivision of the Third
Organ, composed of a TWELFTH, FIFTEENTH, NINETEENTH, and
TWENTY-SECOND. The pipes are of tin and of small scale. All Pedal
Organ MIXTURES which have no breaks or short ranks in their com-
pass are strictly CORNETS. The largest and most remarkable ex-
ample of this class of compound stop is that in the Pedal of the
Organ in the Wanamaker Store in Philadelphia, Pa. It consists of
ten complete ranks of open metal pipes, yielding pure organ-tone,
the principal rank being a DIAPASON, 16 FT. As a corroborating
stop it belongs to the 32 ft. harmonic series, and comprises four
octave, two third-, and four fifth-sounding ranks as follows :
PEDAL ORGAN GRAND CORNET X. RANKS.
1. DIAPASON .
2. QUINT .
3. OCTAVE
4. TIERCE
5. OCTAVE QUINT
Metal. 1 6 feet
" 10% "
8
6% "
6. SUPER-OCTAVE Metal. 4 feet
7. OCTAVE TIERCE " 3^ "
8. TWELFTH . " 2% "
9. FIFTEENTH . " 2 "
10. NINETEENTH " iM "
The compound tone of this stupendous CORNET is that of a mag-
nificent lingual stop, and when registered with the DOUBLE DIA-
PASON, 32 FT., or the CONTRA-BOMBARDE, 32 FT., or with both, it
produces a majestic harmonic tonal structure impossible on any
other Pedal Organ ever constructed. The only other stop of the
same class, in any way approaching that just described, known to
us is the one labeled GRAND BOURDON in the Organ in the Cathedral
of Riga. It is composed of the following five complete ranks of open
wood pipes: PRINCIPAL, i6FT. QUINT, IO%FT. OCTAVE, 8 FT.
TIERCE, 6^ FT. SUPER-OCTAVE, 4 FT.
TONALITY AND REGISTRATION. The value of the compound
harmonic-corroborating CORNET of the DULCIANA or DOLCE class
in artistic registration depends entirely on its tonality the product
of its correct composition, the scales of its several ranks, and, above
all, of its fine voicing, and the scientific graduation of the tones of all
its ranks and their breaks in accordance with the natural laws of
musical sounds. Provided the CORNET is correctly constructed along
all these lines, it is a stop to be treasured in every Organ in which it
is introduced; and will become the delight of the organist in his
registrations, forming with stops of all tonalities combinations of
rare beauty and charm. From long experience with just such a
CORNET we can speak with authority in this direction. We have
78 ORGAN-STOPS
found it to combine with labial and lingual stops of different pitches,
producing numerous very beautiful compound tones which are ab-
solutely unknown to organists who preside at Organs unprovided
with such refined tonal aid; and which have only the crude and
screaming MIXTURES which characterize the prevailing inartistic
and rule-of -thumb methods of organ tonal appointment. Before a
truly satisfactory condition of tonality, offering the maximum of
opportunity for artistic registration, obtains in the Organ, all the
compound stops CORNETS and MIXTURES must undergo a radical
change in treatment: more art and science must be concentrated on
their production, and more time and care must be devoted to their
correct regulation and graduation of tone.
CORNET A PA VILLON, Fr. The name given by Cavailte-
Coll to a lingual stop, of 8 ft. pitch, inserted in the Grand-Orgue of
the instrument constructed by him, in 1841, for the Royal Church
of Saint-Denis. The name was suggested by the form of the resona-
tors employed, just as in the case of the FL^TE A PAVILLON, the
pipes of which are surmounted by bells. Another example exists in
the Organ in the Church of Saint Vincent de Paul, Paris.
CORNETTINO, Ital. Ger., ZINK. A lingual stop of 2 ft.
pitch, and necessarily of short compass. This favored its introduc-
tion in the Pedal Organ. An example, labeled CORNETTINO, exists
in the Erstes Pedal of the Organ in St. Paul's Church, Frankfurt am
Main, built by Walcker. Respecting this stop Schlimbach remarks :
"CORNETTINO. Bin Rohrwerk, sell das zu den Posaunen gebrauch-
liche, unter dem Namen Zinken bekannte Discantinstrument nach-
ahmen. Es ist klein, und sollte eigntlich nur durchs halbe Clavier
gehen, indem man keine Basszinken hat," See ZINK.
CORNO DI BASSETTO, Ital. Fr., COR DE BASSET. Ger.,
BASSETHORN. A lingual stop of 8 ft. pitch, the pipes of which are
formed, in most examples, after the fashion of those of the CLARI-
NET. Topfer describes the stop as made of labial pipes; but, in its
best form, it has always been a lingual stop, made with free- or
striking-reeds, preferably the latter. The stop is by no means a
common one and few examples are to be found. As a manual stop
it is to be found on the third manual of the Organ in the Cathedral
of Lund, Sweden. A fine example exists in the Solo of the Organ in
St, George's Hall, Liverpool; and we inserted one in the Expressive
First Subdivision (Wood-wind) of the Organ installed in the Fes-
tival Hall of the Louisiana Purchase Exposition. As a Pedal Organ
PLATE IT
'17
CORNO DI
BA.SSETTO
'16
FL-CTTE
A PA VILLON HARMONIQUE
THEIR ARTISTIC REGISTRATION 79
stop of 8 ft. pitch it is introduced in the Organs in the Cathedrals of
Ulm and Riga. According to Hopkins, a stop labeled CORNO BASS-
ETTO, SOPRANO, 32 FT., was to be found in the Organ in the Cathe-
dral of Como, but regarding the form of this stop we have been
unable to obtain information.
FORMATION. The form of the CORNO DI BASSETTO or BASSETHORN seems to
have varied greatly in the hands of different organ-builders in Germany and else-
where on the Continent. Topfer describes it as made in the form of a labial stop :
while Locher describes it as a smooth- voiced stop made with free-reeds, and, as a
rule, without resonators, similar to the PHYSHARMONIKA. In all probability, as
their voices are described to be of a "smooth reed-character, " the stops in the Ulm
and Riga Organs are free-reeds with some form of resonators. But these matters
are now of very little practical value.
The CORNO DI BASSETTO, 8 FT., in its best form, as made to-day, is a striking
reed, and, being of the CLARINET family, is commonly furnished with plain cylin-
drical resonators completely open at top, similar to that shown in Fig. 6, Plate I.,
but properly of a larger scale, so as to develop a fuller tone than that of the CLARI-
NET. With a similar aim, we strongly recommend the addition of a resonant
chamber to the top of the cylindrical tube of the form shown in the drawing of the
complete pipe, Fig. 8, Plate II. This should have a screw adjustment for regula-
tion, as indicated, and may be further provided with a shade for fine toning.
Although organ-builders will not look with much favor on this form of resonator,
on account of its troublesome construction, yet every means should be adopted to
perfect a stop of such importance. It can be made one of the most beautiful
lingual stops in the Organ. While the reeds and their tongues generally resemble
those of the CLARINET in form and curvature, they should be slightly larger in all
directions, to accord with the larger scale of the resonators. Larger boots are also
to be recommended, for they have more influence on tone than is commonly under-
stood ; unfortunately large boots cost more than small ones.
Although no authority is furnished by an orchestral instrument, there seems
to be no reason why a very valuable voice should not be given to the Concert-room
Organ in the form of a CONTRA-CORNO DI BASSETTO or DOUBLE BASSET-HORN,
1 6 FT. We are strongly in favor of such a stop being introduced in every Concert-
room Organ just as it has been in the Organ in the Centennial Hall, Sydney, N. S.
W. A beautiful family might be created by the addition of an OCTAVE BASSET-
HORN, 4 FT.
TONALITY AND REGISTRATION. The orchestral Corno di Bass-
etto is, strictly considered, a Tenor Clarinet with a compass from
FF to c 3 , but its tone is fuller and of richer reedy quality; and this
has to be produced very clearly in the organ stop, so as to differ-
entiate it distinctly from the CLARINET. Voiced by a master-hand
and with this aim, the CORNO DI BASSETTO becomes invaluable as a
timbre-creator in artistic registration, combining perfectly and
- producing a beautiful series of tones with all classes of labial stops,
including such a compound stop as the DULCIANA CORNET (q. .).
The stop should be carefully developed by artistic voicers, and it
8o ORGAN-STOPS
should find a place in every important Organ. In the properly ap-
pointed Concert-room Organ its place is in the division appropriated
to the stops representing the wood-wind forces of the grand orches-
tra. There its voice would be valuable both in tonal massing and as
a solo. One has only to turn to the compositions of Mozart and
Mendelssohn to realize the high estimation in which these masters
held the tones of the Corno di Bassetto.
CORNO DOLCE. The name given to a stop of 8 ft. pitch, con-
structed of inverted conical pipes, after those of the DOLCE or DOL-
CAN. The name is misleading, for the tone of the stop has no horn
quality, inclining, on the other hand, to a flute-tone. It is not a
common stop, having nothing characteristic to recommend it. A
few examples exist in English Organs, as in the Organ in the Free
Trade Hall, Manchester.
CORNO FLUTE. This name has been given to two widely
different stops, to neither of which it can be said to be highly appro-
priate; the two constituents forming it conflicting in a tonal sense.
The original CORNO FLUTE is a lingual stop of 8 ft. pitch, formed
with resonators of wood, and voiced to yield a quiet tone inclining
to an Oboe quality. It was invented by the distinguished London
organ-builder, William Hill. An example of this stop exists in the
Organ in the Church of St. Olave, Southwark, Surrey.
The second form, of later date, is a metal labial stop of 8 ft.
pitch, invented by Herbert Norman, organ-builder, of Norwich.
The name, in this instance, seems to have been given to the stop
on account of the Horn-like tonality in its tenor portion and the soft
fluty character of its higher octaves. We are not surprised at this
tonal peculiarity, for we have long been of the opinion that the
characteristic tones of the orchestral Horn will be most successfully
imitated by labial pipes.*
CORNO INGLESE, Ital A lingual stop, of 8 ft. pitch, yield-
ing a reedy tone in imitation of that of the orchestral instrument
bearing the same name. For particulars respecting the stop, see
COR ANGLAIS.
CORNOPEAN. A lingual stop of 8 ft. pitch, the tone of which
is intended to imitate that of the brass, wind-instrument originally
called Cornopean, but now known as the Cornet & Pistons (Fr.)-
* The stop in our Chamber Organ, of the KKRAULOPHONE class, was named by us CORNO DI
CACCIA because of the remarkable imitation of the Horn tones it yielded in its tenor and middle
octaves.
THEIR ARTISTIC REGISTRATION 81
Stone describes the instrument as "intermediate between the
French Horn, Trumpet, and Bugle"; and this description would
point to the quality of tone desirable in the CORNOPEAN of the
Organ, but which very rarely attends it.
The CORNOPEAN was invented by the distinguished organ-
builder William Hill, of London, and was introduced in several of his
organs. The firm placed one in the Swell of the Organ in the Cen-
tennial Hall, Sydney, N. S. W. When made and voiced by a master-
hand the stop is distinct and agreeable, occupying a place, tonally,
between the TRUMPET and the HORN. The resonators of the CORNO-
PEAN are, like those of the TRUMPET, inverted conical in form, and
vary considerably in scale, according to the quality and strength of
voice aimed at. The reeds should be of the closed variety ; and the
tongues should be thick and carefully curved to produce a tqne free
from the brassy clang of the TRUMPET, but of a firm and singing
quality. There is no question respecting the value of a fine lingual
stop, having such an intermediate voice, in refined and expressive
registration in which the assertive clang of the TRUMPET would be
destructive and that of the HORN too dead. In the properly ap-
pointed Concert-room Organ the CORNOPEAN should find its place
in the division appropriated to the brass-wind stops.
Locher describes the CORNOPEAN as " an 8 ft. labial stop of horn-
like tone." Allihn describes it much in the same manner in the
following words: " CORNOPEAN (ital), Paanshorn, ein veraltete
Labialstimme von hornartigem Ton. Zu deutsch etwa : Jubelhorn. ' '
COPULA, Lat. Ger. KOPPEL. This term is used in two senses
in German organ nomenclature. It alludes to the mechanical appli-
ance, commonly known as the coupler, which connects claviers or
other portions of an Organ together temporarily. It is also applied
to certain labial stops having a tonality which seems to have the
property of binding together in a harmonious manner extreme voices
of other stops. The KOPPELFLOTE is an example. See KOPPEL.
CREMONA. This very meaningless and inappropriate name
for an organ stop labial or lingual is obviously a corruption.
Hawkins, in his "History of Music." says: "The names and de-
scriptions of several musical instruments instruct us as to the nature
and design of many Stops in the Organ, and what they are intended
to imitate. For instance, in the Krummhorn, the tone of it origi-
nally resembled that of a small Cornet, though many organ-makers
have corrupted the word into Cremona, supposing it to be an imita-
tion of the Cremona Violin."
6
82 ORGAN-STOPS
Max Allihn gives a more sensible usage of the term in the follow-
ing words: "CREMONA bedeutet eine Labialstimme von streichen-
dem Tone. Die Cremonesergeige wird in Prankreich kurzweg Cre-
mona genannt." Under any conditions, the name of a town must be
inappropriate for an organ-stop, although not more so than the
name of a person written backwards, which has recently been added
to the list of stops. Such a method of naming is ridiculous and
should be discontinued. Surely the name of an organ-stop should
either indicate tone or some special formation of its pipes.
CROMORNE, Fr. Ger., KRUMMHORN. The name used by
French and German organ-builders to designate a lingual stop, the
tone of which somewhat resembles that of the CLARINET, but with a
smoother and mournful character. It is commonly of 8 ft. pitch.
Seidel, in his remarks on the KRUMMHORN, says: "Properly 'CoR-
MORNE,' from cor 'horn' and morne 'mournful, still, soft/ signifying
a soft, quiet Horn, is a lingual stop of a delicate intonation, of 8 ft.
or ,4 ft, pitch, of tin or pipe metal, open or shaded, and sometimes
formed of small-scaled cylindrical pipes " (like the CLARINET). This
stop has been constructed in imitation of an old instrument, called
the Krummhorn (crooked horn), which had six holes, and was, at its
lower end, bent in the form of a half circle.
The CROMORNE, 8 FT., has long been a favorite stop with the
French organ-builders. It appears in the Positif of the Organs in
the Royal Church of Saint-Denis; the Cathedral of Chartres; and
the Basilique du Sacrd-Coeur, Paris: in these Organs there is no
CLARINET. It is inserted in the Rcit of the Organ in the Church of
Saint-Sulpice, Paris, in which there is no CLARINET: and in the
R6cit, along with the CLARINET, in the Organ in the Conservatoire
Royal de Musique, Brussels. It occupies a place, along with the
CLARINETTE-BASSE, 16 FT., and CLARINETTE-AIGUE, 4 FT., in the
Positif of the Organ in the Cathedral of Notre-Dame, Paris* It is
inserted in the Positif of the Concert-room Organ in the Palais du
Trocad&x>, Paris; and in the Positif Expressif of the Organ in the
Albert Hall, Sheffield. We are not aware of any stop under the
name CROMORNE, or having the characteristic tonality of the French
stop, having been inserted in an Organ constructed in England or
America, It may be taken for granted that unless the CROMORNE
had been found valuable in registration by French organists, and
held in high estimation by such distinguished organ-builders as
Cavaill<5-Coll, Merklin, Abbey, and Mutin, it would not have been
so systematically inserted in almost all the more important Organs
THEIR ARTISTIC REGISTRATION 83
constructed in France. Such being the case it is worthy of considera-
tion as to the desirability of adding such a stop to the tonal appoint-
ment of the Organ of the Twentieth Century. To us it would seem
highly desirable.*
CYMBAL. Fr., CYMBALE. Ger., CYMBEL, ZYMBEL. ItaL,
CIMBALO. The name used to designate a compound harmonic-
corroborating stop, usually composed of from four to seven ranks
of open, organ-toned pipes, of high pitch, breaking at every octave.
Its ranks are alternately octave- and fifth-sounding. Its name was
suggested by the likeness of its compound tone the clang of the or-
chestral Cymbals. Such a tone would seem at variance with the
harmonic-corroborating office of the stop; and as noise even if it is
musical noise is not desirable in any form of musical instrument,
there would seem to be little in favor of introducing so space-taking
and costly a stop in the Organ. The CYMBALE seems, however, to
have been esteemed for some reason by the great French organ-
builder Cavaille-Coll. In the Organ in the Royal Church of Saint-
Denis there is a GROSSE CYMBALE, of four ranks, and a CYMBALE,
also of four ranks, in the Grand Orgue; and a CYMBALE, of four
ranks in the Positif. In the Organ in the Church of Saint-Sulpice
there is a CYMBALE of five ranks in the Rcit Expressif . In the Con-
cert-room Organ in the Albert Hall, Sheffield, there is a CYMBALE of
four ranks in the Grand Orgue (Premier Clavier). There is a re-
markable example, of seven ranks, in the Concert-room Organ in
the Music Hall, Cincinnati; the composition of which is here given:
CYMBAL VII. RANKS.
CC to BB . . . . 15 19 22 26 29 33 36.
C to B . . . . 12 15 19 22 26 29 33.
C 1 tO b 1 . . . . 8 12 15 19 22 26 29.
c 2 to b 2 .... i 5 812151922.
C* tO C4 . . . DOUBLE I 5 8 12 15 19.
* "LE CROMORNE (de 1'allemand Krumm-Horn, Cornu 'torse), est encore un quatre-pieds qui
en forme huit, en raison de sa forte languette. On conQoit que, n'ayant pas la hauteur de tuy-
aux que comporte son ton, il rende des sous ne~cessairement moins forts que la Tro-mpette, ou la
hauteur et le ton se trouvent d'accord. C'est n<anmoins le meilleur des jeux d'anches acces-
soires. II a le timbre clair, plus nourri que le Hautbois, tenant du Cor anglais et de la Clarinette
avec une teinte plus me"tallique et une certaine mollesse gutturals qui n'est pas sans grace et qu'on
nomme cruchement. La difficult^ est d'obtenir cet effet & son point: avec trop de mollesse, le
Cromorne rale; avec trop de raideur, au lieu de crucher, il crache. ... On peut le placer k tous
les claviers en lui donnant des tallies diverses, mais il ne faut pas les jouer ensemble : leur timbre
est trop saillant et nerveux pour bien s'accorder 6n se doublant lui-me'me. II ne pourrait pas
meTne se poser en p6dale pour appuyer son chant au grand orgue; quoique, en tout autre cas,
iso!6 en pe"dale, il puisse faire bon effet, ici 6videmment le Cromorne en pdale de"truira le peu
d'inte're't qu'offre le Cromorne manuel. M616 & quelques fonds, il perd son aprete et se pr&t
84 ORGAN-STOPS
The CYMBAL does not appear to have assumed any importance
In the practice of the organ-builders of Germany : we have, after a
somewhat hurried review, been able to find one instance of its intro-
duction. The stop appears on the Erste Manual of the Organ built
by Priedrich Ladegast for the Cathedral of Schwerin in 1871. It is
described thus: "CYMBEL 3 fach, 14 lothig; einen halben Ton enger
als, Mixtur grosster Chor 2! \ c 1 g 1 c 2 , repetirt nur einmal."
CYMBELREGAL, Ger. An obsolete lingual stop, of either 4
ft. or 2 ft. pitch; the tone of which was of a metallic and ringing
character. See REGAL.
CYMBELSTERN, Ger. Literally Cymbal-star. This so-called
organ-stop, which was merely a mechanical device actuated by the
organ-wind, has been properly classed among the several puerilities
of old German organ-building. It was in the form of a star, to the
points of which small bells or metallic "jingles" were attached.
When caused to revolve, at the will of the organist, it gave forth a
tinkling sound. A few examples are said to exist ; as in the Organs
in the Cathedral of Merseburg; the Abbey Church, Weingarten; and
the Church of Waltershausen, near Gotha.
D
DECIMA, Ital. The term used by Italian organ-builders to
designate the TIERCE or TENTH, 33^ FT. The following terms are
used by them for other harmonic-corroborating stops: DECIMA
QUINTA the FIFTEENTH, 2 FT.; DECIMA SETTIMA the SEVEN-
TEENTH, i% FT.; and DECIMA NONA the NINETEENTH, ij^ FT.
DIAPASON. Ger., PRINZIPAL. Ital., PRINCIPALS. Fr., MON-
TRE. The word employed by the early English organ-builders, and
commonly qualified by the prefix OPEN, to designate the stop yield-
ing the foundation tone of the Organ.; and which has been retained,
with the same signification, by English-speaking organ-builders to
the present day. In the term Diapason Normal it is used in the
sense of a standard of pitch. The word is derived from the Greek
SccxTuacr&v, the concord of the first and last tones. In Latin it
signifies an octave.
souvcnt mieux quo les grands jeux d'anches & la diversity des expressions, M61ang6 avec une
Wl&tf ou un Bourdon de huit, ses notes sup6rieures imitent jusqu'& un certain point la Clarinette;
avec le Prestant, ses notes inf6rieures joucnt Ic Basson, Le Cromorne, chantant & plusieurs par-
ties, lea laissc ressortir tcmtes; avantage qui manque sou vent & la Trompette, dont les basses
^crascnt les tlcssua; certains organistcs Vaccouplent & un Nasard, pour lui donner plus d'6nergie
mais settlement dans le cas o& il parle avcc lea fonds." Regnier.
THEIR ARTISTIC REGISTRATION 85
As a name for an organ-stop the word does not seem highly ap-
propriate, and we much prefer the more expressive one, PRINCIPAL;
but it is not likely that the time-honored word will ever be altogether
abandoned. Under certain conditions, and for the sake of distinc-
tion, tonally, it may be desirable to use both words in the same stop
appointment.
PITCH. The stop appears in the Pedal Organ in two pitches
32 ft. and 1 6 ft. When of the former pitch it is properly termed
DOUBLE DIAPASON, 32 FT. ; and when of the latter it may be simply
termed DIAPASON, its -Unison pitch being understood. In the Organs
commanded by the manual claviers it also appears in two standard
pitches 16 ft. and 8 ft. When of the graver pitch it is correctly
termed DOUBLE DIAPASON, 16 FT.; and when of 8 ft. pitch it is
simply termed DIAPASON, its unison pitch being understood. In
preparing an Organ Specification, the pitches of all the DIAPASONS
must be given.
FORMATION. The pipes of the DIAPASON are cylindrical in form and of large
scale, as shown in proper proportions in Fig. 9, Plate II. ; and when properly made
are of good pipe-metal, of ample thickness to withstand firmly the vibrations of the
air within them while speaking. The desire to save money has led to the use of
undesirably thin metal, as well as metal of inferior and insufficient quality. Such
a practice should be rigorously condemned and guarded against by every one
specifying or purchasing an Organ. It is impossible to obtain the necessary full
and firm foundation tone, characteristic of the perfect DIAPASON, from pipes con-
structed of poor or thin metal. The thickness of the upper lip of the mouth,
securing perfect rigidity and favoring correct and artistic voicing, is a factor of the
greatest importance. Different metals have been and still are being used in the
formation of the pipes of this important stop. The old builders used tin almost
pure, or very rich alloys of tin and lead; and, accordingly, we find, as in the cele-
brated Haarlem Organ, pipes one hundred and eighty years old in perfect condi-
tion to-day. How many DIAPASONS, constructed of poor, thin pipe-metal or zinc,
as commonly placed in modern Organs, will be found in good condition after the
lapse of even half that length of time ? Whatever the metal or alloy may be, it is
essential that it be of ample thickness. This is particularly the case when cast
alloys of tin and lead are used, such as the different grades of spotted-metal. Even
when hard rolled metals are used, such as zinc or the Hoyt two-ply metal, the
matter of adequate thickness must be carefully attended to. Organ-builders,
depending upon the firmness of rolled zinc, combined with the desire to save
money, have very commonly used it of too thin gauges for DIAPASON pipes, in-
serting lips of spotted-metal or some such alloy. We strongly advise the disuse of
zinc for DIAPASON stops of 8 ft. pitch, but in the 16 ft. octave of the DOUBLE DIA-
PASON it may be used if of ample thickness. Zinc may, with advantage, be used
for the feet of heavy pipes, provided they are properly lipped and toed with good
pipe-metal. The following Table may be accepted as a reliable guide, showing
the minimum desirable thicknesses of good spotted-metal for DIAPASON pipes.
The thicknesses are given in thousandths of an inch, and these are carried through
86
ORGAN-STOPS
seven octaves, and are calculated on a scale for the CC pipe, 8 feet, not exceeding
seven inches in diameter, voiced on a wind-pressure not exceeding six inches.
TABLE SHOWING MINIMUM THICKNESSES OF GOOD SPOTTED-
METAL FOR DIAPASON PIPES.
16'
8'
4'
2'
I'
6"
3"
c
O.IIO
0.080
0.065
0-055
O.O4O
0.030
0.025
c#
O.I 10
0.080
0.065
O.05O
O.O4O
0.030
0.025
D
0.105
0.080
0.065
0.050
0.040
0.030
0.025
D#
0.105
0.075
0.060
0.050
O.O4O
0.030
0.025
E
O.IOO
0.075
0.060
0.050
0.040
0.030
O.O20
F
O.I 00
0.075
0.060
0.050
0.040
0.030
0.020
P#
0.095
0.070
0.060
0.045
0.035
0.030
O.020
G
0.095
0.070
0.060
0.045
0.035
0.030
O.020
G#
0.090
0.070
0.055
0.045
0-035
0.025
O.020
A
0.090
0.070
0-055
0.045
0^035
0.025
O.O20
A#
0.085
0.065
0-055
0.045
0.035
0.025
0.020
B
0.085
0.065
0-055
0.045
0.035
0.025
0.020
If the Hoyt hard-rolled two-ply pipe-metal is used, the Table may be started,
for the CC, 8 ft., pipe, at gauge 0.070.
SCALE. The scaling of the DIAPASONS in Organs of the different
classes, and which are to be heard under different conditions, re-
quires the careful consideration of the organ expert. When two or
more DIAPASONS are inserted in an Organ, either in one or separate
Divisions, they should be scaled differently, so as to aid the voiccr
in the production of the desirable diversity of tone. Large and in-
ordinate scales have been used by certain organ-builders, apparently
with the view of producing great volumes of sound, but the undue
preponderance of the bass destroyed their general tonal value. As
the true bass of the Organ obtains in the pedal department, the
value of the tenor and higher octaves of the manual stops, and of the
DIAPASONS in particular, should not be overshadowed by any ab-
normal development of the bass octave, Even under the most
scientific treatment the trebles of the finest DIAPASONS are unde-
sirably weak in proportion to their basses. The tonality of the
tenor and middle octaves have to be taken most care of by judicious
scaling and artistic voicing. Experience has gone far to prove that
nothing is gained by using a larger scale than 7 inches in diameter
for the CC, 8 ft., pipe. Indeed, Edmund Schulze, of Paulinzelle, the
artist representing the best school of German organ-building of the
nineteenth century, and who did much splendid work in England^
THEIR ARTISTIC REGISTRATION 87
maintained that for the chief DIAPASON of the largest Church Organ,
63^ inches in diameter was the extreme scale for its CC pipe, pro-
vided it was properly voiced. The CC pipe of his superb DIAPASON
in the Organ in the Parish Church of Leeds, Yorkshire, is only 6^
inches in diameter; yet the stop is the grandest of its class known to
us. The same scale appears in his celebrated DIAPASON in the Organ
in the Church of St. Mary, Tyne Dock. Equal in importance to the
measurement of the scale is its ratio. That favored by the great
German builders Sauer, Ladegast, Walcker, Reubke, and Schulze,
and adopted by the last named for his DIAPASONS in the Leeds and
Tyne Dock Organs, is the ratio 1:^8, which halves on the sixteenth
step or seventeenth pipe. We, however, prefer the ratio i : 2.66,
which halves on the eighteenth pipe, favoring a slight increase of
tone in the treble, always desirable. The following Tables give de-
sirable scales for DIAPASONS suitable for Concert-room, Church,
and Chamber Organs, voiced on winds of proper pressures:
TABLE OF DIAPASON PIPE DIAMETERS IN INCHES, RATIO i : v
PIPES
CC
C
c 1
C 2
C3
C4
L
6.60
3-93
2-33
-39
0.82
0.49
II.
6.32
3.76
2.23
-33
0.79
0.47
III.
6.06
3.60
2. 14
.27
0.76
o-45
IV.
5-56
3-30
1.96
.16
0.69
0.41
V.
5-10
3-03
I. 80
,07
0.64
0.38
TABLE OP DIAPASON PIPE DIAMETERS IN INCHES, RATIO 1:2.66.
PIPES
CC
C
c 1
c 2
C3
C4
L
6.68
4. 10
2-51
1-54
0.94
0.58
II.
6.42
3-94
2.41
1.47
0.91
0-55
III.
6.06
3-78
2.31
1.42
0,87
0-53
IV.
5-56
3-34
2.05
1-25
0.77
0.47
V.
5-io
3.08
1.89
I-I5
0.71
0-43
TONALITY. The true tone of the DIAPASON is that which is
peculiar to and characteristic of the Organ, and which cannot be
produced by any other musical instrument. When at its best, it is
singularly pure and simple, being, like the normal tone of the tuning-
fork, almost entirely free from harmonic over-tones or upper partials.
It is this fact that makes the true DIAPASON ineffective as a melodic
stop, played in single notes; while played in full chords its tones are
rich and impressive and generally beautiful. If prolonged, however,
it becomes cloying and palls upon the ear. It is this simple quality
which makes the DIAPASON tone the proper foundation of the tonal
structure of the Organ; upon which may be laid tonal combinations
of endless variety and beauty.
88 ORGAN-STOPS
The following remarks from the pen of Professor R. H. M.
Bosanquet, an authority on tonal matters, are of interest:
"The scales, character, and voicing of the OPEN DIAPASON vary with fashion,
and are different in different countries. We may distinguish three principal types.
The old English DIAPASONS of the days before the introduction of Pedal Organs
into England were characterized by a rich sweet tone, and were not very powerful.
They were generally voiced on a light wind, having a pressure equivalent to that
of a column of water of from 2 to 2 J^ inches. The scale was in some cases very
large, as in Green's two OPEN DIAPASONS in the old Organ at St. George's, Windsor;
in these the wind was light and the tone very soft. In other cases the scale was
smaller and the voicing bolder, as in Father Smith's original DIAPASONS in St.
Paul's Cathedral. But on the whole the old English DIAPASONS presented a lovely
quality of tone. English travelers of those days, accustomed to these DIAPASONS,
usually found foreign Organs harsh, noisy, and uninteresting. And there arc
many still in England who, while recognizing the necessity of a firmer diapason-
tone in view of the introduction of the heavy pedal bass, and the corresponding
strengthening of the upper departments of the organ-tone, lament the disappear-
ance of the old diapason-tone. However, it is possible with care to obtain DIA-
PASONS presenting the sweet characteristics of the old English tone, combined with
sufficient fullness and power to form a sound general foundation. And there can
be no doubt that this should be one of the chief points to be kept in view in organ
design.
"The German DIAPASON was of an entirely different character from the Eng-
lish. The heavy bass of the pedals has been an essential characteristic of the
German Organ for at least two or three centuries, or, as it is said, for four. The
development of the piercing stops of high pitch was equally general. Thus founda-
tion-work of comparatively great power was required to maintain the balance of
tone; the ordinary German DIAPASON was very loud, and we may say coarse, in
its tone when compared with the old English DIAPASON. The German stop was
voiced as a rule on from 3}^ to 4 inches of wind, not quite twice the pressure used
in England.
" The French DIAPASON is a modem variety. It may be described as present-
ing rather the characteristics of a loud GAMBA than of a DIAPASON. In other
words, the tone tends towards a certain quality which may be described as ' tinny '
or metallic; or as approaching to that of a string instrument of rather coarse
character. Some modern English builders appear to aim at the same model, and
not without success.
"The tone of a DIAPASON must be strong enough to assert itself* It is the
foundation of the whole organ tone, It is the voicer's business to satisfy this con-
dition in conjunction with the requirement that the tone shall be full and of agree-
able quality."*
That we agree with everything stated by Professor Bosanquet is
proved by every word we have written on the subject during the
past forty years. Unquestionably, to the lover of sweet sounds
which we claim to be the true English diapason-tone is the most
lovely ; that of the French MONTRE the most unsatisfactory : yet to
* Encyclopedia Britannica: Ninth Edition. "Organ."
THEIR ARTISTIC REGISTRATION 89
French ears, and those satisfied with power rather than refinement
of tone, it seems to be agreeable. Different tastes are born and
fostered under different conditions.
VOICING AND REGULATING. As there are various tones produced
by stops labeled DIAPASON, so there are different schools of voicing,
not only peculiar to different countries but to varied classes of
voicers. Such being the case, it is obvious that no standard of dia-
pason- or pure foundation organ-tone has been in any way estab-
lished, or even suggested, in modern organ-work. Yet it must be
admitted that for such an all-important foundation organ-tone a
well-defined standard should be set and universally adopted.
At the present time, when wind-pressures of great variety are
used at the caprice of organ-builders, it seems a hopeless thing to
look for anything approaching a standard diapason-tone. As Pro-
fessor Bosanquet has pointed out, it has ranged in the past from the
refined and beautiful tones of the true English DIAPASONS, on
through the loud and coarse tones of the German PRINCIPALS, to
the stringy and unsympathetic voices of the French MONTRES.
While it is obviously desirable that when two or more DIAPASONS
are inserted in an Organ, they should be of different scales, and
voiced to produce varied strengths and, within due limits, different
tints of true organ-tone; there is certainly no call for them to yield
the coarse German quality, on the one hand, or the dry, "tinny,"
or gambaish tonality favored by the French voicers, on the other
hand.
Scale, of course, is a matter of great importance, but it must be
conceded that other conditions are of equal, if not of greater, im-
portance in the production of true diapason-tone. These conditions
may be embraced under the general term Voicing, which includes
wind-pressure, wind-supply, and the artistic treatment of the mouth
and top of every pipe. First in order is the proper wind-pressure.
Experience has shown that for the production of the pure, smooth,
and beautiful tone characteristic of the true English DIAPASON, a
moderate wind-pressure is essential, accompanied by a copious
supply. The finest DIAPASONS in existence to-day speak on wind
of from 3 to 5 inches. For instance, the grandest and most beautiful
DIAPASON known to us is the larger of the two, made by Schulze,
in the Great of the Organ in the Parish Church of Leeds, already
alluded to, which speaks on a copious wind-supply of only 3% inches
pressure. It is just worth while stating, as a lesson to voicers hit by
the high-pressure craze of the present time, the several pressures
used in this fine instrument, which owes so much of its grandeur and
90 ORGAN-STOPS
beauty to the genius of Schulze. All the stops of the Pedal Organ
with its 32 ft. and 16 ft. reeds, are on wind of 3% inches. The stops
of the Great Organ, with the exception of the POSAUNE, are on wind
of 3^ inches; the POSAUNE being on wind of 7 inches. The Swell
Organ is on wind of 3 inches; the Choir Organ is on wind of 2^
inches; the Solo Organ, with the exception of the TUBA, is on wind
of 5 inches, the TUBA being on wind of 8 inches; and the Echo Organ
is on wind of ij^ inches. Although tonally appointed and appor-
tioned in the "good old-fashioned way," this instrument is par ex-
cellence a Church Organ; dignified and refined in tone, and admirable
in its fundamental office the accompaniment of the musical service.
The Organ comprises 77 speaking stops, 1 1 of which are in the Pedal;
having in all 5,060 pipes.
Careful experiment and observation go to prove that in the pro-
duction of a pure, pervading, and musically perfect normal diapason-
tone (or organ-tone) a volume of, and not high-pressure, pipe- wind
is the principal factor. The loud and coarse tones which character-
ize the great majority of the modern DIAPASONS voiced on winds of
undue pressures, have nothing like the pervading, traveling, and
sympathetic qualities and powers of the stops artistically voiced on
a copious supply of low-pressure wind.
It is somewhat surprising that the French voicers have so utterly
failed in realizing the value and beauty of the English diapason-tone.
They seem to have been satisfied with the cutting, "tinny, " tone of
their MONTRES. Even the great Cavaill6-Coll apparently did not
aim at anything better : and much of the stringy tonality of his stops
is due to the practice, he invariably followed, of slotting his pipes at
the top, for convenience in tuning. The true English DIAPASON
was never slotted; and no DIAPASON ever should be. Let modern
voicers take note of these important facts.
A certain method of voicing DIAPASONS has been introduced
during recent years, and is now practised by a certain class of voicers
a method which we would never allow to be used in any Organ
over the construction of which we had any control. We allude to
the leathering of the upper lips of the pipes, with the view of im-
parting to them the necessary thickness and the desirable roundness
and smoothness. It is claimed, by the organ-builders who resort to
it, to greatly improve the tones of their pipes. Probably it does in
comparison with the tones produced by the unduly thin pipe-metal
they may have been accustomed to use. None of the great and
truly artistic organ-builders have resorted to this cheap and un-
desirable expedient. Let the mouths of the pipes be properly
THEIR ARTISTIC REGISTRATION 91
formed with lips of sufficient thickness, as in old work and in that of
the great builders, and let their upper lips be properly adjusted and
smoothly rounded and there will be no necessity for leathering. We
presume the question of durability does not trouble the organ-
builder who calculates that the leathering will, in all probability,
outlast the rest of his lifetime; but the question is of great impor-
tance to the Organ Committee or purchaser of the Organ. In what
condition would the DIAPASONS of the celebrated Haarlem Organ
be in to-day had its builder used thin metal, and leathered the
mouths of their pipes, in the money-saving method now in vogue?
We unhesitatingly affirm that the fine work of the past proves that
there is not the slightest excuse for leathering the lips of organ-pipes :
and the objectionable practice should be discountenanced by every
purchaser of an Organ. If a very thick and well-rounded lip is
desired, let a strip of pipe-metal be folded just as the leather is
folded and slipped on the upper lip and soldered in position. This
method would be most efficient, but unless imperatively specified
there would be no chance of its being adopted by the organ-builder
or voicer. Let lips of sufficient thickness be furnished, as in old
work, and let a proper school of voicing, on winds of moderate pres-
sures, be instituted, and there will be no need to resort to objec-
tionable and perishable leathering for the production of perfect
diapason-tone.
Next in importance to the production of a pure and beautiful
diapason-tone is its perfect regulation. This is a matter requiring a
sensitive ear and the expenditure of considerable time. Accordingly,
it is very seldom that one finds a DIAPASON accurately and artist-
ically regulated throughout its compass. Perfect regulation is essen-
tial to the beauty of an Organ; for a few badly regulated stops will
go far to destroy that beauty : this will especially be the case if the
DIAPASONS are not properly regulated, with due regard to an effec-
tive balance of tone in the treble octaves. There is always a ten-
dency to undesirable preponderance of tone in the bass octave; and
this should be overcome to as large an extent as possible both in the
voicing and regulating. A perfect DIAPASON is a work of science and
art that any pipe-maker and voicer may point to with justifiable
pride.
COMBINATION AND REGISTRATION. The diapason-tone is to the
tonal structure of the Organ what the solid foundation of a building
is to its superstructure. As has already been stated, the pure organ-
tone yielded by the true and properly-voiced DIAPASON is simple in
Jts nature: and such being the case, it naturally lends itself to en-
92 OAGAN-STOPS
richment by the addition of those tones which, by their presence,
impart so great a charm and beauty to the compound sounds of the
stringed instruments of the orchestra and the cultivated human
voice. Accordingly, to the simple prime tone of the DIAPASON must
be added a superstructure of upper partials or harmonic over-tones,
scientifically graduated in strength in accordance with the natural
laws of musical sounds. For this purpose are introduced in the
properly-appointed Organ the several derivatives of the DIAPASON
the Octave, Mutation, and compound harmonic-corroborating
stops, A deficiency in any one of these classes seriously impairs
the desirable range of compound diapason-tones; and, accordingly,
divests the Organ proper of much of its usefulness and beauty.
This important fact is too largely neglected in the tonal appointment
of modern Organs. For the production of the various colors of
foundation-tone, not only have greater or lesser proportions of these
different harmonic-corroborating stops to be added to the DIAPASON ;
but greater or lesser strengths in their voices are required. This
latter requirement points conclusively to the necessity for their tonal
control, through their inclosure in a swell-box. The numerous modes
of combining the DIAPASON and its harmonic derivatives alone, may
be classed under the term Diapason Registration.
In artistic registration the DIAPASON has a very important rdle;
for upon its foundation tone can be built an infinite number of effec-
tive and beautiful combinations the most dignified and impressive
the properly-appointed Organ places at the command of the musi-
cian. Colorings derived through combinations with flute-, string-,
reed-, and brass-toned stops are simply endless in their variety.
Indeed, registration on the foundation of the diapason-tone is a
study in itself, worthy of the organist's earnest attention; for upon it
is based the entire fabric of artistic organ tonal combination. Such
registration and the tonal effects it produces belong exclusively to
the Organ.
TUNING. The tuning of DIAPASON pipes deserves the attention
of those interested in artistic and good organ-building, The practice
of tuning such pipes by means of slotting them near their open ends,
so commonly adopted by the French organ-builders, is to be con-
demned on the ground that such slotting seriously injures the true
diapason-tone. Coning, which either expands or contracts the open
ends of the pipes, flattening or sharpening their tones, has, from
old times, been commonly adopted in tuning pipes of the DIAPASON
class from tenor C (4 ft.) upwards; but the practice has nothing to
recommend it save convenience. It is not desirable on tonal grounds,
THEIR ARTISTIC REGISTRATION
93
because in any case of excessive spreading or contracting of the top
of a pipe the tone undergoes a certain alteration : and, further, the
necessary smart and often heavy blows of the weighty metal cones,
frequently repeated, tend to, and ultimately do, seriously injure the
top, mouth, or feet of the pipe. The most, and, indeed, only, de-
sirable system of tuning DIAPASON pipes is by means of adjustable
slides, which only require to be slightly raised or lowered in tuning;
but it is not to be expected that organ-builders will adopt this sys-
tem unless compelled to do so. Pipes that may be considered too
large to be conveniently tuned by slides should be cut a little longer
than their correct speaking lengths, and tuned by means of broad
tongues, cut in the manner shown in the accompanying illustration,
Fig. 10. If the pipes are properly cut to length, only fine tuning will
FIG. 10
be necessary by the easy and slight manipulation of the tongues.
No injury need ever be done to the pipes. This method can be
carried from the CC to the tenor C pipe, or even farther with ad-
vantage. The method will, of course, be carried through the 16 ft.
and 8 ft. octaves of the DOUBLE DIAPASON.
DIAPASON, WOOD. Although German organ-builders have
proved that true diapason-tone can be produced from quadrangular
wood pipes, no French, English, or American organ-builder seems to
have devoted serious attention to the construction of DIAPASONS of
wood throughout. This is to be regretted, for we are convinced,
from observation of what has been done by German builders, that,
associated with metal DIAPASONS, a wood DIAPASON would be of
great value, doing away with all risk of sympathy, while building up
a grand volume of tone. Our attention was first directed to the
matter when we inspected the fine Organ in the Church of St. Bar-
tholomew, Armley, Yorkshire, constructed by Edmund Schulze.
94
ORGAN-STOPS
In this instrument the bass octave of the MAJOR PRINCIPAL, 8 FT.,
in the Great Organ a grand stop of pure organ-tone is carried
down in wood pipes in a manner so perfect that the ordinary ear
fails to detect the transition from metal to wood. In the accom-
panying illustration, Fig. n, is given the Front View and Longitu-
dinal Section of the lower portion of the BB pipe of this stop, drawn
to scale. It will be observed that the external slopes of the lower
FIG. ii
and upper Hps, shown in the Section, closely resemble those of the
usual metal DIAPASON. The principal feature here is the recessed
and splayed cap A, which is brought almost to a sharp edge where it
forms the lower lip, in this respect resembling the metal lip. The
upper lip is thin and cut thin, as shown. As regards scale, the most
important dimensions are those of the BB pipe which immediately
adjoins the tenor C metal pipe. The tenor C pipe is 3% inches in
diameter, with a mouth 3 inches wide and % inch high, and with a
wind-hole in the foot -f$ inch in diameter. The BB wood pipe
measures, internally, 2}$ inches in width by 3^ inches in depth,
with a mouth 7 /s inch high. In the same Organ, the SUB-PRINCIPAL,
16 FT., has its two lower octaves of wood; and the MINOR PRINCIPAL,
8 FT., in the Choir Organ, has its bass octave of wood. All are speci-
mens of artistic voicing.
THEIR ARTISTIC REGISTRATION 95
Both the DIAPASON, 16 FT., and DOUBLE DIAPASON, 32 FT., of
the Pedal Organ are commonly made of wood; but it is seldom one
hears the true diapason-tone produced by such stops as made at the
present time. The desire to obtain a powerful intonation, through
the use of large scales and wind of undue pressure, goes far to de-
stroy the tone which should always be sought for in such foundation
stops, and especially in those of the refined Church Organ. The now
common practice of applying the harmonic-bridge, with the view of
securing promptness and a certain clearness of speech, imparts a
stringy quality to the tone, in itself by no means disagreeable, and
which may, perhaps, be considered desirable in this age of borrowed
and poverty-stricken Pedal Organs. If properly voiced on a copious
flow of wind of a pressure not exceeding 6 inches, scales between 8
inches by 10 inches and 12 inches by 14^ inches for the CCC (16 ft.)
pipe will be ample. The German and other Continental builders,
in their appreciation of the true nature and office of the Pedal Organ,
and also of its adequate stop-apportionment, never, so far as we
have been able to learn, used inordinate scales. The largest scale
known to us in a German Organ is 12 inches by 14 inches; while the
majority of Pedal Organ PRINCIPALS, 16 FT., rarely exceed 9 inches
by n inches. The late W. T. Best, of Liverpool, the best judge of
organ-tone in his day, invariably specified the scale of 10 inches by
12 inches. We have conclusively proved the scale of 8 inches by
10 inches to be perfectly suitable for a Chamber Organ, voiced even
on the low pressure of 2^ inches. On a higher pressure say 4
inches it would be sufficient for a Church Organ of ordinary size.
' It must be borne in mind that this stop has to provide a properly
balanced bass to the Great Organ DIAPASON, 8 FT.
The most satisfactory scales for the DOUBLE DIAPASON, 32 FT.,
are not proportionately so large as those desirable for the DIAPASON,
1 6 FT. It is undesirable, on the ground of scientific and artistic tonal
balance, that the sub-octave pitch should be as assertive as that of
the fundamental unison.
The most suitable minimum scale is n inches by 13 J^ inches for
the CCCC pipe, halving on the eighteenth pipe: while the maximum
scale may be that adopted by Schulze for the fine stop in the Organ
in St. Bartholomew's Church, Armley ; namely, 14^ inches by iSJ^
inches. This scale is larger than that adopted by the same builder
for his DOUBLE PRINCIPALS, in the Organs in Bremen Cathedral and
St. Mary's Church, Wismar, which measure only 12 inches by 15
inches. The equally celebrated builder, Ladegast, uses a still smaller
scale nYs inches by 14^3 inches for the CCCC pipe of the
96 ORGAN-STOPS
DOUBLE PRINCIPAL, 32 FT., in his celebrated Organ in Schwerin
Cathedral.
DIAPASON PHONON. The name introduced during late
years to designate a metal labial stop of 8 ft. and 16 ft. pitch, and
large scale, voiced to yield a powerful and pure organ-tone. The
pipes of the stop are of the same form and construction as those of
the standard DIAPASON. The stop and its name were introduced by
Hope-Jones, who adopted the cheap method of forming thick and
smoothly rounded upper lips to the mouths of its pipes so desir-
able in all pipes" of the DIAPASON class by covering the thin metal
with perishable leather. For further remarks on this practice, see
DIAPASON. The DIAPASON PHONON will, naturally, be favored by
those who love loud sounds; and who advocate the construction of
brick-and-mortar or reinforced-concrete swell-boxes.
DIAPASON, STOPPED. This name is an example of loose
terminology : the stop so named does not belong to the class yielding
pure organ-tone, and is not a DIAPASON in any sense of the word.
The so-called STOPPED DIAPASON belongs to the Covered Flute-
work of the Organ. Long usage, however, is likely to prevail; and
in all probability, as there is no English equivalent for the German
term GEDECKT, it will be retained by English and American organ-
builders, in organ-stop nomenclature, for many years to come. For
form and other particulars, see STOPPED DIAPASON.
DIAPHONE. The name of a stop invented by Hope-Jones.
It is of peculiar construction. The tone of the DIAPHONE pipe is
created by pulses or vibrations generated in its resonator by the
rapid motions of a pallet actuated by compressed air (pipe-wind)
that enters the boot, which contains the mechanical portion of the
pipe. The pallet simply closes and opens the lower orifice of the
resonator so many times in a second, according to the size and pitch
of the pipe, acting much in the same manner as the striking tongue
in an ordinary lingual pipe. As the vibrating column of air in the
resonator controls the action of the pallet, when adjusted, the pitch
of the pipe is not affected by a change in the wind-pressure, while
the strength and quality of the tone are altered. The tone pro-
duced by the best examples is full and commanding; but owing to
the complex character of the mechanical portion of the pipes, and
the difficulty of regulating the action of the same, it is almost im-
possible to secure uniformity throughout even the short compass of
the Pedal Organ. There are also very grave questions regarding the
THEIR ARTISTIC REGISTRATION 97
durability of the stop. These facts will effectually militate against
anything approaching a free introduction of the DIAPHONE. In all
probability it is destined to swell the ranks of the curiosities of
organ-building.*
DIVINARE. The somewhat inappropriate term used to desig-
nate a covered stop of 4 ft. pitch, belonging to the Covered Flute-
work, the tone of which is singularly soft and singing, f Such a stop
is very desirable in the Choir of a Church Organ, in a Chamber
Organ, and in the Echo or Ancillary Aerial Organ of a Concert-
room instrument. Its presence would be productive of many charm-
ing effects in refined registration. The stop is made of both wood
and metal : when of wood, the pipes are of small scale, and made
narrow and deep so as to secure small mouths. Low-pressure wind
is desirable.
DOIFLOTE, DUIFLOTE, Ger. Names used to designate the
covered, flute-toned stop, the pipes of which have double mouths,
commonly called DOPPELFLOTE (q. v.).
DOLCAN. The name given to an open labial stop of 8 ft. pitch,
the pipes of which are of metal in the best examples tin the
bodies of which are in the form of a slender, inverted truncated cone,
as shown in Fig. 12, Plate II., which depicts a CC pipe in correct
proportions. The bass octave has been formed of inverted pyra-
midal pipes of wood, but the metal ones are to be preferred. The
scale of the DOLCAN varies, as in other labial stops, according to the
volume and quality of the tone desired. The following scales in the
ratio of 1 12.519 halving on the nineteenth pipe may be accepted
as productive of satisfactory tones :
DOLCAN PIPE DIAMETERS IN INCHES RATIO 1:2.519.
PIPES CC C C 1 C a C3 C4
AT MOUTH 3.38 2.13 1.34 0.84 0,53 0.33
AT TOP 4.96 3.13 1.97 1.24 O./S 0.49
ATMOUTH 3.25 2.05 1.29 0.8l 0.51 0.32
AT TOP 5.16 3.25 2.05 1.29 O.Sl 0.51
* Particulars respecting the origin of the DIAPHONE, its forms and construction, accompanied
by fully detailed illustrations, are given in our work, "The Art of Organ-Building, " Vol. I,, pp.
399-400. Vol. II., pp. 619-622.
t " DIVINARE, flute de bourdon de quatre-pieds, fl&te divine, est une singuliere traduction de
ce mot, et c'est la seule indiquee pour exprimer la qualite" divinement superieure de cette flute
... a peu pres inconnue. Regnier.
"DIVINARE (vom Lat. divinus, gSttlich), sondern es gedecktes Flotenwerk zu 4'. Die In-
tonation dieser Stimme muss dem Namen nach sehr schon sein." Seidal.
9 ORGAN-STOPS
The width of the mouth may be two-ninths or one-fifth the
internal circumference of the pipe at its mouth-line, according to
the strength of tone required; and its height may vary from one-
fifth to one-fourth its width, subject to the wind-pressure employed
and the tone aimed at by the voicer. When a very delicate tone is
required, on a low pressure of wind, say 2j/ inches, a mouth of
one-fifth the circumference should be adopted, having a height about
one-quarter its width ; and the languid should be finely nicked. The
upper lip should be of good substance, cut straight, and smoothly
rounded.
TONE AND REGISTRATION. When properly formed and artist-
ically voiced, the DOLCAN yields a tone which is freer or more open
in quality than that of the true English DULCIANA, having a some-
what plaintive and singing character, which is highly effective in
soft accompanimental music, and extremely valuable in the more
delicate and refined school of registration. The tone combines in
the most satisfactory manner with all the varieties of flute-tones,
giving them a peculiar charm; and also with the voices of the softer
lingual stops. The DOLCAN 's place is in the softer accompanimental
division of the Organ, where it need not displace the DULCIANA.
DOLCE, Ital. This name has been used to designate stops of
different formation and tone; but, as the name implies, all being
similar in one important direction the possession of sweetness of
voice. The stop is introduced in important German Organs in both
8 ft. and 4 ft. pitch, as in the instrument in the Cathedral of Ulm.*
The DOLCE has been made by German organ-builders of both wood
and metal, and of both cylindrical and inverted conical pipes, after
the fashion of those of the DOLCAN (Pig. 12, Plate II.). A beautiful
example of this latter form, made by Edmund Schulze, exists in the
Echo of the Organ in the Parish Church of Leeds, England. It is of
small scale and its pipes are slightly conical, yielding a tone of a
quiet nasal quality; and, being on wind of only ij^ inches pressure,
has a tendency to be slow, imparting a peculiar intonation to its
speech. In certain German examples the tones are inclined to be
stringy; but it is questionable if this is desirable. In our opinion, a
smooth, extremely soft, and horn-like tonality is to be preferred,
differentiating it from that of the DOLCAN and DULCIANA. The
DOLCE, formed of inverted conical pipes, seems to have been intro-
*"DotCE 8' von Metall, ist eine Stimme, welche im aten Manual cler neuen Orgel in der
Peterskirchc zu Petersburg disponiert ist and einen ausscrst weichen Toncharakter hat, Im
3tcn Manual wtcht dasselbe Register von Zinn zu 4', Auch haufiger in Sauerschen Orgeln findcn."
S oid el.
THEIR ARTISTIC REGISTRATION 99
duced at an early date in English Organs; for it is stated on good
authority that John Snetzler inserted one in the Organ he built, in
1741, for the Parish Church of Chesterfield, Derbyshire.
It is to be regretted that the DOLCE, in its proper form, has not
been more frequently introduced in important Organs : but it would
seem that the space it requires on the wind-chest, and the additional
metal and labor its construction calls for beyond what a small-scaled
cylindrical stop, such as the kindred DULCIANA, demands, have
militated against its more general adoption by organ-builders.
There are, however, several good examples to be found in English
Organs constructed by leading English builders. A fine one exists
in the Organ in Emmanuel Church, Leicester. A DOLCE, 16 FT., of
very beautiful tone voiced by the artist, W. Thynne is to be
found in the Chapel of St. Katherine's Convent, London. A DOLCE,
8 FT., is inserted in the Swell of the Organ in the Centennial Hall,
Sydney, N. S. W. Roosevelt inserted a DOLCE, 8 FT., in the Swell of
the fine Organ he built for the First Congregational Church, Great
Harrington, Mass. The proper place for the DOLCE in the Organ
would seem to be quite undecided by organists and organ-builders,
as the following particulars show; yet, in artistic tonal apportion-
ment, it has its proper and logical position. In the Great of the
Organ in the Second Church, West Newton, Mass. In the Swell of
the Gallery Organ in Emmanuel Church, Boston, Mass. In the
Choir of the Concert-room Organ in the Carnegie Music Hall, Pitts-
burgh, Pa. In the Echo of the Organ in St. Peter's Episcopal
Church, St. Louis, Mo.
The scale of the DOLCE, 8 FT., varies in different examples,
though not to a large extent. The following scale of a representative
stop may be accepted as satisfactory, with low wind-pressures:
DOLCE PIPE DIAMETERS IN INCHES RATIO 1:2.66.
PIPES CC C c 1 c a c3 c*
AT MOUTH 3.08 1.89 I.I5 0.71 0.43 0.27
ATTOP 4.10 2.51 1.54 0.94 0.58 0.35
The width of the mouth should not exceed one-fifth of the in-
ternal circumference of the pipe at its mouth line; and its height
need not exceed one-fourth its width unless a leaning toward a flute-
tone is desired.
DOLCE CORNET. This stop, as the name implies, is com-
pound, harmonic-corroborating, and sweet-voiced. It is properly
formed of several ranks of very small-scaled open metal pipes, yield-
ioo ORGAN-STOPS
ing a soft, singing quality of tone. When extreme softness is re-
quired, the ranks should be octave- and fifth-sounding only: but
when a more assertive tone is desired, a third-sounding rank should
appear in every break, as in the following example of five ranks:
DOLCE CORNET V. RANKS.
CC tO f * . 12* 15 1/* 19 22.
f# to f a 8 12* 15 17*- 19.
f#* to 1)3 1 8 12* 15 17*.
e>i to c^ ... . i 8 ro 12 15.
It will be observed that the introduction of the SEVENTEENTH
makes the stop a SESQUIALTERA in all save the top octave. Artist-
ically voiced and scientifically graduated in tone this CORNET would
be extremely valuable in refined registration. The following is a
satisfactory composition for a CORNET of four ranks, octave- and
fifth-sounding :
DOLCE CORNET IV. RANKS.
CC to BB
C to B
C 1 tO g3
g# 3 to c<
A DOLCE GRAND CORNET composed of complete, through ranks
of true DOLCE pipes, of small scale, would be of great value in any
of the softer toned divisions or subdivisions, expressive, of the Con-
cert-room Organ, in which its presence would tend to complete the
harmonic structure. It may be composed of an OCTAVE, 4 FT.; a
TWELFTH, 2% FT.; a FIFTEENTH, 2 FT.; a SEVENTEENTH, i% FT,;
and a NINETEENTH, i J^ FT. The SEVENTEENTH may be omitted if
considered too harsh in tone.
DOLCE FLUTE. A softly-voiced stop of unimitativc flute-
tone, more commonly known by the Italian name PLAUTO DOLCE
(2- "-)
DOLCETTE. The term, as a diminutive, properly applied to
designate the OCTAVE DOLCE, 4 FT. This is a valuable stop in re-
fined registration, corroborating the first upper partial tone of all
soft-voiced unison open stops, and imparting a foreign element to
the voices of the softly-toned covered stops, such as the LIEBLICH-
GEDECKTS, FLi>TE A ClIEMINfeE, COR DE NUIT, etc,
DOLCIANO PROFUNDO. The distinctive name we have
suggested for a CONTRA-DULCIANA, 32 FT,, when such a stop ia
THEIR ARTISTIC REGISTRATION 101
commanded by a manual clavier. We have shown, in our tonal
scheme for the Concert-room Organ of the Twentieth Century, how
this grave and most desirable soft-toned, open metal stop, of 32 ft.
pitch, can be derived from the Pedal Organ CONTRA-DULCIANA,
adequately extended in compass, and be commanded, as an auxiliary
stop, by the First or Great Organ Clavier.*
DOLCISSIMO, Ital. The term appropriately used to designate
the softest flute-toned stop made. The extended term, FLAUTO
DOLCISSIMO, will, however, be found more expressive, especially if
the stop is voiced to yield an imitative tone that of the orchestral
Flute played pianissimo. In its best form, the DOLCISSIMO is of
8 ft. pitch, constructed of small-scaled hard wood pipes, having very
narrow inverted mouths. It should be voiced on wind of low pres-
sure, not exceeding 2]/ 2 inches, preferably of ij/g inches. It is an
ideal stop for a refined Chamber Organ; and for the Choir or Echo
of a Church Organ, and the Ancillary Aerial Organ or other very
softly-toned division of a Concert-room instrument. In registration,
the DOLCISSIMO, 8 FT., will be found extremely valuable in imparting
a beautiful effect and a. desirable firmness and body to the Vox
HUMANA and all the other softer-toned lingual stops, without de-
stroying their characteristic tonalities, which a loud-voiced FLUTE,
8 FT., would certainly do in combination. f
DOLZFLOTE, DULZFLOTE, Ger. The name given by old
German organ-builders to an open wood stop of 8 ft. and 4 ft. pitch,
the pipes of which are of medium scale, yielding a soft and sweet
unimitative flute-tone. Seidel describes the stop thus :
" DOLZFLOTE, DULZFLOTE, Flauto dulcis (susse Flote), auch
Tibia angusta, ist eine eng mensurierte offene Flotenstimme von
Holtz zu 4' und 8', von ausnehmend lieblichem und angenehmem
Tone. Im Pedal soil diese Stimme zu 16' unter dem Namen Flau-
tone vorkomen." See FLAUTO DOLCE.
DOPPELFLOTE, Ger. The term commonly used to designate
a covered wood stop of 8 ft. pitch, the pipes of which have two
mouths placed directly opposite each other, hence the name. As
usual with stops that require unusual conditions for their accommo-
dation on the wind-chest, and which call for special skill and more
than ordinary labor in their construction, the DOPPELFLOTE has
never been a favorite with English and American organ-builders;
* See "The Organ of the Twentieth Century." Chap. XI., pp. 297 and 319.
t See the Aerial Organ in "The Organ of the Twentieth Century," pp. 329-331.
102 ORGAN-STOPS
while its great tonal value seems to have been unrealized by the
organists of both countries. One, assuming to be an authority on
organ-stops, writing on the DOPPELFLOTE, displays a strange igno-
rance of its tonal character and value in registration. So far as we
have been able to learn, the first DOPPELFLOTE that appeared in an
English Organ was the one we inserted, in 1883, in our own Chamber
Organ, under the Italian name, FLAUTO PRIMO, 8 FT. It occupied a
prominent position in the Second Expressive Subdivision of the
First Organ.* This beautiful stop was presented to us by the late
Hilborne L. Roosevelt, of New York. A few examples have since
appeared in English Organs, for the most part made by Continental
builders, Roosevelt held the DOPPELFLOTE in high estimation and
invariably placed one in the Great of all his important Organs. It
is to be regretted that so few examples appear in even large Organs
constructed in this country during the present period. One has been
wisely inserted in the Great of the Organ in the Carnegie Music Hall,
Pittsburgh, Pa.
FORMATION. The DOPPELFLO'TE, 8 FT., is usually formed of pipes of two
different scales and treatments, those from CC to BB being medium-scaled GE-
DECKT pipes with single mouths, voiced to yield a full, round tone to carry clown, as
closely as practicable, the characteristic quality of the principal portion of the
stop. From tenor C to the top, each pipe has two mouths, placed on the opposite
and narrower sides of a block in which a semicircular depression has been cut,
leaving two thin lower lips to the mouths. The mouths and caps are of the English
form, the lower lips being somewhat widely nicked, and the upper lips arched and
carefully formed and rounded. These and all other details are correctly shown
in the accompanying illustration, Fig. 13. In the -Drawing i, is given a Longitu-
dinal Section of a pipe, cut through the mouths, at A, and showing the forms of
the block, cap, and wind-ways. Above, at B, is given a section of the stopper,
edged with cork and covered with soft leather. A Front View of one of the mouths
is given in Drawing 2, showing its height, the arching of the upper lip, the nicking
of the lower lip, and the lateral splays. In the Drawing 3 is given a Transverse
Section through the mouths, showing the depression in the block, the lower nicked
lips, the windways, and the upper edges of the caps. All the parts are drawn in
correct proportion and to scale. It would seem to have been the practice of the
German organ-builders to use different woods in the formation of their DOPPEL-
FLdTEs. In that on the Third Manual of the Organ built by Ladcgast for the
Cathedral of Schwerin the pipes are of pine fronted with oak; and in that on the
First Manual of the Organ of St. Paul's Church, in the same city, the pipes are of
pine fronted with oak and mahogany. In the Roosevelt DoppELFLtiXE, to which
allusion has been made, the pipes have sides of sugar-pine and fronts and caps of
close-grained mahogany. These are examples of fine pipe-making,
* We may here record the fact that, for the first time in the history of organ-building, three
independent tonal subdivisions were placed on a single clavier that of the First Organ;
namely, one unexpressive and two independently expressive. So far as we know this remarkable
and beautiful arrangement of tonal apportionment remains unique to-day (1920), See "Tha
Organ of the Twentieth Century," p. 334,
THEIR ARTISTIC REGISTRATION
103
SCALE. The pipes of the DOPPELFLOTE, having double mouths,
require a scale providing a considerable depth in proportion to width.
This is exemplified by the following internal measurements taken
from the Roosevelt standard scale: Tenor C pipe 2)^ inches in
width by 3^ inches in depth; middle c 1 pipe i% inches in width by
2^- inches in depth; c 2 pipe % inch in width by iffe- inches in depth.
This is an example of an irregu-
lar scale, arrived at by experience.
It will be observed that the pro-
portions of depth to width vary
as the scale ascends; the C pipe
being a little over one and a half
its width in depth; while the c 2
pipe is a little under twice its
width in depth. The height of the
mouth at the spring of its arched
upper lip ranges between one-
third and one-half its width, ac-
cording to the wind-pressure and
the volume of tone required.
The arching and thickness of the
upper lip are also important fac-
tors in tone production that the
voicer must decide.
TONE AND REGISTRATION.
The DOPPELFLOTE, when proper-
ly scaled and artistically voiced,
yields a singularly full and pure
unimitative flute-tone, having
more filling-up power and better
mixing quality than any other
unison covered stop of medium
power in the Organ. It is especi-
ally this mixing quality that gives
the stop its great value in refined
registration. Owing to the rarity
of the stop, organists have few opportunities of learning its import-
ance in tonal coloring. But having a beautiful DOPPELFLOTE in our
own Organ, and having carefully observed the use made of it in regis-
tration by a large number of distinguished English, French, and Ameri-
can Organists, during a period of eight years, supplemented by our
own studies, we have been able to learn the great value of the stop
Fig- 13
104 ORGAN-STOPS
as, perhaps, no other man has done. The DOPPELFLOTE is specially
effective in combination with the softer-toned lingual stops of all
classes, imparting to their voices firmness and fullness, and in some
cases considerable richness, without injuring their characteristic
tonalities. In this direction, however, the DOPPELROHRGEDECKT or
DOPPELROHRFLOTE may be preferred (q. v.}.
DOPPELFLOTENBASS, Ger. A medium-scaled covered wood
stop of 1 6 ft. pitch, formed with double mouths. This stop, which
furnishes the true bass to the DOPPELFLOTE, 8 FT., has its proper
place in- the Pedal Organ, where its full and smooth fluty voice
is of great value for accompaniment and in combination with the
lingual stops. A fine example of this uncommon stop exists in
the Second Pedal Organ of Schulze's important instrument in the
Marienkirche, at Liibeck, It is described thus:
*' DOPPELFLOTENBASS 16'. Holz. Eine weite, gedeckt Stimme
mit doppelten Labien. Der Ton ist etwas voller und runder, als
der des SUBBASS, und bildet zu den sanften Stimme der Orgel den
schonsten Bass."
DOPPELGEDECKT, Ger. The name given to a covered wood
stop of 8 ft. and 16 ft. pitch, the pipes of which are of large scale,
and made, after the fashion of the DOPPELFLOTE, with two
mouths: the scale, however, differs from that of the DOPPELFLOTE,
in not having so great a depth in proportion to width. The tone is
fuller and richer than that of the ordinary single-mouthed GEDECKT,
8 FT., or BOURDON, 16 FT.
DOPPELROHRPLOTE, Ger. A labial stop of 8 ft. and 4 ft.
pitch, which, as the name implies, is a modification of the DOPPEL-
FLOTE in so much as it is what is known as a half-covered stop, while
thje DOPPELFLOTE is a wholly-covered one. It is properly made of
wood throughout when of 8 ft. pitch; and when of 4 ft, pitch, of
wood save in its top octave, which is properly of metal, the pipes
there being too small to be conveniently made of wood. The wood
pipes are constructed in all essentials similar to those of the DOPPEL-
FLOTE (Pig. 13), with the exception of the stopper, which is per-
forated as in the normal ROHRFLOTE (q. v.).
TONE AND REGISTRATION. Owing to the perforation of the
stoppers of its pipes, the tone of the DOPPELROHRFLOTE is brighter,
lighter, and more open than that of the DOPPELFLOTE; and on this
account is better adapted for combination with the tones of the
softer lingual stops. This fact should incline the designer of a Con-
THEIR ARTISTIC REGISTRATION 105
cert-room Organ to place the DOPPELROHRFLOTE in the division
devoted to the stops representing the wood-wind forces of the or-
chestra. Placing the stop there, he will properly insert the fuller-
toned DOPPELFLOTE in the First or Great Organ, where its good
mixing voice will prove valuable in registration. The position of the
stop, of either of its pitches, in the Church Organ will properly de-
pend on the general tonal apportionment of the instrument. Ex-
amples of the DOPPELROHRFLOTE of both 8 ft. and 4 ft. pitches are to
be found in the Great of Mullet's Organ (1843) in the Catholic
Church, at Katcher. Other examples, of both pitches, are inserted
in the Echo of the Grand Organ in the Cathedral of St. John, Bres-
lau.* The DOPPELROHRFLOTE, 4 FT. , is an octave stop having all the
properties which render a labial stop of that pitch valuable in refined
registration. It is both a good harmonic-corroborating and timbre-
creating stop, the latter on account of its clear fluty voice. It can
be properly associated with the DOPPELFLOTE, 8 FT., in any suitable
division of the Organ; or, as in the instances given, with the unison
stop of its own class.
DOPPELROHRGEDECKT, Ger. While this term may be
accepted as another name for the DOPPELROHRFLOTE (q. v.), the
stops alluded to being similar in general formation while differing
somewhat in scale proportion, the term may be properly applied to a
stop, of 8 ft. pitch, yielding a fuller tone than either that stop or the
DOPPELFLOTE. To produce this more assertive voice, the scale of
the DOPPELROHRGEDECKT is larger than, and not so deep in propor-
tion to width as are the scales of the other stops alluded to. In
preference to duplicating the DOPPELFLOTE, it may, on the leading
principle of tonal variety, be desirable to introduce the DOPPEL-
ROHRGEDECKT, 8 FT., along with the DOPPELROHRFLOTE, 4 FT., in a
contrasting division of the Organ.
DOPPELSPITZFLOTE, Ger. As the name implies, this stop
is a variation of the SPITZFLOTE (q. .) its pipes having double
mouths. It has been made of metal and wood and of 8 ft. and 4 ft.
pitch. An example, of the latter pitch and of wood, exists in the
Echo of the Grand Organ in the Church of St. Mary Magdalene,
* In our remarks on the formation of the DOPPELFLOTE, we mentioned the use of different
hard woods in the construction of its pipes as displaying conscientious care on the part of Ger-
man organ-builders. Here, again, we have similar evidence of this care, and examples of fine
pipe-making. The DOPPELROHRFLSTES in the Organ in the Catholic Church, at Katcher, are
chiefly of oak, pine "being used only in the treble of the 8 ft. stop. In the same stops in the Organ
in the Cathedral of Breslau all the pipes are made of maple. When will such a commendable
practice be followed by the organ-builders of this country?
io6 ORGAN-STOPS
Breslau. Wood was, in all probability, used on account of the con-
venience it afforded, in the quadrangular pipes, for the formation
of the double mouths. It is also probable that, as in the case of the
other double-mouthed wood stops, the pipes of the DOPPELSPITZ-
FLOTE were made of greater depth than width; a treatment which
could not be applied to metal pipes. We are of opinion that very
beautiful and valuable voices could be added to the Organ by stops
of 1 6 ft. and 8 ft. pitch, of wood, of medium scale, and constructed
in the pyramidal form of the DQPPELSPITZFLOTE. But we are afraid
that the time and trouble such stops would entail, beyond what the
construction and voicing of ordinary straight and single-mouthed
wood stops call for, will effectually prevent organ-builders advocat-
ing their introduction.
DOUBLE BASSOON, The lingual stop of 16 ft. pitch, the tone
of which imitates as closely as practicable that of the Double Bas-
soon of the orchestra, the compass of which extends from BBBb to
F, and, accordingly, covers the compass of the Pedal Organ, with the
exception of the two top notes, to which department of the Organ
the stop may be considered to properly belong. See BASSOON and
CONTRAFAGOTTO.
DOUBLE CLARINET. A lingual stop of 16 ft, pitch, the
pipes of which are similar in formation to those of the unison CLARI-
NET (q. v.). The stop was, so far as we can learn, first made by
Wedlake, organ-builder, of London, in 1863, and inserted in an
important Chamber Organ. Artistically voiced, and placed in an
expressive division, the DOUBLE CLARINET would prove a valuable
voice in the Concert-room Organ, affording a different tonal coloring
from that of the DOUBLE BASSOON, 16 FT. It would also be valuable
as a Pedal Organ stop, carrying down the manual CLARINETS.
DOUBLE DIAPASON. The stop formed of large-scaled open
cylindrical metal, or quadrangular wood, pipes, voiced to yield pure
organ-tone. When introduced in a manual Organ it is of 16 ft.
pitch, and when placed in the Pedal Organ it is of 32 ft. pitch. In
almost all satisfactory examples of the manual stop the pipes are
formed of metal throughout their compass. In all the finest existing
examples the pipes are of tin or high-class alloy; their larger pipes
being displayed in the cases, forming salient towers or other effec-
tive features. A notable exception to this general and desirable
rule is to be found in the remarkably fine stop, labeled SUB-PRINCI-
PAL, 16 FT., in the Great of the Organ in St. Bartholomew's Church,
THEIR ARTISTIC REGISTRATION 107
Armley, Yorkshire, which is formed of open wood pipes from CC to
B, and of metal from C T to the top note. Though perfectly successful
in this instance, such a method should not be ventured upon by any
one less skilled than the master who made and voiced the Armley
stop.
The manual DOUBLE DIAPASON, 16 FT., has its pipes formed in
all respects similar to those of the manual DIAPASON, 8 FT. (q. .),
while its relative scale should be smaller; its tenor C (8 ft.) pipe
being from two to four pipes less in scale than the CC (8 ft.) pipe of
the unison stop. The voices of the respective stops should follow
the same relative proportions. The tone of the double stop should
never dominate that of the foundation unison stop.
The DOUBLE DIAPASON, 32 FT., as introduced in the Pedal Organ,
is formed throughout of either metal or wood; but on account of its
great cost metal has been comparatively seldom used. In Organs of
the first magnitude the stop exists in both materials complete, as in
the Concert-room Organs in St. George's Hall, Liverpool, and the
Centennial Hall, Sydney, N. S. W. The DOUBLE DIAPASON scale
varies greatly in different metal stops : for instance, the CCCC pipe
in the Liverpool Organ measures 25 inches in diameter; that in the
Organ built by Walcker for the Music Hall, Boston, Mass., in 1863,
was 22% inches in diameter; that in the Organ in the Monastery
Church, at Weingarten, is about 15^ inches in diameter; and that
in the celebrated Haarlem Organ is 15 inches in diameter. In the
last named three Organs English tin was used for all the displayed
pipes of the stop including those of which the diameters are given.
The bottom octave of the stop in the Liverpool Organ is of thick zinc.
Several stops having scales ranging between those just quoted are to
be found in other important Organs. While it is not possible or
desirable to lay down a hard and fast rule for general adoption in
the scaling of this exceptional stop, we can strongly recommend the
use of the moderate scales favored by the most advanced German
and other Continental organ-builders. It is, however, quite safe to
say that with proper voicing and a copious winding there can be no
necessity to exceed the scale of the SUB-PRINCIPAL, 32 FT., of the
Haarlem Organ.
The Pedal Organ DOUBLE DIAPASON, 32 FT., formed of wood
throughout is the prevailing form, and when properly scaled and
voiced is all that can be desired in so grave a stop. It must be borne
in mind that its office is not to disturb the unison fundamental tone
of the Pedal Organ, but to enrich it by harmonic creation. As in
the case of the metal stop, inordinate scales have been adopted with
io8
ORGAN -STOPS
no appreciable advantage. One extreme example may be given: the
DOUBLE DIAPASON, 32 FT., in the Organ in the Wanamaker Store in
Philadelphia, Pa., has its CCCC pipe measuring internally 22%
inches in width by 27^ inches in depth.* The Scales adopted by
two of the most celebrated German builders may be safely accepted
as the most desirable maximum and minimum ones for general use.
The fine stop in the Organ, by Schulze, in St. Bartholomew's, Arm-
ley, measures 14^ inches in width by i8J^ inches in depth. This
scale is larger than that adopted by the same builder for his DOUBLE
PRINCIPALS in the Organs in Bremen Cathedral and St. Mary's
Church, Wismar, which measure 12 inches in width by 15 inches in
depth. The equally distinguished builder Ladegast used a still
FIG. 14
smaller scale n % inches in width by 14*^ inches in depth for the
32 ft. stop in his Organ in the Cathedral of Schwerin. We can speak
from experience of the satisfactory character of a stop of a similar
small scale in an Organ constructed to our Specification. As it is of
more importance in so large and grave a stop that it should speak
promptly and distinctly than that its exact timbre should be con-
sidered, steps were taken by the German builders to secure this de-
sirable result by the most effective means at their disposal, leading
to the application of the harmonic-bridge. Different forms of the
bridge were employed, which, by creating harmonic upper partial
tones, added greatly to the clearness of speech and tonal value of
the stop. The unique method in which the bridge was applied by
Schulze, is shown at A in Pig. 14, which is a Section through the
* This monster pipe called for in its construction over one thousand square feet of j inch
sugar pine. It weighs 1,735 Ibs,
THEIR ARTISTIC REGISTRATION 109
mouth of the largest pipe of the DOUBLE PRINCIPAL, 32 FT., in his
Armley Organ. In Section B is shown the usual form and manner
of applying the harmonic-bridge. The bridge . imparts not only
prompter speech to the pipes but a string quality to their tones.
This must have been sufficiently marked in the stop in the Schwerin
Organ to decide Ladegast to label it VIOLON, 32 FT.
^ In German Organs the DOUBLE DIAPASONS bear the more appro-
priate names PRINZIPAL, 16 FT., PRINZIPALBASS, 16 FT. In the Pedal
Organ they appear under the names PRINZIPALBASS, 32 FT., GROSS-
PRINZIPAL, 32 FT., and GROSSPRINZIPALBASS, 32 FT. In French Or-
gans the stops appear as: MONTRE, 16 FT. (when mounted in the
case), PRINCIPAL BASSE, 16 FT. and 32 FT., CONTREBASSE, 16 FT.,
FLUTE OUVERTE, 16 FT. and 32 FT., and simply FLUTE, 32 FT. In
Italian Organs the stops are commonly named PRINCIPALS BASSO,
1 6 FT. and 32 FT. In Spanish Organs the usual name is FLAUTADO.
DOUBLE DULCIANA The name properly employed to desig-
nate a manual stop of 16 ft. pitch, formed in all respects similar
to the unison DULCIANA. When correctly made and artistically
voiced by a master-hand it yields a pure organ-tone of extreme
beauty and refinement. It is greatly to be regretted that voices of
this class are so much neglected by organ-builders and so little
desired by organists; but these tacts may be accounted for by the
modern and present craze for loud tones and high wind-pressures.
Purity, refinement, and delicacy of intonation seem to be at a dis-
count in the organ-building of to-day. The DOUBLE DULCIANA
forms a perfect Double for the Choir Organ, especially should it be
in an uninclosed division; and it is the most beautiful open stop of
16 ft. for the Chamber Organ. On account of its small scale, it fur-
nishes desirable display pipes, while its soft and beautiful tone favors
such an exposed disposition. When inserted in the Pedal Organ, it
is correctly labeled DULCIANA, 16 FT., being simply a unison stop in
that department. In the Pedal Organ, a DOUBLE DULCIANA would
be a small-scaled open metal stop of 32 ft. A DOUBLE DULCIANA,
1 6 FT., exists in the Swell of the Organ in St. Mark's Church, St.
John's Wood, London. For further particulars see CONTRA-DUL-
CIANA and DULCIANA.
DOUBLE MELODIA An open wood stop, of medium scale
and of 1 6 ft. pitch, belonging to the manual department of the
Organ. On account of the great length of the pipes in its lower
octave, and especially when height is limited for its accommodation,
the stop is usually formed of open pipes from tenor C (8 ft.) to the
no ORGAN-STOPS
top note; the bass octave being in LIEBLICHGEDECKT or some suit-
able form of covered pipes, of soft tone, voiced to carry down as
closely as possible that of the open pipes. The open pipes, like those
of the unison MELODIA, have inverted mouths. A fine example of
the DOUBLE MELODIA, 16 FT., exists in the Great (tower section) of
the Organ in the Cathedral of the Incarnation, Garden City, Long
Island, N. Y. See MELODIA.
DOUBLE TROMBONE. The Pedal Organ lingual stop of
32 ft. pitch, similar in all respects to the CONTRA-TROMBONE, 32 FT.
(2- )
DOUBLE TRUMPET. The manual lingual stop of 16 ft.
pitch, formed of metal pipes of the same construction, and voiced in
the same manner, as the pipes of the unison TRUMPET, 8 FT., of
which stop it is the true Sub-octave. It is the most appropriate and
generally useful lingual stop of 16 ft. pitch for the Great Organ;
forming with the TRUMPET, 8 FT., and CLARION, 4 FT., the complete
family of the Trumpet-toned stops. To be of full value in combina-
tion, the voice of the DOUBLE TRUMPET must be markedly sub-
ordinate to that of the unison stop, so that it can be freely used
without destroying the dominance of the unison tone of the TRUM-
PET. This subordination also renders the DOUBLE TRUMPET ex-
tremely valuable in general registration with the foundation stops
of the Great Organ, and especially so if it is inclosed so as to be flex-
ible and expressive. It may be accepted as an axiom that in proper
organ-building and tonal appointment every lingual stop must be
endowed with powers of flexibility and expression. Without such
powers, lingual stops are practically valueless in solos and in refined
and artistic registration. For particulars of formation, see TRUMPET.
DOUBLETTE, Fr. In French Organs the DOUBLETTE is an
open cylindrical metal stop of 2 ft. pitch, belonging to the harmonic-
corroborating series of the manual foundation-work, properly yield-
ing pure organ-tone.* The term is used by German and English
* The following particulars, from the pen of a distinguished French authority on the Organ
will be read with interest and profit by the Organist:
"La DOUBLKTTK, ou simplemexit le deux-pieds, car c'est le ton de son plus grand tuyau.est
par cette raison m&me la double octave du huit-pieds pris pour base g6ne>ale du ton d'orgue,
puisque le huit-pieds, avons-nous dit f cst a 1'unisson de la voix commune de 1'hornnie. La dou-
blette est done 1'octave du prcstant, ouvcrte comme lui, de taillc m6diocre commelui.etcomme
lui detain fin. Cependant, la finesse cle l'6tain est g6n6ralement ne"glig6e, parce qutsl'exiguitd
de diametre et de hauteur compense, ou est cens6e compenser le tranchant que donne la finesse
du mc'tal.
"Elle embrasse toute l'6tendue du clavier, parce que son harmonic sert de liaison entreles
.ieux de mutation et les fonds, et donne aux jeux d'anches me"lang6s de jeux de fond un degr6
THEIR ARTISTIC REGISTRATION in
organ-builders to designate a stop composed of two ranks of open
metal pipes of different pitches. In the Organ in the Cathedral of
Merseburg, a DOUBLETTE of 4 ft. and 2 ft. ranks is inserted in the
Hauptwerk: and in the same Division of the Organ in the Refor-
mirten Kirche, Elberfeld, there is one of 2% ft. and 2 ft. The
DOUBLETTE is not a common stop in English Organs, but examples,
of 2 ft. and i ft. ranks, exist in the Great and Swell of the Organ in
St. George's Hall, Liverpool. We are strongly in favor of DOU-
BLETTE s being made timbre-creating by their ranks being formed of
pipes of contrasting tonalities: they would then belong to the tonal
appointment of the entire Organ, not specially to its foundation-
work,
DRUMS. Ital., TIMPANI. Ger., PAUKEN. Drums were intro-
duced in many old Organs, and were, in some cases, mounted on the
case and mechanically beaten by figures of angels. After having
long been classed among the curiosities of the organ-builders' art,
Drums are again being introduced in Theater Organs and in a cer-
tain type of Concert-room Organ. To what extent such accessories
are desirable may be left to individual opinion: we would rather see
them remain as obsolete curiosities,
DUIFLOTE, Ger. The old and practically obsolete name for
the double-mouthed, covered wood stop now known as the DOPPEL-
FLOTE (q. v.).
DULCET. An open metal labial stop of 4 ft. pitch. In its
proper and most desirable form it is a DULCIANA OCTAVE, and has
been sometimes called DULCIANA PRINCIPAL in English organ
nomenclature. The pipes are of small scale, and formed and voiced
in all respects similar to those of the unison DULCIANA, 8 FT. Its
scale may properly vary according to the tonal apportionment of
the division of the Organ in which it is placed: but when associated
with the DULCIANA in any manual division its scale should be two
or three pipes less, accordingly, the CC pipe of the DULCET, 4 FT.,
d'acuit6 particuliere, Cependant, quoique placee aux claviers de grand orgue et depositif, elle ne
se trouve guere sur celui de r'ecil, ou ne ngurent que des jeux de solo, ni sur le clavier connu id.
Paris surtout) sous le nom de clavier de bombardes. La doublette ne s'emploie guere seule que
par accident, par exemple pour imiter 1'effet d'un sifflet adouci. Unie a la quarte de Nazard,
qui est son unisson, elle siffle avec une grande vigueur. Melangee aux fonds de qua tre-pieds, elle
y produit 1'effet semblable ; mais sit6t qu'on la mele aux huit-pieds, surtout sans y joindre comme
transition le quatre-pieds, elle crie, siffle ou gemit desagreablement et fait perdre toute noblesse
aux r6gistres de fonds. Cependant, j 'ai oul souvent la doublette, accouplee a un seul bourdon de
seize, produire sous des doigts habiles et dans certains passages un effet singulier, que je n'oserais
direagr6able, de peur de donner aux organistes mediocres 1'idee de 1'essayer; 1'effet ne serait pas
le nifime." Regnier.
112 ORGAN-STOPS
would be of the same diameter as the D or D# pipe of the unison
DULCIANA. The most desirable material for the stop is tin, and
next to that Hoyt's Two-ply Pipe Metal, which lend themselves to
the delicate manipulation necessary for the production of the light
and singing "silvery tone," characteristic of the stop. The DULCET,
4 FT., was introduced by Samuel Green (probably between 1780 and
1790), when soft and refined organ-tones were appreciated by music
lovers, and was used by him, in association with the DULCIANA, in
certain of his Swell Organs under the name DULCFANA PRINCIPAL.
Octave and Super-octave stops of the DULCET class are prac-
tically unknown in modern Organs; the prevailing craze for high
wind-pressures and the crude taste (or want of taste) for musical
noise, having swept away such beautiful and winning voices. Or-
ganists have to learn what tonal wonders lie in refined registration
into which such voices enter with tonal effects absolutely unknown
to them on the noisy Organs of fco-day.
DULCIAN, DOLCIAN, Ger. The name given to a lingual stop
yielding a soft tone, resembling that of the BASSOON, and of 8 ft.
and 1 6 ft. pitch. It is formed with striking- and free-reeds, and with
slender inverted conical and pyramidal resonators of metal and
wood. Examples are found, in either of the pitches, in different
divisions of the Organ. The DULCIAN, 8 FT., exists in the Echo of
the celebrated Haarlem Organ ; and in the Pedal of the Organ in the
Evangelical Church, Minister. The DULCIAN, 16 FT., is introduced
in the Oberwerk of the Organ in the Cathedral of Konigsberg; in the
Pernwerk (Third Clavier) of the Organ in St Petri-Kirche, Berlin;
in the Rtickpositiv of the Organ in St. Catherine's Church, Ham-
burg; as a free-reed stop, in the Pedal of the Organ in the Cathedral
of Merseburg; and in the Piano Pedal of the Organ in the Cathedral
of Schwerin, where it is a free-reed furnished with resonators of wood.
DULCIANA. A stop formed of open labial pipes of cylindri-
cal form and small scale, of 8 ft. pitch in the manual Organs and
1 6 ft. pitch in the Pedal Organ, The stop appears to have been in-
troduced for the first time by John Snetzler in the Organ he built in
t754, under the direction of the distinguished Dr. Burney, for the
Church of St. Margaret, Lynn Regis, Norfolk, liver since then the
DULCIANA has been a great favorite with English organ-builders,
and numerous beautiful examples exist in English Organs, We have
not been able to find the DULCIANA, 8 FT., in its true form in any
French Organ; and neither it, nor any equivalent under another
name, is mentioned by Rcgnier, We find a DOLCIANE, 8 FT., in the
THEIR ARTISTIC REGISTRATION 113
Rcit of the Chamber Organ built for the late Alexandre Guilmant
by Mutin, of Paris : this was in all probability suggested by the Eng-
lish stop, with which M. Guilmant was familiar. On the other hand,
however, we find an octave stop, labeled DULCIANA, 4 FT., in several
important Organs; for instance, in the Recit of the Organ in the
Church of Saint-Sulpice, and in the Bombarde of the Organ in the
Church of Saint-Eustache, Paris. We have not been able to find
any trace of the DULCIANA in connection with old German Organs;
but it appears, invariably on the First Clavier (Hauptwerk), in
several of Walcker's large Organs, as in those in the Cathedral of
Riga, and St. Petrikirche, Liibeck.
FORMATION AND SCALE The pipes of the DULCIANA are cylindrical and
formed similar to those of the DIAPASON, 8 FT., of which it is, strictly considered,
the proper diminutive. Its scale being small and its voice having to be extremely
clear and perfectly uniform in tone, the pipes must be carefully made of tin or
high-class metal. The scale of the true DULCIANA, 8 FT., varies slightly, according
to the class of Organ for which it is designed, the position it is to occupy in the
Organ, and the acoustical conditions under which it has to be heard. In so delicate
and sensitive a stop as the true English DULCIANA every condition should be
carefully considered affecting its scaling and voicing. The following scales, in two
desirable ratios, will be found, with careful formation and artistic voicing of the
pipes, to produce satisfactory results under favorable conditions:
DULCIANA PIPE DIAMETERS IN INCHES, RATIO 1:2.519.
PIPES CC C c 1 c 2 c^ c4
I. 3.25 2.05 1.29 0.81 0.51 0.32
II. 3-38 2.13 1.34 0.84 0.53 0.33
DULCIANA PIPE DIAMETERS IN INCHES, RATIO 1:2.66.
PIPES CC C c 1 c 2 c3 c4
I. 3.34 2.05 1.25 0.77 0.47 0.29
II. 3.48 2.13 1.31 0.80 0.49 0.30
The smaller scales to the ratio i : 2.519 give a comparatively full treble; while
tnose to the quicker ratio 1 : 2.66 keep the lower octaves comparatively full toned.
The selection made by the tonal artist will, or should, depend on the special stop-
appointment of the division of the Organ in which the DULCIANA is to be placed.
In the tradesman organ-building of to-day, and the general don't-careism on the
part of organists, such considerations of tonal propriety and refinement are either
not realized or are ignored. When will the organist arise who will gloat over the
beauties of his glorious instrument, as the violinist gloats over the wonders of his
Stradivarius? The width of the mouth of the DULCIANA pipe should not exceed
one-fifth the internal circumference, while, except in a stop intended for a true
Chamber Organ, it should rarely be less. The height of the mouth may vary from
one-fifth to one-third its width, according to the method of the voicer, the wind-
pressure employed, and the character of the tone desired. The mouths of all the
pipes, except, perhaps, those of the top octave, should be furnished with ears of
ri4 ORGAN-STOPS
slight projection, aiding clean articulation. On no account arc the pipes to tc
slotted for tuning if the true tone of the DULCIANA is desired. The larger pipey
should be furnished with tuning-slides, as recommended for the DIAPASON, This
expedient will prevent the pipes being injured by repeated tuning with the cones ;
and will secure uniformity in their voices, their open tops remaining at all times of
the correct internal diameters. Such matters as these should be attended to by
the organ expert in his Specifications.
TONE AND REGISTRATION. The true English DULCIANA has a
pure organ-tone of a sweet singing, silvery, quality a tone seldom
heard in Organs built during late years. To ears vitiated by con-
stant association with powerful and more or less coarse musical
sounds the product of inordinate wind-pressures- such refined
tones as those of the DULCIANA, in their purity, seem poor and in-
sipid; accordingly, voicers proceed to improve the stop by what they
call giving it color; removing it from its proper and time-honored
place in the tonal appointment of the Organ, and throwing it among
the SALICIONALS and quasi string-toned stops. The DULCIANA is
either a DULCIANA or it is not one; and every voicer should be re-
quested to remember that important fact, and, at the same time,
asked to improve his overblown labial and lingual stops, leaving the
DULCIANA queen of the soft organ-toned cantabile group.
The place for the DULCIANA, 8 FT., is in the Choir or chief accom-
panimental division of the Church Organ; certainly not in the Great
(or Hauptwerk) where Walckor has invariably placed it. In the
Concert-room Organ its proper position is in the softest-toned divi-
sion, whatever it may be. In such approved situations it will prove
most valuable in registration, furnishing a foundation or background
for numerous combinations of tone-colors. Its reposeful and can-
tabile voice renders it of great use in quiet solo effects; and in giving
body to combinations of special tones without in any marked manner
affecting their normal character. In registration with soft-toned
open, covered, and half-covered FLUTES, 8 ft. and 4 ft. pitch; with
delicately voiced string-toned stops; with the more refined lingual
stops; and with its Octave, the CCELKSTINA, 4 FT., and such a com-
pound stop as DULCIANA CORNET, or, better still, the HARMONIA-
^ETHERIA, the DULCIANA contributes to the production of numerous
compound tones of great beauty, refinement, and charm, Let not
the true DULCIANA disappear from the Organ.
DULCIANA CORNET. The compound harmonic-corroborat-
ing stop, commonly formed of five ranks of high-pitched DULCIANA
pipes; yielding delicate silvery tones, representing the higher upper
partials of the prime unison, suitable for combination with the softer
THEIR ARTISTIC REGISTRATION 115
labial and lingual voices of the Organ, and, accordingly, entering
into countless tonal combinations into which the ordinary full-toned
MIXTURES could not possibly be introduced. The stop has been
formed of octave- and fifth-sounding ranks, and also of octave-,
third-, and fifth-sounding ranks, in all cases, owing to their high
pitch, requiring several breaks in their compass. The following is
the composition of an example of proved excellence under all tests
in combination and artistic registration:
DULCIANA CORNET V. RANKS.
B
to b
C t0 B 12 15 1 7 10 22.
8 12 1; 19 22.
This stop carefully regulated and graduated in tone in strict
accordance with the natural laws of compound musical sounds,
becomes one of the most useful agents in the building up of expres-
sive tonal structures of a refined and fascinating character. It is so
sympathetic that it can be used with a single DULCIANA or any soft
unison stop in the Organ. The DULCIANA CORNET or some equiva-
lent, should be introduced in every Organ of any importance ; and
it should be the only compound stop in the true Chamber Organ.
DULCIANA PRINCIPAL. The term employed by Samuel
Green, organ-builder to King George the Third (1780-1796), to
designate, according to the English nomenclature, a stop of 4 ft.
pitch and of the DULCIANA class : what would now be more correctly
called DULCIANA OCTAVE. It has been said: ''The organs built by
Green are characterized by a peculiar sweetness and delicacy of tone,
entirely original; and, probably, in this respect he has never been
excelled." It is to be hoped that the mantle of Green will some day
fall on the shoulders of one of the aspiring organ-builders of the
twentieth century.
DULCIMER. The name given by Thomas Schwarbrook to a
stop, composed of metal strings, inserted in the Organ erected by
him, in 1733, in the Church of St. Michael, Coventry. It was, in all
probability, sounded by a hammer action, as the real Dulcimer was
played. Respecting this notable Organ, Dr. Rimbault remarks;
" This noble instrument (Schwarbrook's masterpiece) cost 1400.
It originally contained three remarkable stops the HARP, LUTE;
and DULCIMER; but, in consequence of the 'difficulty of keeping the
strings in tune/ they were removed in 1763."
ORGAN-STOPS
DUOPHONE. The name given by G. W. Till, of Philadelphia,
to an open labial stop, of 8 ft. pitch, recently invented by him; the
pipes of which are of wood, having inclined sides, and mouths fur-
nished with metal upper lips and cylindrical harmonic-bridges.*
The tone of the stop is dual, the prime tone and its first upper partial
being produced in almost equal volume, and is remarkable on ac-
count of its penetrating and traveling quality without undesirable
loudness.
E
ECHO BOURDON. A BOURDON, 16 FT., of small scale and
soft intonation, the pipes of which are of wood, preferably of oak or
maple, finished as thin as conditions will permit, and otherwise con-
structed like the ordinary BOURDON. This stop differs from the
I ;EBLICHGEDECKT in its tone, chiefly on account of the different
proportions and treatment of the mouths of its pipes, and, accord-
ingly, in a special style of voicing. The best tone is obtained on
wind of low pressure, between I J^ and 2 J/o inches. The stop may be
considered the English equivalent of the German BOURDONECHO,
and is an ideal stop for the true Chamber Organ.
ECHO DIAPASON. The appropriate name for a pure organ-
toned unison stop, the scale of which is midway between that of the
full-toned DIAPASON, 8 FT., and that of the true DULCIANA, and tlK
voice of which i* also a medium one, Such a medium stop is ex-
tremely valuable in certain divisional tonal apportionments, fur-
nishing foundation organ-tone commensurate with the requirements
of the divisions,
The pipes of the ECHO DIAPASON are formed precisely as those of
the foundation DIAPASON; their only difference lying in the subdued
character of their tones largely due to their being voiced on wind of
low pressure not exceeding 2% inches, but preferably less. The
following scale is suitable for the stop ;
ECHO DIAPASON PIPJB DIAMETERS IN INCHES, RATIO 1:2.66.
PIPES CC C e* c 9 c* c*
4.82 2.95 1,81 i. ii 0,68 0.42
The mouth, in width, should not exceed two-ninths of the inter-
* For full particulars and drawings of thif, stop, See " The Organ of the Twentieth Century,
^w,oa 109*451-2,
THEIR ARTISTIC REGISTRATION 117
nal circumference of the pipe, and should be cut up only sufficient to
produce the pure organ-tone required. The upper lip must not be
leathered; and the pipes must not be slotted for tuning.
TONE AND REGISTRATION. Like all unison stops yielding pure
organ-tone, the ECHO DIAPASON furnishes a foundation or back-
ground for fine effects of tonal coloring, to which it gives richness
and neutral body. The delicate tone of this stop is extremely recep-
tive and lends itself to the most refined nuances and effects of tonal
light and shade. It combines perfectly with all the softer tonal
colors that can be thrown upon it by the contrasting voices of labial
and lingual stops; in this direction resembling the still softer and
equally pure-voiced DULCIANA (q. .).
ECHO DULCIANA. A small-scaled, open, metal stop, of 8 ft.
pitch, the pipes of which are formed similar in all respects to those
of the English DULCIANA, of which stop it is the proper diminutive
in scale and tone. It may be said to occupy an intermediate posi-
tion between the DULCIANA and the Vox ANGELICA, 8 FT., when
these stops are properly related. It should have a pure organ-tone
of a cantabile character, rendering it peculiarly suitable for the
softest division of the true Chamber Organ. The following scale
may be accepted as suitable for the stop, relative to the scales given
for the DULCIANA (q. .), ratio 1 : 2.66:
ECHO DULCIANA PIPE DIAMETERS IN INCHES, RATIO 1:2.66.
PIPES CC C c 1 c a c* c4
2.84 1.74 i. 06 0.65 0.40 0.24
The width of the mouth should not exceed one-fifth the internal
circumference of the pipe; while its height may vary from one-fifth
to one-third its width, according to the wind-pressure employed,
the volume of the tone desired, and the special method of voicing
followed.
REGISTRATION. As the ECHO DULCIANA yields a tone which is
properly a diminutive of that of the DULCIANA, its offices in regis-
tration are practically similar with regard to the stops of soft tonal-
ity suitable for combination, and, accordingly, reference may be
made to the remarks under DULCIANA. As a single echo voice this
stop is the most desirable in the Organ. It completes the family,
which stands thus : DULCIANA, 8 FT. ; ECHO DULCIANA, 8 FT. ; DUL-
CET, 4 FT. ; and DULCIANA CORNET, V. RANKS.
ECHOFLOTE, Ger. The name that has been used to designate
an extremely soft flute-toned stop of 8 ft. or 4 ft. pitch, commonly
H8 ORGAN-STOPS
of wood and covered or half-covered, made still softer by inclosure
in a box. Sometimes the stop has been simply labeled ECHO; and
Locher remarks : "When this word alone appears on a draw-stop
knob, it indicates an exceedingly soft, flute-like stop, which is often
placed in a swell-box, separate from the main body of the Organ."
On the same single term, Seidel says: "A simple Echo consists of a
single flute-toned register of soft intonation' which stands behind the
Organ, and receives its wind through long conveyances. A com-
pound Echo consists of several stops, standing behind the Organ in a
separate box, the inside of which is lined with felt or cloth, so that
if this Echo-work is played upon, the tones will seem to come from
without the church. The Echo often contains also a MIXTURE, for
instance a CORNETT, which is then called the ECHO-CORNETT,"
ECHO GAMBA. The name found in certain English Organs,
designating a stop of the GAMBA class yielding a soft and somewhat
cold string-tone. An example, of 8 ft. pitch, exists in the Swell of
the Organ in York Minster, It is usually made of small-scaled metal
pipes from tenor C, the bass octave being added, very inappro-
priately, in covered wood pipes, as in the Organs in St, Saviour's
Church, Eastbourne, and All Saints' Church, Wokingharn. In its
original tonality the stop is of little interest or value.
ECHO OBOE. The name given by Edmund Schulzc, organ-
builder, of Paulinzclle, to an open wood labial stop, of 8 ft. pitch,
invented by him, the pipes of which are of small scale and about
twice their width in depth* The mouth is furnished with a small,
snaip-edged harmonic-bridge, which rests on a sunk and sloped
cap; the edge of the cap at the wind-way is thin; and the upper lip
is cut sharp.* The voice of the stop is a peculiar compound of reed-
tone and string-tone, and is singularly delicate and pleasing, though
not highly suggestive of that of the Oboe of the orchestra. A beau-
tiful example, made by Schutee, exists in the Echo of the Organ in
St. Bartholomew's Church, Armley, Yorkshire: it is of tenor C
compass, being grooved into the bass of the Vox ANGELICA, 8 FT.
Another fine example, by Abbott & Smith, of Leeds, exists in the
Echo of the Organ in the Parish Church of Leeds, Both these re-
markable stops speak on wind of i l /i inches pressure.
ENGELSTIMMB, Ger. The name originally given to a lingual
stop of the Vox HUMANA class; and subsequently to a stop also
* A full description and drawings of the Ecuo OBOIS arc given in our work, " The Art of Organ.
Building/' Vol. II., pp. 481-48**
THEIR ARTISTIC REGISTRATION 119
called Vox ANGELICA ( 2 . v.). Of the stop Seidel remarks: "ANGEL-
ICA (vox), die ENGELSTIMME, ein angenehmes, aber jetzt veraltetes
Rohrwerk zu 8'. Es geht nur durch die oberen Oktaven vom ein-
gestr. c an. Dieses Register scheint eine fruhere Art der Vox HU-
MANA zu sein."
ENGLISH HORN. Ger., ENGLISCH HORN. FT., COR ANGLAIS.
Ital, CORNO INGLESE. This stop derives its name from the orches-
tral instrument, which is a reed of the Oboe family, having a com-
pass extending from tenor E two octaves and a fifth. Accordingly,
the stop need not be carried below tenor C; but, as incomplete stops
are undesirable in a properly appointed Organ, it is proper to insert
it, as a stop of 8 ft. pitch, of the full compass. For further particu-
lars see COR ANGLAIS.
ENGPRINZIPAL, Ger. (fromeng narrow). The name which
has been sometimes used by the old German organ-builders to desi fe
nate a PRINCIPAL, 8 FT., of small scale and soft intonation, so as to
distinguish it from the foundation PRINCIPAL of full scale and tone.
ERZAHLER, Ger. The name given by Ernest M. Skinner,
organ-builder, of Boston, to an open metal labial &top, of 8 ft. pitch,
introduced by him in 1904. The ERZAHLER pipe is similar to that
of the GEMSHORN in being conical in form; but differing from it in
having the diameter of its top opening only one-fourth of the dia-
meter at its mouth line, in being slotted near the top, and having a
mouth width equal only to one-fifth of the larger circumference of
the body.
The tone of the stop is compound and singularly bright; this is
due to the octave being distinctly produced in combination with the
unison or prime tone. In addition to this there is a peculiarity in the
tone which suggested the somewhat fanciful name to the introducer
of the stop. Examples exist in several Skinner Organs.
EUPHONE, EUPHONIUM. Fr., Eupn6NE. Ger., EUPHON.
The name derived from the Greek word so^ovoq, and properly
given to a lingual stop, the tone of which is intended to imitate that
of the brass instrument of the same name, a member of the Saxhorn
family, the compass of which is a little over three octaves from CC.
The stop is, accordingly and properly, of 8 ft. pitch, extending, in
keeping with the Euphonium, over the richest portion of the manual
compass of the Organ, but necessarily carried through the higher two
octaves. Fine stops of the same tonality and 16 ft. pitch have been
made, and these are, perhaps, the more valuable. A EUPHON, 8 FT.,
I2O
ORGAN-STOPS
exists on the First Manual of the Organ in the Cathedral of Riga:
and representative examples of the stop of 16 ft. pitch exist in the
Roosevelt Organs in the Cathedral of Garden City, L. I., and the
Auditorium, at Chicago. All are free-reed stops,
FORMATION, The EUPHONE as made by German
and French organ-builders arc invariably free-reca
stops, and their tones are smooth and pleasing. Differ-
ent forms of resonators have been devised for the pur-
pose of producing the desired tone, some of very pec-
uliar form, but those most approved are cither of the
plain inverted conical form, shorter and of smaller
scale than those used for the TRUMPET, as adopted
by Roosevelt; or of the same form surmounted by
some contracting or shading device, preferably a short
truncated cone, as shown at the top of the resonator IP
the drawing of the complete pipe, 2, Fig. 1 5. In other,
and less effective, examples the inverted conical res-
onators are closed at top and slotted for the emission
of wind and sound. The drawings in Fig. 15 are
accurately made to scale from a C (4 ft. pitch) pipe
constructed by the well-known German organ-pipe
maker, August Luakhuff. All free-reed pipes re-
quire for their prompt and satisfactory speech large
boots; and this is shown in the Section i, which also
shows the free-reed, the block in which it is inserted,
and the tuning- wire. The length of the boot is 13^4
inches, while the entire length of the resonator is only
25 Ji inches. The largest diameter of the resonator
is 2 7 / H inches, and that of the opening at top i J^
inches. The length of the tongue is 2.25 inches and
its width 0,27 inch.
TONE AND REGISTRATION. The tones
produced by the several stops, of different
formation, that have been named EUPHONE
have varied greatly. It appears that in the
first and earlier forms no attempt was made
to imitate the voice of any particular wind
instrument. The first stop to which the name
appears to have been given was that inserted
in the Organ of the Cathedral of Beauvais,
FIG. 15 in 1829. During the following year, Sebas-
tian Erard inserted aEuPudNK, as an expres-
sive free-reed, in the Organ of the Royal Chapel of the Tuileries,
Paris. Regarding the tones of these stops nothing is clearly known,
but, judging from the peculiar forms of their resonators, which were
THEIR ARTISTIC REGISTRATION 121
short, covered, and slotted, their tones must have been somewhat
muffled and probably inclining toward the Bassoon quality. The
tones of the later examples, such as those made by Roosevelt, Luak-
huff, Zimmermann of Paris, and others, are open, smooth, and full;
but not strongly resembling those of the Euphonium.*
A soft, full tone, imitative of that of the Euphonium played by
a master, is greatly to be desired in the Organ; but, like that of the
orchestral Horn, it seems almost impossible to obtain it from lingual
pipes. It might, however, be produced by a dual stop, formed by
the artistic combination of a labial and a lingual rank of pipes.
Such a stop would be very valuable in impressive solos; just as the
Euphonium is par excellence a solo instrument: and one can easily
realize that it would also be extremely valuable in refined registra-
tion, especially in its proper 8 ft. pitch. In its 16 ft. pitch it would
be an important addition to the expressive forces of the Pedal
Organ.
FAGOTTO, Ital A lingual stop, of small scale, and 8 ft. pitch,
voiced to yield a tone imitative of that of the orchestral instrument
of the same name. See BASSOON.
FAGOTTONE, Ital. This term, which has the Italian aug-
mentative ending, has been used to designate a lingual stop of the
FAGOTTO family, properly of 32 ft. pitch, and of a full and impressive
voice. It is a Pedal Organ stop, and practically carries down the
voice of the CONTRAFAGOTTO, 16 FT., an octave lower. It has, how-
ever, no equivalent in the orchestra. An example exists in the. Organ
of the Church of San Alessandro, Milan. The name has also been
given to stops of 16 ft. pitch, the scales of which are larger and the
voices fuller than those of the imitative CONTRAFAGOTTO. An
example exists in the Swell of the principal Organ in the Cathedral
of Como.
FELDFLOTE, FELDPFEIFE, FELDPIPE, Ger. Lat., FIS-
TULA MILITARIS. An open labial stop, of a penetrating flute- or
fife-tone, hence its names. It has been made by German organ-
builders, according to Schneider (Organist of the Cathedral of
Merseburg, 1835), of metal and wood, and of 4 ft., 2 ft., and i ft.
* The beautiful stop made by Roosevelt for our Chamber Organ, and named by him Eu-
PHONE, had a voice which seemed to incline so much, in its higher compass, to the tone of the
Bass Saxophone that we labeled it CONTRA-SAXOPHONE, 16 FT.
122 ORGAN-STOPS
pitch. Owing to its very assertive voice, the stop seems never to
have been a favorite, and it does not appear in modern German
Organs. Organs of every class are better without high-pitched
screaming voices; and one can understand the disappearance of the
FELDFLOTES of 2 ft. and i ft. pitch.
FERNFLOTE, Ger. Literally Distant or Echo Flute. Prop-
erly applied, this name indicates a small-scaled labial stop of 8 ft.
and 4 ft. pitch, yielding the softest flute-tone produced by organ-
pipes; but the name has been applied to softly- voiced stops of differ-
ent classes, and both of wood and metal, open and covered. In the
Organ in the Church of St. Mary, Tyne Dock, constructed by Ed-
mund Schulze, the PERNFLOTE, 8 FT., is in the form of an extremely
soft SPITZFLOTE.
FERNHORN, Ger. Literally Distant or Echo Horn. The
name that has been given to a half-covered labial stop of metal and
of 8 ft. pitch. It is practically an ECHO NACHTHORN. The stop
would be suitable for the true Chamber Organ, and for a soft an-
cillary division of a properly appointed Concert-room Organ. It
should be voiced on wind of a pressure not exceeding 2% inches.
FIFE. Fr., FIFRE. Ger., PFEIFE. An open metal labial stop
of 2 ft. or i ft. pitch, yielding a shrill flute-tone, resembling that of
the military Fife, the compass of which extends about two octaves
from d 2 . As an organ-stop l the FIFE, 2 FT., differs from the PICCOLO,
2 FT., only in the character and loudness of its voice. The latter
stop is much to be preferred; accordingly the FIFE is seldom met
with. An example of the FIFRE, i FT., as made by the French build-
ers, and of comparatively soft tonality, exists in the Postif of the
Organ in the Cathedral of Abbeville. It is not a desirable stop for
the Organ of to-day.
FIFFARO, Ger. The name sometimes given by the old German
organ-builders to a stop, the voice of which imitated that of the or-
chestral Flute. Schneider thus briefly describes it: " FIFFARO,
QUEERPFEIFE, eine kleine Stimme von 4 oder 2 Fuss. " See QUER-
FLOTE.
FIFTEENTH, SUPER-OCTAVE. The names used to desig-
nate open, metal labial stops, yielding pure organ-tone, and sounding
two octaves above the unisons. Accordingly, in the aianual divi-
sions the FIFTEENTH is of 2 ft. pitch, and in the Pedal Organ of 4 ft.
pitch. The stop belongs to the foundation-work of the Organ, rep-
THEIR ARTISTIC REGISTRATION 123
resenting the third upper partial tone of the prime unison of the
department in which it is placed. Its scale is, accordingly and
properly, calculated from that of the foundation DIAPASON or PRIN-
CIPAL, 8 ft. or 1 6 ft., being usually made two or three pipes less in
diameter; the voicer taking care of the necessary reduction in the
strength of tone. The old English organ-builders, including Ber-
nard Smith, of Temple Church fame (1684), and John Harris, made
their FIFTEENTHS only one pipe less in scale than the corresponding
pipes of their DIAPASONS, simply voicing them softer than the in-
termediate OCTAVES, 4 FT. In modern Organs, as a rule, the FIF-
TEENTH, 2 FT., is made of too large a scale, and voiced much too
assertive in tone; builders, desiring to get as much sound as possible
out of the pipes, overlooking the scientific side of the matter, and
the fact that the true office of the stop is primarily a harmonic-cor-
roborating one, and that it should be treated as such in the tonal
appointment of the Organ. In proportionate strength, the tone of
the FIFTEENTH should be about 50% that yielded by the fundamen-
tal DIAPASON, and be clear and singing in quality.
TONE AND REGISTRATION. The tone of the FIFTEENTH, 2 FT.,
being of the pure organ class, as required by its primal office in the
tonal structure of the Organ, proves to be of considerable value in
registration, especially so when of the proper relative strength, as
recommended above. Its pure organ-tone practically free from
harmonics blends perfectly with every other tone produced by
organ-pipes, labial and lingual; imparting the elements of clearness
and brilliancy to the heavier and duller tones of many unison and
double stops, without sensibly impairing their characteristic voices.
This is surely a valuable property and one to be made much use of
in artistic registration. Of course, to be of maximum value, the
FIFTEENTH must be voiced to the proper strength of tone, as stated
above.
FIFTH, QUINT. Fr. and Ger., QUINTE. Ital, QUINTA.
The name properly applied to a stop the voice of which is at the in-
terval of a perfect fifth above the unison. In the Pedal Organ the
FIFTH is of 10% ft. pitch, and in the manual divisions it is of 5^ ft.
pitch. These stops, under different qualifications, may be of metal
or wood, and open, covered, or half-covered. See QUINT.
FLACHFLOTE, Ger. The name given to a labial stop of 8 ft.,
4 ft., or 2 ft. pitch, the pipes of which are of metal or wood, and
either conical or of the same measurement throughout, according
to the fancy of the builder and the quality of the tone desired.
124 ORGAN-STOPS
According to Wolfram the stop resembles the GEMSHORN and
SPITZFLOTE.* Seidel describes its tone as somewhat thin but not
disagreeable, f In his illustration, the FLACHFLOTE pipe is shown
cylindrical, terminating in a truncated cone having a small opening
at top, closely resembling the SPINDEFLOTE pipe. The mouth is
shown very narrow; but as the illustration is badly drawn this detail
cannot be depended on. Although the stop seems to have usually
been made of metal, there is one, of 4 ft. pitch, made of pear-tree
(Birmbaumholz) in the Organ of the Church of St. Boniface, at
Langensalza. It would seem probable, from the name, which means
Flat Flute, that the original FLACHFLOTE was made of wood, its pipes
being flat in form, with the mouth on a wider side. Under other
conditions the name would seem meaningless; and German organ-
builders generally had some sensible reasons for naming their stops.
It would be well if such a practice obtained here to-day; for some
names, having no relation to form or tone, which have appeared in
certain stop lists of late, if they were not objectionable in their ab-
surdity, would be only laughable.
FLAGEOLET. Ger. } FLAGEOLETT. Ital, FLAGEOLETTA.
Span., FLAUTIM. Lat., FISTULA MINIMA. A metal labial stop of
2 ft. and I ft. pitch, the pipes of which are cylindrical in form and of
a medium scale. The proper tone of the stop is liquid, clear, and
somewhat penetrating, in imitation of the voice of the old English
Flageolet, the instrument for which it is generally understood Han-
del wrote the obbligato in the song, "0 ruddier than the cherry"
("Acis and Galatea"). In some English Organs the FLAGEOLET,
2 FT., is a wood stop, of small scale, and a clear fluty voice. The
metal stop appears in several German and Dutch Organs in both
pitches. Of 2 ft. pitch, the FLAGEOLET exists in the Organs of the
Cathedral of Magdeburg, and the State Church, Triebel : and of I ft.
pitch in the Organs in the Marienkirche, Liibeck, and the Church of
St. Stephen, Nymengen. In the celebrated Haarlem Organ the stop
appears in the Echo of I }4 ft. pitch. A FLAGEOLET, 2 FT. , is inserted
in the Swell of the Roosevelt Organ in the Cathedral of the Incarna-
tion, Garden City, L. I. ; but it is rarely inserted in modern Organs.
*" FLACHFLSTE, eine offene sehr angenehme P16tenstimme 4 oder 8 Fuss mit kegelfdrmigen
K5rpern. Sie erhalt breite Labien rait engem Auf schnitt, und ist ubrigens dem GEMSHORN und
der SpmFLtfTE ahnlich." Wolfram.
t" FLACHFLOTE ist eine Labialstimme, deren Pf eif en nach Art des GEMSHORNS spitz zulauf en
Die Pfeif en haben breite Labien, einen weiten Aufschnitt und Seitenbarte, daher ist der Ton der
Stimme nichtvoll.sondern nach, ubrigens geradenicht unangenehm. Sie wirdzu 8, 4, 2, und i'
und zwar konisch und cylindrisch zugleich angetroffen, die 8- und 4 fussige heisst zuweilcn
GROSSFLACHFLQTE. "Seidel.
THEIR ARTISTIC REGISTRATION 125
TONE AND REGISTRATION. The position of the FLAGEOLET, in
the tonal economy of the Organ, is somewhat undecided; but as a
stop of 2 ft. pitch, having a pure flute-tone of less power than that
of the ORCHESTRAL PICCOLO, 2 FT., it would unquestionably prove a
valuable addition to an Organ, properly occupying a place in a
manual division to which the softer-toned accompaniment al stops
are apportioned. Voiced to imitate as closely as possible the tone
of the true Flageolet, it would become the best stop of super-octave
pitch for a solo or an accompanimental obbligato, just as the Flageo-
let was employed by Handel in ' ' Acis and Galatea. ' ' In registration,
it would be found more generally useful than either the FIFTEENTH
or the ORCHESTRAL PICCOLO; both of which are too assertive in any
combinations save those of full tonality.
FLAGEOLET HARMONIQUE, Fr. An open metal labial
stop, yielding a bright tone, closely resembling that of the true
Flageolet. Its pipes are cylindrical, of medium scale, and double
the standard speaking length; and have the perforation in their
bodies, as usual in harmonic pipes. See FLUTE HARMONIQUE. A
FLAGEOLET HARMONIQUE, 2 FT., appears in the Postif of the Grand
Organ in the Royal Church of Saint-Denis, constructed by MM.
Cavaill6-Coll, in 1841.
FLAUTADA, Span. The name applied by Spanish organ-
builders to open lingual stops yielding flute- and organ-tone. Thus :
FLAUTADA DE 13 signifies a FLUTE or PRINCIPAL, 8 FT.; FLAUTADA
DE 26 signifies a FLUTE or DOUBLE DIAPASON, 16 FT.; and FLAU-
TADA DE 52, the stop belonging to the Pedal Organ, is in all essen-
tials similar to the DOUBLE DIAPASON, 32 FT. Examples are to be
found in the Organs of the Cathedrals of Burgos, Seville, and Valla-
dolid. and also in other important Spanish instruments.
FLAUTINO, Ital. This word, terminating in the Italian dim-
inutive, mo, is employed to designate a flute-toned stop of small
size. In its best form, it is an open metal stop of 2 ft, pitch, formed
of cylindrical pipes of small scale, voiced to yield a flute-tone softer
than that of the proper FLAGEOLET, 2 FT. Examples exist in the
Organs in the Cathedral of Milan and the Church of San Gsetano,
Florence. In the Choir of the Organ in the Jesuit's Church, at
Cologne, is a FLAUTINO of 4 ft. pitch. In modern Organs the name
should be given only to the stop of 2 ft. pitch. The stop has been
frequently and incorrectly labeled FLAUTINA.
REGISTRATION. The FLAUTINO, 2 FT., when properly voiced is
126 ORGAN-STOPS
a very valuable stop; and being so closely allied to the FLAGEOLET,
of the same pitch, in tonality it can be used instead of that stop in
artistic registration. In this direction reference may be made to the
remarks under FLAGEOLET.
FLAUTO AMABILE, Ital. The suggestive name given to a
small-scaled labial stop, of either 8 ft. or 4 ft. pitch, properly con-
structed of open wo'od pipes, yielding an extremely delicate and
sweet unimitative flute-tone. In its best form, its pipes are slender,
about one and a half times their width in depth, and have inverted
mouths. The pipes should be made of clear spruce, fronted with
mahogany or white maple so as to allow their mouths being carefully
and accurately formed. Examples of the stop exist in the Organs
of the German Church, at Montreux, and the Church of Saint-
Martin, at Vevey.
Voiced on wind of low pressure, the FLAUTO AMABILE 4 FT., is an
ideal stop for the true Chamber Organ ; and of either 8 ft. or 4 ft.
pitch it is suitable for insertion in the softest division of a Church or
Concert-room Organ.
FLAUTO AMOROSO, Ital. This name has been employed to
designate an open metal stop of small scale and 4 ft. pitch, voiced
to yield an extremely soft and singing quality of unimitative flute-
tone. A fine FLAUTO AMOROSO, 4 FT., exists in the Echo of the im-
portant Organ in the Church of SS. Peter and Paul, Liegnitz, in
Silesia, built by Buckow, in the year 1839. Stops of this refined class
are held in little esteem by organ-builders of to-day in their craze
for high-pressures and loud voicing. When will the musical world
protest against such crude and inartistic treatment of the Monarch
of all Instruments ?
FLAUTO D'AMORE, Ital. Pr., FLUTE D'AMOUR. The name
given to special stops formed of small-scaled wood pipes, yielding a
peculiarly delicate and fascinating flute-tone. The stop is generally
of 4 ft. pitch, which is that most desirable in solo passages and regis-
tration; but it has been made of 8 ft. pitch. An example, labeled
FLUTE D'AMOUR, 4 FT., exists in the Choir of the small Organ in the
Marienkirche, at Goerlitz, built by Buckow, in the year 1838. It is
the only stop of 4 ft. pitch in the Choir formed entirely of soft-
voiced stops.
FORMATION. The pipes of the FLAUTO D'AMORE, 4 FT., have been made of
different forms and proportions, according to the fancy of the organ-builder,
resulting, of necessity, in slight differences in tonality. The most satisfactory
THEIR ARTISTIC REGISTRATION
127
n
form, in our estimation, is that here briefly described. The pipes partake of the
character of those of the LIEBLICHGEDECKT and the ROHEFLOTE ; being small in
scale and in general formation like those of the former; and having perforated
stoppers like those of the latter stop. The tone of the FLAUTO D'AMORE is, accord-
ingly, a combination of the voices of the two stops just named, but properly softer
than either. The construction of the FLAUTO D'AMORE pipe is clearly shown in
the accompanying illustration, Fig. 16. In the Longitudinal Section, i, the for-
mation of all portions of the mouth is correctly delineated ; and the manner in
which the stopper, A, is formed and perforat-
ed is shown. It will be observed that the
vertical perforation (Fr. cheminee) does not
extend through the entire length of the stop-
per, but opens into the larger transverse
perforation , B . This is an important exped-
ient. As the perforations of the stoppers af-
fect the quality of the tone of the pipes, their
lengths must, of necessity, be graduated reg-
ularly throughout the compass of the stop :
and as this graduation of length would be
difficult, and would make the stoppers in
the higher octaves inconveniently short, if
the perforations were carried through them
and dictated their lengths, the expedient
of the transverse perforation was happily
adopted, rendering correct graduation possi-
ble without interfering with the convenient
length of the stoppers. The transverse per-
forations would not be required in the stop-
pers of the lower pipes, for they could be per-
forated throughout and made of the proper
graduated lengths. The diameters of the
vertical perforations may vary according to
the scale of the pipes and the quality of the
tone desired : that of the largest pipe need not
exceed % inch, while that of the smallest pipe
may be y% inch. The Front View, 2, shows
the desirable proportions of the mouth ; but its height may vary according to the
wind-pressure used and the quality of tone desired. For its most characteristic
tones, the FLAUTO D'AMORE should be voiced on wind of from i l A to 2^2 inches.
FIG. 16
TONE AND REGISTRATION. The tone of the FLAUTO D'AMORE,
when correctly made and voiced by a master, is the most beautiful
of those produced by the half-covered stops; occupying a place in
the tonal economy of the Organ that neither the FLUTE A CHEMINEE
nor the ROHRFLOTE can fill. Its extremely delicate and sympathetic
voice fails to recommend it during the present craze for loud and
screaming stops: its appearance will, accordingly, be rare in Organs
in this country never, probably, voiced on the proper low wind-
128 ORGAN-STOPS
pressures.* Its. proper place is in the accompanimental division of
the Church Organ and in the softest-toned division of the Concert-
room Organ, where it will be invaluable, as an octave voice, in re-
fined and artistic registration. It imparts a beautiful tonality, as
we have proved, to the Vox HUMANA, especially for a soprano solo.
FLAUTO DI PAN, Ital. The name given to an organ-stop,
probably on account of the similarity of its tone to the sounds of the
time-honored Pandean Pipes one of the precursors of the Organ.
It has been made of 2 ft. and i ft. pitch, but is very uncommon m
either, having been found, in all probability, to have little to re-
commend it tonally. As a stop of 2 ft. pitch it appears in the Ober-
werk, and as one of I ft. pitch in the Pedal of the Organ in the
Cathedral of Lund, Sweden. It is likely to be classed among ob-
solete stops.
FLAUTO DOLCE, Ital. Fr., FLUTE DOUCE. Ger., DOLZFLOTE,
DULZFLOTE. An open labial stop of small scale, the pipes of which
are of metal or wood, and of 8 ft. or 4 ft. pitch. As the name implies,
the tone of the stop is soft, and when at its best it resembles that of
the orchestral Flute in piano passages. Regnier describes it thus:
<c FLAUTO DOLCE, la Flftte douce, Dulzfl&te, que les facteurs de
France ont eu tort de confondre avec la flute allemande beaucoup
plus fine et vive, est un registre egalement ouvert et de tres-menue
taille, mais d'un ton plus rond et eminemment doux."
The FLAUTO DOLCE, in either pitch, is suitable from every point
of view for insertion in the Chamber Organ; or in the Choir or Echo
of larger Organs when extreme delicacy and refinement of intona-
tion is desired. An example of 8 ft. pitch, of wood, exists in the
Choir of the Organ in the Marienkirche, at Goerlitz.f A beautiful
example of 4 ft. pitch, of metal, exists in the Echo of the Organ in
the Parish Church of Leeds, where it speaks on the appropriate wind
* We made one of fine cedar for our own Organ, voiced on wind of 2 % inches, and its tone
was very delicate and beautiful.
f The Organ in the Marienkirche, at Gorlitz, in Silesia, constructed by the celebrated organ-
builder,Buckow, in 1838. It is a small instrument, containing only fifteen speaking stops, three
of which are given to the Pedal Organ. It is, however, to the stop-apportionment of the Choir
Organ that we desire to direct the attention of organists and others interested in refined organ-
building. It comprises the following four stops: VIOLA DA GAMBA, 8 FT., metal; DOLCIANO,
8 FT., wood; FU)TE DOUCE, 8 FT., wood; and FL{)TE D'AMOUR, 4 FT., wood. It requires little
imagination on the part of anyone, with some knowledge of organ-stops, to realize the beauty
and charm which must characterize the tones of this little Choir: and it, as the work of a dis-
tinguished artist, certainly gives a lesson to twentieth-century lovers of musical noise. But it
seemshopeless to look for any immediate improvement so long as organists continue to approve
of, or remain indifferent toward, the inartistic and tasteless doings in the great majority of
voicing-rooms to-day.
THEIR ARTISTIC REGISTRATION 129
pressure of i J^ inches. This fine stop was made by Edmund Schulze
FLATJTO MAGGIORE, Ital. Ger., MAJORFLOTE. The term
appropriately employed to distinguish the most important or dom-
inating flute-toned stop in the Organ the stop holding the same
commanding position in the flute-work as the PRINCIPAL or DIA-
PASON holds in the foundation-work of the instrument. The FLAUTO
MAGGIORE is properly an open wood and metal stop of 8 ft. pitch
and large scale, voiced to yield a rich unimitative flute-tone of con-
siderable volume. We find the principal flute-toned stop, of 8 ft.
pitch, in the Great of the Organ in the Cathedral of Breslau, labeled
MAJORFLOTE : and the same name is given to a corresponding stop
in the Choir Organ. The term has also been applied to a full-toned
covered stop of 8 ft. pitch, in the Great of the Grand Organ in the
Church of St. Elizabeth, at Breslau. There is much to recommend
the name for general adoption, when confined to a unison stop of
dominating unimitative flute-tone, for it clearly denotes its position
in a tonal scheme.
FLAUTO MINORE, Ital. The name appropriately employed
to designate an unimitative flute-toned stop of secondary impor-
tance as regards strength of tone, placed in any division of the Organ
in which a more powerful flute-toned stop, labeled FLAUTO MAG-
GIORE (q. v.), is inserted. The German equivalent MINORFLOTE
has been adopted to designate a flute-toned stop of 4 ft. pitch, as in
the Organ in the choir of the Cathedral of Breslau. In this latter
sense the term is undesirable; for the terms FLAUTO MAGGIORE and
FLAUTO MINORE should refer to tone, the only matter of impor-
tance to the organist, not to the sizes of the stops only, which call
for other distinguishing names.
FLAUTONE, Ital. Fr., FLUTON. The name given by the
Italian organ-builders to a large-scaled, flute-toned stop, yielding a
smooth and somewhat subdued voice. The pipes of this stop are
either open or covered, according to the ideas of the builder, or,
probably, according to the stop-apportionment of the divisional
Organ in which it is placed. The FLAUTONE in the North Organ in
the Cathedral of Milan is of metal with a covered wood bass; while
that in the Choir of the Organ in the Church of San Gaetano,
Florence, is a metal stop of 8 ft. pitch, the pipes of which are appar-
ently open.*
* "A la pedale, les Italiens 1'emploient sous le nom de Flaulone, Fluton, ou grosse flute douce:
Celle de huit-pieds, qui est la plus usit6e, chante admirablement les adagios en solo ou en choeur,
130 ORGAN-STOPS
PLAUTO TEDESCO, Ital. The name given by the old Italian
organ-builders to an open wood stop, of medium scale, and 8 ft.
pitch. The stop resembles, both in the form and tone of its pipes,
the English CLARABELLA. Examples are to be found in several
Italian Organs; as in the North Organ in the Cathedral of Milan;
and in the Organ in the Church of Santa Maria des Vignes, Genoa.
It also existed in the old Organ, by Hermann, in the Church of
Santa Maria di Carignano, Genoa; but was removed when the
present Organ was constructed by Bianchi in 1863.*
PLAUTO TRAVERSO, Ital. Pr., FLUTE TRAVERSIERE. Ger.,
TRAVERSFLOTE, QUERFLOTE. The names used by different organ-
builders to designate the stop of 8 ft. and 4 ft. pitch, the pipes of
which are properly of wood and harmonic, yielding a tone closely
imitating that of the orchestral Plute. Ever since the invention of
harmonic pipes, this important stop has deservedly become a great
favorite, and many fine examples are to be found in well-appointed
Organs, notably in those of the great German builders; as in the
Ladegast Organ in Schwerin Cathedral. The stop is made of pine
and pear-tree, and is thus described by J. Massmann, Grossherzogl,
Musikdirector in Wismar :
"FLAUTO TRAVERSO 8', off en aus Tannen- und Birnbaum-holz;
hat einen weichen und schonen Ton, der Querflote des Orchesters
sehr ahnlich; die Pfeifen der hochsten Octaven tiberblasend." See
ORCHESTRAL FLUTE.
FLOTENBASS, Ger. An open labial stop of 8 ft. pitch, the
pipes <3t whicri are of wood, of large scale, and yield a powerful un-
imitative flute-tone. It is properly a Pedal Organ stop, appearing
in 1 that department of the Organs in the Gewandhaus, at Leipzig,
and the Domkirche, Lubeck. As a covered stop, an example ap-
pears in the Piano Pedal of the Organ in the Cathedral of Schwerin.
As a stop of 1 6 ft. pitch, examples exist in the Walcker Organs in the
Cathedrals of Riga and Vienna. The stop is the same as the BASS
FLUTE (q. v.) of English and American Organs.
FLOTENPRINCIPAL, Ger. The name given by Walcker to
the principal unison and the attendant octave stop placed on the
Fourth Manual (Echo) of the Organ in the Cathedral of Riga. The
mais seule a la main; 11 ne faut pourtant pas trop augmenter le nombre des parties, la confusion
s'y jetterait, et adieu 1'effet instrumental. " Regnier.
* In adopting a consistent Italian stop nomenclature for our own Chamber Organ, we applied
the name PLAUTO TEDESCO to what was practically a soft-voiced English CLARABELLA, 8 FT.
See "The Organ of the Twentieth Century," pp. 334~S-
THEIR ARTISTIC REGISTRATION 131
stops are of open pipes, yielding flute organ-tone, and are valuable
in imparting a distinctive tonality to the division, which is strongly
flute-toned throughout. Of its seventeen stops, nine are flute-toned
and the most important labial stops in the division. The two stops
specially alluded to are labeled respectively FLOTENPRINCIPAL, 8 FT.,
and FLOTENPRINCIPAL, 4 FT.
FL0TE A EEC, Fr. The name that has been given to a flute-
toned stop of ordinary form the tone of which was supposed to imi-
tate that of the old Flute a Bee. Engel, speaking of this instrument,
says: "The most common FMte a bee was made with six finger
holes, and its compass embraced somewhat more than two octaves.
. . . There was often a key on this instrument in addition to the
finger-holes. This flute was much in favor in England; hence it was
called in France ' Flute d' Angleterre. ' The flageolet, the smallest
fiUte a bee, was formerly played in England even by ladies." The
instrument was also called the Whistle Flute, and was played by a
small ivory mouthpiece or beak, hence its French name. The Ital-
ians appear to have called this instrument Flauto a Becco or Flauto
Suabile.
FLtTE A CHEMINfiE, Fr. Eng., CHIMNEY FLUTE. Ger.,
ROHRFLOTE. The stop in its proper form, as constructed by the
French organ-builders, is of metal and of large scale; its pipes being
furnished with sliding caps, from the centers of which rise small
tubes, or so-called chimneys, which give the stop its peculiar name.*
The FLUTE A CHEMINEE appears in certain French Organs of both
8 ft. and 4 ft. pitch, although it is properly a unison stop. There can
be no doubt, however, regarding the value of the octave voice in
effective registration. An example of 4 ft. pitch exists in the Choir
of the Organ in the Cathedral of Abbeville. For particulars respect-
ing the formation of the pipes of the stop, see ROHRFLOTE.
* La FLt)TE X CHEMINEE tient le milieu entre les jeux bouch6s et les flutes ouvertes. Si elles
e"taient tout a fait bouche" es, il ne leur faudrait que la demi-hauteur des tuyaux ouverts; mais
ouvertes en partie, elles doivent avoir environ les deux tiers de hauteur des fonds ouverts de
mme ton. Dans cette hauteur des Rohrflozten, on comprend celle de la chemine*e posSe au centre
de la calotte. Nous 1'avons eatplique" : une partie du son se propage par cette ouverture, 1'autre
par la bouche, apres s'e'tre heurte"e k la partie couverte du sommet. Le son qui sort par la chem-
in6e demande la m6me 616vation de tuyau que dans les jeux ouverts; et il faudrait bien la lui
dormer si la partie calotte"e du sommet ou viennent se heurteur les vibrations, pour ressorth ^n
partie par les levres du tuyau, ne ramenaient pour leur part ce tuyau & l'6tat de bourdon, et n'en
diminuaient par consequent la hauteur. Pour donner plus de facilit6 au son de se propager par la
bouche, on ouvre celle-ci davantage, on tgueule* comme on dit, les flutes k chemine'e plus queles
flutes ouvertes et que les simple^ bourdons. II va sans dire que plus on les veut sonores, plus
on 61argit la taille, qu'en ge"ne"ral on tient tres-grosse. Ces flutes se font detain et i, barbes de
chaque cdte" de la bouche, pour en diriger 1'intonation et 1'accord. " Regnier.
132 ORGAN-STOPS
FLtJTE A FUSEAU, Fr. A metal labial stop, of 8 ft. and 4 ft.
pitch, yielding an unimitative flute-tone, the pipes of which are
tapered or spindle-shaped, hence its name. An example exists on the
Fourth Manual of the Organ in the Church of Saint-Nicholas, at
Blois. The FLUTE A FUSEAU is practically the same as the German
SPINDELFLOTE (q. v.).
FLtfTE A PAVILLON, Fr. The name given by the French
organ-builders to a full-scaled metal labial stop, of 8 ft. pitch, the
pipes of which have cylindrical bodies surmounted with bells of in-
verted conical form. Hence the distinctive and appropriate name.
It is to be regretted that the labor attending the construction of its
pipes, combined with the difficulty of finding proper standing room
for them on wind-chests of ordinary dimensions, has seriously inter-
fered with a desirable introduction of this fine stop in important
Organs. The FLUTE A PAVILLON was first brought before the notice
of the English organ-builders by Ducroquet, of Paris, in the Organ
he sent to the London Exhibition of 1851. It was received with
great favor, and several fine examples were made by different build-
ers and pipe-makers, especially by John Courcelle, the celebrated
reed voicer, and teacher of George Willis, of equal fame.* So fine
were his stops, that in one Organ the name COURCELLINA was given
to the stop. An example, under this name, exists in the Organ of
the Church of St. John, Portsea. One, under the proper French
name, exists in the Back Great of the Concert Organ in the Town
Hall, Leeds, Yorkshire; and one, under the English name, BELL
DIAPASON, in the Great of the Organ in the Oratory, Brompton,
London. Notable French examples exist in the Grand Orgue of the
Organ in the Church of Saint-Sulpice; and in the same division of
the Organ in the Church of Saint-Eustache, Paris. Both in France
and England the stop appears to have fallen into disuetude. We
introduced a FLUTE A PAVILLON, 8 FT., in the scheme of the Fourth
Organ (Solo) of the Concert-room Organ installed in the Festival
Hall of the Louisiana Purchase Exposition, 1904.
FORMATION. The FL^TE A PAVILLON pipe comprises two leading portions, a
cylindrical body, and a bell (pavilion) in the shape of an inverted truncated cone,
attached to a short cylindrical portion which slides on the open end of the body
for the purpose of tuning, as neither coning nor slotting is admissible for that
operation. The cylindrical body is formed in all respects similar to that of the
PRINCIPAL or DIAPASON, 8 FT., as shown in Fig. 17, Plate II. The short cylindrical
* In our own Chamber 1 Organ the TROMBA and CLARINETTO were by Courcelle and the OBOE
by Willis.
THEIR ARTISTIC REGISTRATION 133
portion, to which the bell is attached, fits closely to the body , its lower edge being
diagonally cut, so as to have a screwlike motion against a small projecting button
of solder on the body, as indicated. Accordingly, in tuning the pipe it is only
necessary to raise or lower the bell by turning it slightly of the right or left. By
such a simple method of tuning, the timbre of the pipe will remain constant, and
the bell will remain firmly in position : both important matters. The mouth of the
FLUTE A PA VILLON is of the same width as that of a DIAPASON pipe of the same
scale, but is cut higher; the height depending largely on the pressure of the wind
used and on the volume of tone desired. The pipes should, in all cases, be copi-
ously winded and high pressures avoided. It is questionable if any of the French
or English stops were voiced on wind of higher pressure than 3 % inches. The
proportions of the bell vary according to the quality of the tone required, but its
desirable dimensions are one and one-half the diameter of the pipe in height, and
one and two-thirds the diameter of the pipe in its diameter at top, as shown in
Fig. 17. The proportions have been modified by different pipe makers.
TONE AND REGISTRATION. The stop has been labeled by certain
English organ-builders BELL DIAPASON, as by Bishop in the Organ
of the Oratory, Brompton; but the term is misleading, for the tone
of the stop is not the pure organ-tone of the true DIAPASON. The
stop belongs to the flute-work of the Organ, and this fact the French
organ-builders recognized when they gave the original name to it.
The proper tone of the stop is a combination of pure organ-tone and
unimitative flute-tone, characterized by a singular richness and
fullness, yet not unduly assertive. It inclines, in certain fine ex-
amples, to a horn timbre, which can be intensified, if desired, by
slotting or perforating the bell after the fashion of the KERAULO-
PHONE tuning-slide. There can be no question as to the tonal value
of the FLUTE A PAVILLON in full registration, in which important
lingual stops take prominent part, and call for effective backgrounds
of labial tones. Accordingly, while the stop finds a proper place in
the Great or Grand Orgue of French and English instruments, its
rich compound tone makes it a valuable, and, perhaps, a more valu-
able, voice in another division of the Organ richer in lingual stops.
It was that consideration led us to insert it in the Fourth Organ of
the instrument installed at the Louisiana Purchase Exposition,
which contained seven very important lingual stops. If the science
and art of tonal combination and artistic registration were carefully
studied by organ-builders, and better understood by organists than
they seem to be in many quarters to-day, a better and more rational
system of stop-apportionment would soon appear in new Organs.
FL0TE A PYRAMIDS, Fr. The term used to designate an
open labial stop of 8 ft. and 4 ft. pitch, the pipes of which are square
and inverted pyramidal in form, resembling those of the DOLCAN in
134 ORGAN-STOPS
being larger at the top than at the mouth-line, and especially in the
pipes of the bass octave, which have been made of wood (See DOL-
CAN). The name would be more appropriately applied to a stop
the pipes of which are directly pyramidal; that is, square, and larger
at the mouth-line than at the top, in this respect resembling the
pipes of the SPITZFLOTE, GEMSHORN, and CONE GAMBA. We have
made wood pipes of this form, having extremely low inverted
mouths, which, voiced on wind of 2^ inches, yielded exquisite tones
having a rare combination of imitative flute- and string-tones, both
of which could be distinctly heard as if fighting for supremacy.
There is much to be done in this direction that has not been dreamt
of : but the trouble and skill involved in the formation of such pyra-
midal pipes will effectively condemn them in the organ-building
world.
FLtTE CONIQUE, Fr. The name given by Cavaille-Coll to
open stops, of 16 ft. pitch, inserted by him in the Grand-Orgue and
Solo of the Organ in the Church of Saint- Sulpice, Paris. The stops
are flute-toned, and are practically DOUBLE SPITZFLOTES.
FLtTTE CREUSE, Fr. The term that has been used by
French organ-builders to designate a stop, of full scale, yielding a
powerful unimitative flute-tone: it is in all essentials, as the name
implies, similar to the German HOHLFLOTE (q. v.).
FLftTE DOUCE, Fr. An open labial stop, of small scale,
yielding a soft and pure flute-tone, approaching an imitative quality.
An example exists in the Positif of the Organ in the Church of Saint-
Sulpice, Paris. Another example is to be found in the same division
of the Organ in the Cathedral of Notre-Dame. Both stops are of 4
ft. pitch; evidently the one preferred by the distinguished builder,
Cavaill<-Coll, although he introduced one, of 8 ft. pitch, in the
Positif of the Organ in the Madeleine, Paris. See FLAUTO DOLCE.
FLCTE HARMONIQUE, Fr. Eng., HARMONIC FLUTE.
The principle of formation which obtains in the FLUTE HARMONIQUE
was long known as applied to wood pipes, the tones of which were
more or less imitative of those of the orchestral Flute. The first
metal stops to which the principle was applied were constructed by
MM. Cavaill6-Coll, of Paris, and inserted in the Organ erected by
them, in the year 1841, in the Royal Church of Saint-Denis, near
Paris. The series of HARMONIC FLUTES in the Organ is so note-
worthy that the stops which belong to it deserve enumeration here.
In the Grand Orgue FLUTE TRAVERSIERE HARMONIQUE, 8 FT., and
THEIR ARTISTIC REGISTRATION 135
FLUTE OCTAVIANTE HARMONIQUE, 4 FT. In the Positif FLUTE
HARMONIQUE, 8 FT., and FLAGEOLET HARMONIQUE, 2 FT. In the
Recit-ficho Expressif FLTJTE HARMONIQUE, SFT., FLUTE OCTAVI-
ANTE HARMONIQUE, 4 FT., and FLUTE OCTAVIN HARMONIQUE, 2 FT.
The stop derives its name from the fact that the pipes which form
the larger portion of its compass are so formed and voiced as to yield
their first harmonic upper partial tones instead of the tones which
normally belong to their full lengths; the tones so produced having
a distinctive timbre of special tonal value, which has secured the
stop universal adoption.
FORMATION. The FLUTE HARMONIQUE, properly so-called, is formed of open
cylindrical pipes of metal, or straight quadrangular pipes of wood, of large scale.
From about the note f r the pipes are made of twice the standard speaking length.
In the HARMONIC FLUTE, 4 FT., the harmonic pipes are carried to about G; and the
HARMONIC PICCOLO, 2 FT., should be harmonic throughout. Indeed, it is desirable
in the unison stop to commence the harmonic pipes on middle c 1 . Open pipes of
wood or metal of the normal speaking lengths are used for the lower portions of
the stops of 8 ft. and 4 ft. pitch ; of these, which present no unusual treatment, it is
unnecessary to speak here. The FLUTE HARMONIQUE, 8 FT., we select for descrip-
tion and illustration is that in the Great of the Concert-room Organ by Cavaille*-
Coll in the Town Hall of Manchester, a stop we had the opportunity of examining
and measuring. The lowest harmonic pipe of this stop is g 1 , as shown in correct
proportions in Fig. 18, Plate II. The pipe is 2.37 inches in diameter and 29^
inches in effective length from the mouth line. At a distance of 13 inches from the
lower Hp of the mouth, a hole y% inch in diameter is pierced, as indicated, opposite
A, in the illustration. The mouth is i% inches in width and H inch in height,
having a straight upper lip, and the languid closely and finely nicked. Through
the agency of the small perforation in the body, which prevents the formation of a
node in the middle of the internal column of air, and by the pipe being slightly
overblown, a note is produced which is about an octave of that which normally
belongs to a pipe of the length of 29 % inches. The bass and tenor of this stop are
of open wood pipes the CC pipe having the scale of 5.00 inches in width and 6.50
inches in depth. The metal non-harmonic pipe c 1 , has a diameter of 2. 75 inches,
and a speaking length of 22 inches, belonging to a very large scale. Certain organ-
builders prefer, and, we think, wisely, to form the HARMONIC FLUTE, 8 FT., entirely
of metal, commencing the double-length harmonic pipes at f *.* For particulars
respecting the formation of harmonic pipes of wood, see ORCHESTRAL FLUTE.
TONE AND REGISTRATION. The tones produced by the har-
monic pipes of the FLUTE HARMONIQUE differ considerably from
those yielded by open pipes of the standard speaking lengths: this is
due to the large scale allowed by the perforated double-length pipes,
which would be excessive and undesirable in pipes of half their
length, yielding notes of similar pitch under ordinary conditions;
* For further details respecting formation of harmonic pipes, see "The Organ of the Twen-
tieth Century," page 388.
136 ORGAN-STOPS
and also to the prominence of certain upper partial tones, generated
by the necessary and special voicing and overblowing of the pipes,
which impart a peculiar and characteristic timbre to the compound
sounds they produce. Although the tones of the metal FLUTE
HARMONIQUE, as originally and usually formed and voiced, are not
strictly imitative they differ considerably from those of the ordinary
stops which belong to the flute organ-tone class : they are clearer,
more penetrating, and to some extent more valuable in combina-
tions and registration of an assertive character. In artistic registra-
tion with all classes of lingual stops, except the Vox HUMANA, the
FLUTE HARMONIQUE, 8 FT., fulfills its highest office; and, accordingly,
should find a place in the division of the Organ which is richest in
lingual stops. This is especially desirable in a properly stop-appor-
tioned Concert-room Organ, in which provision for the greatest
range of tonal contrasts by means of effective and convenient regis-
tration should be made.*
^ OCTAVIANTE, Fr. The term used by French organ-
builders to designate an open flute-toned stop, of either metal or
wood, and of octave, or 4 ft. pitch. This stop in its most effective
form is constructed of harmonic pipes from tenor F, or better still
from tenor C, the lower pipes being of large scale and standard
speaking lengths. In this form it is called FLIJTE OCTAVIANTE HAR-
MONIQUE ; the first example of which was inserted by MM. Cavaille-
Coll in the Organ erected by them in the Royal Church of Saint-
Denis. See FLUTE HARMONIQUE.
FLOTE OUVERTE, Fr. The general name given by French
organ-builders to large-scaled open stops, formed of metal or wood,
and yielding an indeterminate flute-tone: hence its convenience can
be realized. The FLUTE OUVERTE is of 32 ft., 16 ft., 8 ft., and 4 ft.
pitch, as shown by the tonal appointment of the Pedal Organ of the
Organ in the Royal Church of Saint-Denis, in which examples of all
these four pitches are to be found. In English nomenclature, these
four Pedal Organ stops would be labeled DOUBLE DIAPASON, 32 FT.,
DIAPASON, 16 FT., OCTAVE or FLUTE, 8 FT., and SUPER-OCTAVE^
4 FT.
FLtTE POINTUE, FLtTE A POINTE, Fr. The name given
to an open metal stop of 8 ft. and 4 ft. pitch, the pipes of which are
conical in form, resembling those of the SPITZFLOTE. A flute-toned
* See the tonal apportionment of the Fourth Organ, in Chapter XL of "The Organ of the
Twentieth Century, " page 311.
THEIR ARTISTIC REGISTRATION 137
stop of 4 ft. pitch bearing this uncommon name exists in the Grand
Orgue of the instrument built by Clerinex for the Church of Saint-
Martin, at Liege.
FLUTTUAN, Get. The uncommon name given to a short com-
pass manual stop of 16 ft. pitch, the pipes of which are of open
wood. An example, having the compass from middle c 1 to g 3 , con-
structed of pear-tree, exists on the Second Clavier of the Organ in the
Church of Neu-Ruppin. Seidel describes the stop and its tonality
thus:
<c FLUTTUAN 16' steht in der Orgel zu Neu-Ruppin im Mittel-
klavier von c 1 bis g 3 und hat sehr schwache Bretter von Birnbaum-
holz. Diese Stimme ist sehr weit mensuriert, hat engen Aufschnitt,
starke Intonation und einen hornartigen Klang; die Bassoktaven
sind durch Quintaton erganzt."
FOURNITURE, Fr. Eng., FURNITURE. The name given to a
compound harmonic-corroborating stop composed of three or more
ranks of open metal pipes of medium scale, the ranks usually being
alternately octave- and fifth-sounding. With the aim of obtaining
greater assertiveness, a third-sounding rank has sometimes been
added; but this rank will not be called for if there is a SESQUIALTERA
on the same manual. The general pitch of the FOURNITURE is,
properly, higher than that of any other compound stop introduced
in the Organ ; and as the pipes composing the stop are of necessity
very small, the ranks have to break at every octave. When prop-
erly voiced and scientifically graduated in strength of tone, the
stop adds great richness and brilliancy to all combinations in which
it enters. This fact has been pointedly commented on by Regnier.*
The following examples of FOURNITURES of four and five ranks may
be accepted as representative compositions suitable for insertion in
important Organs. The first, of four ranks, is composed of octave-
* "La FOURNITURE, ou mixture, est un jeu multiple, ordinairement quadruple, de menue
taille, fait de 1'etain le plus fin et le plus doux. Le plus grand tuyau de la Fourniture mesure sa
hauteur sur celle des tuyaux rang6s sur le mSme sommier et correspondant au mSme clavier.
" Ainsi le premier ut d'une Fourniture plac6 au grand orgue d'un seize-pieds y a quatre-pieds,
et deux au positif . L 'harmonic de la Fourniture est brillante, fine, et destinee a des roulements
rapides. On peut dire que son effet sur 1'oreille est comparable en harmonie a celui que ferait
sur la vue une parure d'acier poli et talle en diamants. On lui donne toute 1'etendue du clavier,
'
ut^oa b |juu.i i cj-n cu.ia.Lc;, viuj.ii.uie nuuo <aij<Jiis j.c vuj.ii 10. ^vjiiipus>i uiun u.c let yj.ciij.icic IIVLC. i MI> jjuuo.
la Fourniture a trois rangs est celle-ci: wt 1 , 50/ 1 , wf 2 : on ajoute le sol 2 , si la Fourniture a quatre
tuyaux par note : I'wf3, si elle en a cinq; le sols, si elle en a six; enfin l'w4, si elle en a sept, comme
nos plus grandes orgues. Au positif, on prend la Fourniture & 1'octave superieure. Si le positif
n'est qu'un quatre-pieds, on prend la Fourniture a la super octave. " Regnier.
138 ORGAN-STOPS
and fifth-sounding ranks : the second, of five ranks, is composed of
octave-, third-, and fifth-sounding ranks :
FOURNITURE IV. RANKS.
CC to BB 22 26 29 33.
C to B 19 22 26 29.
c r to b 1 15 19 22 26.
C 2 to b 3 12 15 19 22.
C^ tO C4 8 12 15 19.
FOURNITURE V. RANKS.
CC to BB
C to B ....
c 1 to b 1 ....
c 2 to 'b 2 ....
c 3 to c 4
As all the intervals in these examples belong to the 8 ft. harmonic
series, both stops are suitable for insertion in any manual division of
the Organ in which extreme brightness of harmonic structure is
desired: relative strength of tone and perfect regulation being ob-
served in all cases.
The FOURNITURE was introduced by the old English builders in
several of their Organs, commonly of few ranks. Renatus Harris,
in 1670, introduced one, of three ranks, in the Great of his Organ
in the Church of St. Sepulchre, London, John Avery, in 1794, in-
serted one, of three ranks, in the Choir of his Organ in Croydon
Church; and one of two ranks, in the Choir of his Organ in St.
Margaret's Church, Westminster. Although, for obvious reasons,
the stop has not been a favorite with later builders, Willis inserted
FOURNITURES, of five ranks, in the Pedal Organ and the Swell of his
instrument in St. George's Hall, Liverpool; and Hill has inserted
FOURNITURES, of five ranks, in the Great and Swell of the Organ in
the Centennial Hall, Sydney, N. S. W.
The FOURNITURE appears in numerous Organs by French build-
ers. Confining our remarks to instruments constructed by France's
greatest organ-builder, Aristide Cavaille-Coll, we find a GROSSE
FOURNITURE, of four ranks, in the Grand-Choeur of the Organ in
the Church of Saint-Sulpice, Paris; a FOURNITURE, of five ranks, in
the Grand-Orgue of the Organ in the Church of Saint-Sernin, Tou-
louse; a FOURNITURE, of five ranks, in the Bombarde of the Organ
in the Church of Saint-Ouen, Rouen; and a FOURNITURE, of five
ranks, in the Grand-Orgue of the Concert-room Organ erected by
him in the Albert Hall, Sheffield, England.
THEIR ARTISTIC REGISTRATION 139
FRENCH HORN. The name commonly given to a lingual stop
of 8 ft. pitch, formed and voiced to yield tones imitating as closely
as practicable those of the orchestral Horn. It is undesirable to
continue this full name; for, as Dr. W. H. Stone points out, "The
designation 'French' is commonly added to the name of the orches-
tral Horn, from the fact that a circular instrument of this nature,
without crooks or other appliances, was, and still is, used in France
for hunting. Its tones are, "coarse and boisterous, only fit for the
open air and for woodland pastimes." For description of the organ-
stop see HORN.
FUGARA. The name that has been used to designate an open
stop of metal or wood, commonly of 8 ft. and 4 ft. pitch, the tone of
which is somewhat indeterminate in character; in some examples
inclining to a cutting string quality, and in others to a combination
of string and horn tones. Locher says: "The FUGARA has much in
common with the GAME A, while in quality of tone it stands between
it and the GEIGENPRINCIPAL."* The stop appears to have been
held in high favor by German organ-builders of the later school.
Examples, of 8 ft. pitch, exist in the Organs in the Cathedrals of
Ulm, Vienna, Lubeck, Merseburg, and Schwerin; and, of 4ft. pitch,
in the Organs in the Cathedrals of Riga and Schwerin. The FUGARA
does not seem to have been valued by the French organ-builders,
very seldom appearing in their Organs. One, of 4 ft. pitch, is in-
troduced in the Positif of Merklin's Organ in the Church of Saint-
Sulpice, Paris.
FULLFLOTE, Ger. The name given by German organ-build-
ers to a covered metal stop of large scale, yielding a powerful un-
imitative flute-tone. It is described by Seidel thus: "FULLFLOTE,
4 Fusston gedeckt, aus 10 lot. Metall, steht im Hauptwerk der von
Buckow in Jahren i8f-J in der Stadtpfarrkirche zu Triebel in der
Niederlausitz erbauten Orgel von 25 klangbaren Stimmen."
FULL MIXTURE. A compound harmonic-corroborating stop,
formed of full-scaled open metal pipes, yielding pure organ-tone of
* " La PUGARA est une varie"te de la longue et incisive famille des registres e~troits et f ermes de
metal. Elle a plus de clarte et non moms de mordant que la Viole. Malgre sa douceur de voix,
elle rappelle plus aussi le son de Violon que la Gambe de huit-pieds. La Fugara se fait gen6raie-
ment de huit et de quatre, rarement de seize." Regnier.
4 ' FUGARA, VOGARA, auch Tibia aperta (offene Plate), ist ein zieralich bekanntes offenes
Fldtenwerk zu 4' und 8' im Manual, von enger Mensur aus Holz oder Zinn, schwerer Ansprache
und fast so schneidendem aber hellerem Tone wie die Garaba. Diese Stimme soil, wiewohlsehr
selten, auch zu 16' vorkommen/'' Seidel.
140 ORGAN-STOPS
volume and strength commensurate with the requirements of the
tonal appointment of which it forms a part. It belongs to the tonal
structure of the foundation-work, and, accordingly, finds its proper
place in the First or Great Organ, imparting great richness and full-
ness to all combinations in which it enters. The following is the
composition of an effective stop of five ranks:
FULL MIXTURE V. RANKS.
CC tO G 12 15 19 22 26.
G# tO f 1 8 12 15 19 22.
fft 1 to f 5 8 12 15 19.
f# 2 to c< i 5 8 12 15.
The intervals in the two higher breaks fft 1 to c 4 belong to the
16 ft. harmonic series, and give great fullness and firmness to this
usually weak portion of the compass. A MIXTURE, of four ranks,
can be formed by omitting the fifth rank, without seriously impairing
the tonal value of the stop which lies in the ranks of lower pitch.
This, like all other compound harmonic-corroborating stops, must
be graduated in strength of tone as it ascends the manual compass.
FDXLQUINTE, Ger. The name used by some German organ-
builders to designate the manual QUINT, 5J^ FT., which belongs to
the 1 6 ft. harmonic series, and should appear only in a tonal division
in which there is a stop of 16 ft. pitch. The PULLQUINTE may be of
wood or metal, and of open or covered pipes. When drawn with a
full-toned stop of 8 ft. pitch, it produces the differential 16 ft. tone.
GAME A, Ital. Fr., GAMBE. Ger., GAMBE. This term ap-
plied alone in stop nomenclature, as it very frequently has been and
still is, is senseless; it literally signifies leg, and it would seem difficult
to apply that term to an organ-stop with any degree of propriety.
It is merely a wrong abbreviation of the name Viola da Gamba given
to the old instrument which was the precursor of the Violoncello.
Long usage, however, has so established the single term GAMBA,
that it will probably remain, along with other undesirable terms, for
a long time in the modern organ-builder's nomenclature. We
strongly recommend the abandonment of the senseless term, and
the adoption of the expressive term VIOL in its place.* The stop
* Viol was the English generic name of the stringed and bowed instruments which, developed
from the mediaeval Fiddle, preceded the Violin, and were used in England between the fifteenth
THEIR ARTISTIC REGISTRATION 141
which has passed under the simple name GAMBA has commonly
been formed of open cylindrical pipes of metal, of medium scale, and
8 ft. pitch, voiced to yield a more or less pronounced unimitative
string-tone. In its original form, as made by German organ-builders,
the stop was slow of speech and had a disagreeable "spitting" effect
when commencing to speak: these defects have been removed in all
later examples. We find the term used alone in the Organ in York
Minster, and in the generality of the Organs made by Walker &
Sons. The term is frequently used in combination with other terms
having reference to the shape of the pipes forming the stop or some
peculiarity in its tonality, thus: BELL GAMBA, CONE GAMBA, ECHO
GAMBA, CONTRA-GAMBA, 16 FT., etc. Translate these terms into
English and realize how absurd they are. See VIOLA DA GAMBA.
GAMBENBASS, Ger. The labial stop of 16 ft. pitch, formed of
open cylindrical metal pipes of medium scale, which furnishes the
true Pedal Organ bass to the manual VIOLA DA GAMBA, 8 FT. The
stop has been labeled in full, showing this relationship, as stated by
Seidel: "Die gewohnliche Tongrosse der Gamba ist im Manual
gewohnlich 8', selten 4', wo sie im letztern Palle Viola heisst, mit
der Altviola oder Bratsche im Einklang steht und der wahre Rep-
rasentant derselben ist, im Pedal kommt sei zu 16' unter dem Namen
Viola di Gambenbass oder Gambenbass vor."
GAMBETTE, Fr. An open metal stop of the VIOL DA GAMBA
family, of 4 ft. pitch, the pipes of which are of the same form and
tone as those of the unison stop. An example of the GAMBETTE,
4 FT., exists in the Great of the Organ in the Marienkirche, Lubeck.
It also exists in the Great of the Organ in the Cathedral of the In-
carnation, Garden City, L. L, and other large Roosevelt Organs.
In Organs of any importance the GAMBETTE should accompany the
VIOLA DA GAMBA, 8 FT., giving desirable aid in tonal grouping and
artistic registration.
GEDAMPFTREGAL, Ger. The name used to designate an
old lingual stop of the REGAL family, characterized by a muted or
muffled tone. Both the stop and its name are now obsolete. See
REGAL.
century and the end of the eighteenth. The "Chest of Viols" occupied an honored place in
almost every lordly mansion. It commonly comprised a Bass Viol, Tenor Viol, Alto Viol, and a
Treble Viol. In more important " Chests" these four instruments were duplicated. In some
"Chests" the Altos were omitted. The curious can see all these instruments in the Crosby
Brown Collection, in The Metropolitan Museum of Art, New York City.
142 ORGAN-STOPS
GEDECKT, Ger. The term is here given in its correct orthogra-
phy (past participle of decken to cover). The form GEDACKT is
frequently used both by German and English-speaking organ-
builders, but, being incorrect, should be abandoned in stop nomen-
clature. The simple term GEDECKT is used generally to indicate a
covered stop of wood or metal and of four, eight, sixteen, or thirty-
two feet pitch. In the Walcker Organ in the Cathedral of Ulm three
GEDECKTS of 8 ft. pitch and two of 16 ft. pitch are so labeled. These
five stops have, beyond their pitches, no distinctive tonal coloring,
all yielding normal covered flute-tone. The only other stop-name in
which the term appears is KLEINGEDECKT, 4 FT. To distinguish the
several covered stops, the pipes of which vary either in form or the
quality of covered tone they yield, German organ-builders have
introduced such compound names as the following; ANGENEHM-
GEDECKT, DOPPELGEDECKT, DOPPELROHRGEDECKT, GELINDGEDECKT,
GROBGEDECKT, GROSSGEDECKT, HUMANGEDECKT, KLEINGEDECKT,
LlEBLICHGEDECKT, MUSICIERGEDECKT, K.OHRGEDECKT, STARK-
GEDECKT, STILLGEDECKT, and WEITGEDECKT. See these names.
The GEDECKTS are made of all pitches from 16 ft. to 2 ft. ; but rarely
of the last high pitch.
GEDECKTBOMMER, Ger. The term, rarely used, to desig-
nate a covered stop of 8 ft. foundation tone, voiced and overblown
so as to sound the harmonic twelfth, if not exclusively, more promi-
nently than the prime tone. A stop of this name and tonality was
inserted in the Hauptmanual of the Organ built by Eugenius and
Adam Casparini (1703) for the Church of SS. Peter and Paul, Gor-
litz. It is described by Seidel and other authorities as a QUINT-
ATEN.* The name given the stop would seem to imply a disturbed
or uncertain tonality, caused by its dual voice. The only covered
stop of this class introduced in modern Organs is the ZAUBERFLOTE
(q. v.), invented by the late W. Thynne, of London.
GEDECKTFLOTE, Ger. A covered wood or metal stop of
medium scale and of 8 ft. or 4 ft. pitch, but commonly of the latter.
Its tone is of an unimitative flute quality as free from the harmonic
Twelfth as possible. Generally its tone resembles that of the DOP-
PELFLOTE, but, necessarily, less full owing to its pipes having single
mouths and being smaller in scale.
* 4 ' GBD ACT-POMMER. Pommer soil so viel sein als Bombarda. In der Gdrlitzer Orgel hinge-
gen soil, wie Adelung anffthrt, nach Boxbergs Beschreibung, diese Stimme eine starke QuiNTA'
gewesen sein." Schlimbach-Becker, Leipzig, 1843.
THEIR ARTISTIC REGISTRATION 143
TONE ^AND REGISTRATION. Such a voice as that described is
valuable in registration, combining well with, and imparting de-
sirable body to, imitative string-toned and lingual stops. As a rule,
there are too few stops of 4 ft. pitch introduced in modern Organs
for effective tonal combination and artistic registration: too much
dependence being placed on octave coupling, which, in its only
desirable- form, interferes with the independence of the coupled
claviers. An octave obtained on any clavier by coupling on itself is
inadmissible in' tonally correct and artistic registration.
GEDECKTQUINTE, Ger. A covered harmonic-corroborating
stop of 5^ ft. pitch, belonging to the 16 feet harmonic series, and,
accordingly, suitable for insertion in any division of the Organ in
which there are important stops of 16 ft. pitch. When combined
with a stop of 8 ft. tone it creates the differential 16 ft. tone. Of
10^ ft. pitch it belongs to the 32 ft. harmonic series, and is suitable
for the Pedal Organ, where in combination with the DIAPASON,
1 6 FT., it generates the differential 32 ft. tone. These differential
tones, however, are not so effective as those generated by open stops.
GEIGENOCTAV, Ger. Eng., VIOL OCTAVE. The appropri-
ate name for the open metal stop of 4 ft. pitch which is the true
Octave of the GEIGENPRINCIPAL, 8 FT. (q. .). The scale of its pipes
should be a little smaller than that of the corresponding pipes of the
unison stop. Its voice should be distinctly softer than, while similar
in tonal character to, that of the GEIGENPRINCIPAL, rendering it
extremely valuable in artistic registration.
GEIGENPRINCIPAL, GEIGENPRINZIPAL, Ger Eng.,
VIOLIN DIAPASON. A metal labial stop of 8 ft. pitch, the pipes of
which are cylindrical, and of medium scale in the best examples.
Its tone is, as the name implies, a combination of organ-tone and
string-tone, the former predominating in a decided manner. The
German name is to be preferred, for the stop is not, strictly classed,
a DIAPASON. Better English terms, in any case, would be those we
have used; namely, VIOL PRINCIPAL, GRAND VIOL, and VIOL DIA-
PASON; the last being less appropriate than the others, while, in
part, it recognizes old terminology. In form, the pipes of the GEI-
GENPRINCIPAL are similar to those of the PRINCIPAL or DIAPASON,
only being smaller in scale and slightly different in mouth-treat-
ment. The scale of the CC (8 ft,) pipe varies in different examples
from 4 inches to sJ^ inches in diameter, the mean being the most
desirable scale for general adoption. While the GEIGENPRINCIPAL
144 ORGAN-STOPS
is rightly a unison manual stop, and invariably appears as such in
English and American Organs, German organ-builders have formed
it of 1 6 ft., 8 ft., and 4 ft. pitch, as shown in Walcker's Organ in the
Cathedral of Riga. In that important instrument, comprising 116
speaking stops, there is a GEIGENPRINCIPAL, 16 FT., on the Second
Manual, and GEIGENPRINCIPALS, 8 FT. and 4 FT., on the Third
Manual. It would be more expressive to label the 4 ft. stop GEIGEN-
OCTAV. The stop does not seem to have found favor among French
organ-builders, for we have not been able to find a single instance of
its insertion, or that of an equivalent, in any French Organ. The
French organ-builders appear to have been, and indeed are,
strangely conservative in all matters of tonal appointment.
TONE AND REGISTRATION. The tone of the GEIGENPRINCIPAL
varies considerably in examples made by different builders ; in this
as in almost all the other stops, there being no standard studiously
worked up to. It is easy to imagine an organist saying: " I have a
beautiful GEIGENPRINCIPAL in my Organ which I find very valuable
in registration"; while another may, with equal propriety, say:
"I rarely use my GEIGENPRINCIPAL, for its tone is neither agreeable
alone, nor sympathetic in combination with other stops." This
should not be. The proper voice of the stop is a combination of pure
organ-tone with a bright string-tone in due subordination; the latter
imparting that richness to the foundation tone which has won the
stop universal approval among German- and English-speaking organ-
builders and organ-lovers. This compound voice, in which certain
concordant upper partial tones are present, is extremely valuable
in artistic registrations, in which fullness combined with a delicate
string quality is called for, and in which so refined a string-tone as
that of the VIOLA D'AMORE would be too pronounced. We admit
that such tonal refinements are not appreciated by the general class
of organists in the noisy craze of to-day.
GEIGENREGAL, Ger. An old lingual stop, of 8 ft. pitch, the
tone of which, from a slight string quality, somewhat resembled that
of the Geige or Violin. Both stop and its name are now obsolete.
See REGAL.
GELINDGEDECKT, Ger. (from gelind mild). A covered
stop, of 16 ft. or 8 ft, pitch, the pipes of which are of the usual forms,
and constructed of wood or partly of wood and partly of metal. As
its name implies, the stop, when properly scaled, and voiced under a
moderate wind-pressure, yields a soft and refined tone which should
be less assertive than that of a LIEBLICHGEDECKT, inserted in the
THEIR ARTISTIC REGISTRATION 145
same Organ, and of the same pitch. It is rarely, if ever, desirable to
duplicate stops of the same character having the same strength of
tone in an Organ, even should they be inserted in different tonal
divisions.
TONE AND REGISTRATION. The GELINDGEDECKT, 8 FT., is a
stop of great value in refined registration, specially so when its voice
partakes of the tonality of the QUINTATEN. This desirable voice is
best obtained from small-scaled wood pipes, voiced on wind of very
low pressure: and to be of the greatest value it should occupy a
place in the softest tonal division of the Organ.* In such a position
it would combine in a charming manner with practically every other
stop, labial or lingual, in the division.
GEMSHORN, Ger. Fr., COR DE CHAMOIS. The name given
to an open labial stop, the pipes of which are conical in form when of
metal and pyramidal when made of wood. As a manual stop, it is
made of 8 ft., 4 ft., and 2 ft. pitch; and as a Pedal Organ stop it is of
1 6 ft. pitch, properly designated GEMSHORNBASS (q. v.). The tone
of the true GEMSHORN is rich, clear, and penetrating, having a beau-
tiful timbre which may be classed as between a normal reed-tone
and a string-tone. This most desirable tone differs according to the
tastes of the organ-builders of different countries and the ideas of
different voicers.f Of 8 ft. pitch, the stop is inserted in the Haupt-
werk (First Manual) of all Walcker's Organs of any importance from
that in the Cathedral of Riga, downwards. It is invariably of 8 ft,
pitch, and, we presume, of metal throughout, as it should be in all
good Organs. In the Choir of the Organ in the Town Church of
Fulda it exists as an 8 ft. stop, the bass and tenor octaves of which
are of wood and the others of tin. The GEMSHORN, 8 FT., in the
Hauptwerk of the Organ in St. Paul's Church, Schwerin, has its
bass octave of covered wood pipes, the remainder of the stop being
of tin. In the Great of the Organ in the chief Protestant Church of
* We have placed the GELINDGEDECKT, in our tonal scheme for the Concert-room Organ of
the Twentieth Century, in the Ancillary Aerial Organ, all the nineteen stops of which are on wind
of iH inches. We give the scale of the CC (8 ft.) pipe 1.78 inches in -width by 2.28 inches in
depth, the scale ratio being i: 2.3, halving on the twenty-first pipe. See "The Organ of the
Twentieth Century," pp. 329-330.
f " Le COR DE CHA.MOIS ou Ge-mshorn est une des jolies flutes e*troites et ouvertes dont 1'har-
monie suit celle du Sa-licional, mais plus d61icate encore. Ses tuyaux, plus pointus que ceux dela
Spitz-Flcete, les sons clairs mais lointains qui s'6chappent de ses levres senses, lui ont sans doute
valu le nom qu'il porte. On le fait d'6tain ou d'6toff e tant qu'il ne descend pas plus bas quehuit-
pieds. A seize on peut employer le bois. On le conf ond, mais a tort, avec deux registres d'assez
semblable timbre, mais a tuyaux Douche's, Coppel-Flcele, Spielfloete; enfin, il ne dedaigne pas de
figurer comme quinte (Gemshorn-Quinte ou Cylinder-Quinte} et de produire alors sur une mass*
de violes un assez curieux eftet. Le Gemshorn chantant se melange volontiers avec la Hohljlcete,
qui lui donne beaucoup de force par son harmonie voi!6e, mais profonde," Regnier.
10
146 ORGAN-STOPS
Utrecht GEMSHORNS of both 4 ft. and 2 ft. pitch are inserted. The
stop has been made of 16 ft. and several mutation pitches, but these
have been seldom resorted to.*" As a Pedal Organ stop of 16 ft.
pitch it deserves the attention of every artist organ-builder and
expert.
FORMATION. The pipes of the GEMSHORN are invariably of the form of
elongated and slender truncated cones; their open tops having diameters equal to
one- third of the diameters at their mouth lines. This is a generally recognized
rule; but, like all rules in pipe-proportions, it is open to slight modification under
artistic treatment and special voicing. The scale of the GEMSHORN like that of
all other stops, varies according to the ideas and aims of different organ-builders,
and the volume and qua) ity of the tone desired. A scale suitable for a Church or a
Concert-room Organ stop, in the ratio 1:2.519, gives the CC (8 ft.) pipe a diameter
at the mouth line of 4.96 inches, and a diameter at the top of 1.62 inches; theC
pipe diameters of 3.13 inches and 1.02 inches; and the c 1 pipe diameters of 1.9
inches and 0.64 inch. The illustration, Fig. 19, Plate III., shows a CC pipe accu-
rately drawn to this scale. For a GEMSHORN placed in a softly-toned division of a
large Organ or inserted in a true Chamber Organ a smaller scale should be used.
TONE AND REGISTRATION. As before stated, the proper tone
of the GEMSHORN is rich, clear, and penetrating, having a beautiful
timbre which may be classed between a normal unimitative reed-
tone and a viol-tone. It is this compound tone, more or less rich in
the lower harmonics, which renders the stop, in its different pitches,
so valuable in artistic registration. The tone, however, differs under
the treatments and ideas of different voicers; and is, of necessity,
affected by different scales and the relative proportions of the lower
and upper diameters of the pipes. Under all conditions the tone of
the true GEMSHORN is so desirable and distinctive that it favors the
introduction, when possible, of the entire family of 16 ft., 8 ft., 4 ft.,
2% ft., and 2 ft. stops. Such a family would give practically limit-
less means for effective and refined registration, and, accordingly, the
creation of numerous beautiful timbres in combination with labial
stops of the FLUTE and VIOL classes, and with all the softer-voiced
lingual stops. The great value of gathering families of stops of dis-
tinctive tonalities, and placing them in contrasting divisions, is
absolutely unrecognized in the organ-building and organ-designing
world to-day, which continues to be quite satisfied with the old-
*" GEMSHORN ist eine allgemein bekannte, brauchbare und sehr angenehme Pldtenstimme
welc.he, mit oben spitzig zulaufenden Pfeifen versehen ist. Dei Pfeifen haben zuweilen Seiten-
barte zu beiden Seiten des Auf schnittes. Man findet dies Register im Manual zu 8', 4', 2" und i ',
zuletzten beiden Grdssen zuweilen unter dem Namen SUPER-GEMSHORN und im Pedal zu 16'
wo es GEMSHORNBASS heisst. Als Quintstimme zu 10^', $ 1 A'> *%' und I %' kommt dieses
Register ebenfalls im Pedal und Manual unter dem Namen GEMSHORNQUWTE vor. Die Pfeifen
sind gewdhnlich von Zinn oder Metall, bei den 16 fussigen durfte vielleicht hier und da Holz in
Anwendung gebracht werden." Seidel.
I
PLATE III
l\9
GEM SHORN
HORN
I 2*
KERAULOPHONE
26
ORCHESTRAL OBOE
THEIR ARTISTIC REGISTRATION 147
fashioned and systemless manner in which stop-appointments are
made.
GEMSHORNBASS, GROSSGEMSHORN, Ger. The Pedal
Organ stop, of 16 ft. pitch, which furnishes the proper bass to the
manual GEMSHORN, 8 FT. The pipes of this important and beautiful
stop are either made of metal throughout, or of wood and metal,
the former material being confined to the 16 ft. octave. This stop
should be inserted in every Pedal Organ of the first rank, furnishing
a beautiful bass, of a medium weight, to countless manual tonal
combinations. By extension to forty-four notes, a very desirable
OCTAVE can be derived from it, extremely valuable in Pedal Organ
registration. The subject of such registration has not received the
attention it deserves; mainly, if not altogether, owing to the gener-
ally insufficient and too often miserable tonal appointments of
modern Pedal Organs. See GEMSHORN.
GEMSHORNQUINTE, Ger. An open labial stop, the pipes
of which are of metal throughout, or, as in the case of the largest
stop, of wood and metal. The pipes are formed in all respects simi-
lar to -those of the GEMSHORN (q. v.). This stop is fifth-sounding, as
its name implies; and, accordingly, is found of 10% ft., 5}^ ft., 2^
ft. , and i J^ ft. pitch ; the last pitch being rarely used. A GEMSHORN-
QUINTE, 10^ FT., of metal throughout, exists in the Pedal of the
Grand Organ in the Cathedral of Breslau; one of 51^ ft., also of
metal, exists in the Pedal of the Organ in the Church of St. Eliza-
beth, in the same city; and one of 2^ ft. is introduced on the Second
Manual of the Organ in the Cathedral of Schwerin.
GERMAN GAMBA. The name given by English organ-
builders to an unimitative string-toned labial stop made by old
German organ-builders, the speech of which was so slow as to render
it necessary to draw along with it another quick-speaking stop or
KOPPEL (q. v.), commonly in the form of a KOPPELFLOTE. The
GERMAN GAMBA, as made by a German organ-builder, was first
introduced into England by Schulze, of Paulinzelle, in the Organ he
built for the Parish Church of Doncaster, Yorkshire. The stop, un-
desirable on account of its tardy speech as well as its peculiar in-
tonation, has properly fallen into desuetude, being interesting only
as the precursor of the modern string-toned stops.
GLOCKENSPIEL, Ger. Fr., CARILLON. Ital., CAMPANELLA,
CAMPANETTA. Correctly considered under its original German
name this is a mechanical or percussion stop, of short compass,
148 ORGAN-STOPS
formed of dish-shaped bells, spiral rods, stee A bars, or bell-metal
tubes, sounded by a hammer-action, somewhat resembling that
of a pianoforte, actuated by a pneumatic or electro-pneumatic me-
chanism in the modern Organ. In old Organs the GLOCKENSPIEL
was generally little better than a curiosity in organ-building. Ex-
amples still exist in some old German Organs, as in those in the
Churches of St. Catherine, St. Nicholas, and St. Jacobi, at Hamburg.*
The percussion GLOCKENSPIEL, or what is commonly and appro-
priately called the CARILLON in its modern form of tubular bells,
although it cannot be strictly considered a legitimate organ-stop,
has now become a recognized adjunct to the tonal appointment of
both Church and Concert-room Organs: and it must be admitted
that when artistically used it is not without its claim for recognition,
as an effective element in tone production in combination with cer-
tain labial stops, and occasionally in solo-effects. Indeed, it might
be extended in compass and depth of tone so as to admit of the
proper rendition of bell-music, such as that composed and performed
by the great carillonneur, Matthias van den Gheyn, of Louvain.
GLOCKLEINTON, GLOCKENTON, Ger. An open metal
labial stop properly of large scale and 2 ft. pitch, voiced to yield a
tone which has a ringing quality suggestive of that of bells. It is
described by Seidel thus: "GLOCKLEINTON (Tonus fabri) ist ein
weit mensuriertes Pfeifenregister, welches so klingen soil, als wenn
man mit einem Hammer auf einen wohlklingenden Amboss schlagt.
*" GLOCKENSPIEL. Carillon. C'est chez les Allemands un jeu compost de clochettes au lieu
del'Stre de tuyaux. Ordinairement, on le place dans 1'interieur derriere le principal en montre;
quelquefois il est a 1'exterieur ou Ton volt des anges places dans une gloire tenant d'une main une
clochette sur laquelle ils frappent avec un marteau qu'ils portent dans 1'autre main. . . . Les
carillons ne s'entendent Ordinairement que dans les deux octaves superieures du clavier; cepen-
dant, il parait qu'il s'en trouve de quatre octaves, et que celui de 1'eglise Saint-M ichel a Ohrdruff
a cette etendue. II en existe aussi & la pe~dale. Au lieu de timbres en forme de cloche, on emploie
quelquefois des tiges m6talliques tourne'es en spirales et assuj^ties sur une caisse sonore qui aug-
mentel'intensit^ deleurssons. Un des inconvenients des carillons, est den'etre presquejamais
d'accord avec 1'orgue dont la temperature fait varier continuellement les jeux de fond dans des
proportions qui ne sont point dans le me'me rapport que celles des variations dss me'taux. Les
marteaux qui frappent les timbres ou les tiges metalliques sont repousses par un ressort apresles
avoir mis en vibration, afin de n'en pas arr^ter le son. " Hamel.
"GLOCKENSPIEL, Carillon, ist ein Register, welches statt der Pfeifen abgestimmte Glocken
hat. Gewohnlich ist es im Innern der Orgel hinter den Prinzipalpfeifen angebracht, urn den
Klang recht nahe zu bringen, zuweilen wird es auch von Engeln, welche in einer Glorie ange-
bracht sind und in der einen. (beweglichen) Hand ein Hammerchen, in der andern eine Glocke
halten, geschlagen. Manche Glockenspiele smd mit einem Dampfer von Leder, Tuch u. dergl.
versehen, urn das Ineinanderschwirren der Tdne zu verhindern. Die Glockenspiele gehen ge-
wohnlich nur durch die zwei oberen Oktaven des Manuals, doch giebt es auch Glockenspiele im
Pedal. Da der Glockenton sich auf eine hochst auffallende Art von dem Pfeifenton absondert,
und die Glocken nie mit den Pfeifen genau zusammenstimmen, auch die Andacht der Gemeinde
durch dieses Klimperwerke gestort wird, so ware es zu wiinschen, dass dergleichen unnutze
Register gar nicht mehr verfertigt wurden und die hier und da vorhandenen .ausser Gebrauch
kamen. " Seidel-Kothe.
THEIR ARTISTIC REGISTRATION 149
Diese Stimme steht zu 2' im Oberwerk der Gorlitzer Petri-Orgel."
A stop of this class and pitch is hardly to be recommended for
insertion in the proper tonal appointment of g,ny Organ constructed
to-day. As a SUPER-OCTAVE it would be much too assertive in tone
to be used in refined and artistic registration; except, perhaps, in
full organ effects, and even in such it might disturb perfect tonal
balance.
GRAND BOURDON, GROS BOURDON, Fr. A covered stop
of wood, of large scale, and of 32 ft. pitch, the pipes of which are
formed in all respects similar to those of the ordinary BOURDON,
1 6 FT. (<?. v.). It appears in the Pedal of the Organ in the Church
of Saint- Vincent de Paul, Paris. See CONTRA-BOURDON.
The term GRAND BOURDON has been employed by Walcker to
designate a very important Pedal Organ compound harmonic-cor-
roborating stop, of five complete ranks, belonging to the 32 ft. har-
monic series. The composition of this stop and the scales of its
CCC pipes are as follows :
GRAND BOURDON V. RANKS.
PRINCIPAL, 16 FT Width, 220 mm. Depth, 280 mm.
QUINT, 10% FT.
OCTAVE, 8 FT
TIERCE, 6% FT. At mouth,
At top,
SUPER-OCTAVE, 4 FT.
no
108
74
48
65
153
1 06
64
80
It will be observed that all the ranks are formed of wood pipes;
and that the TIERCE 62/5 FT., is formed of pyramidal pipes of the
SPITZFLOTE type, doubtless with the aim of imparting a lighter and
timbre-creating voice to this third-sounding rank; as is always desir-
able in the third-sounding ranks of compound harmonic-corrobo-
rating stops.
GRAND CORNET. The name properly employed to designate
a compound harmonic-corroborating stop, the several ranks of which
belong to the 16 ft. harmonic series, and extend throughout the
manual compass without a break. An example, of V. ranks, exists
in the Bombarde of the Cavaill-Coll Organ in the Church of Saint-
Ouen, Rouen. The stop is desirable only in a tonal division in which
there are two or more assertive stops of 1 6 ft. pitch, preferably lin-
gual. But when the GRAND CORNET is made, as it should be, a
timbre-creating stop, its value is great in combination with both
labial and lingual stops. An example of such a stop is here given:
150 ORGAN-STOPS
TIMBRE-CREATING GRAND CORNET V. RANKS.
I. FL^TE A CHEMINEE . . . Metal 8 Feet
II. SPITZQUINTE Metal sJ^ "
III. VIOLETTA Tin 4
IV. GEMSHORN Metal 2% "
V. PICCOLO Metal 2
The composition of timbre-creating compound stops will, of
necessity, be dictated by the stop-apportionments of the Organs in
which they are placed.
GRAND PRINCIPAL, GRAND DIAPASON. Names that
have been employed to individualize the most important unison
(8 ft.) stop of pure organ-tone, occupying a place in the foundation-
work of an Organ. It was used in the First or Great Organ of the
Concert Organ installed at the Louisiana Purchase Exposition, in
1 904. The term is desirable when there are three or more DIAPASONS
introduced together in any Organ, indicating the tonal preeminence
of one of the series.
GRAND VIOL. The name employed to individualize the most
important unison (8 ft.) stop yielding unimitative viol organ-tone,
inserted in an Organ: in this respect occupying a more command-
ing position, tonally, than the GEIGENPRINCIPAL, 8 FT. An example
exists in the unexpressive Subdivision of the First or Great Organ
of the principal Organ in the Church of Our Lady of Grace, Hoboken,
N. J.
The term grand has been applied to other stops for the purpose
of indicating their relative importance; we, accordingly, find such
terms as GRAND FLUTE, GRAND OCTAVE, GRAND CORNET, and
GRAND MIXTURE.
GRAVISSIMA, Lat. This term, which appears in the tonal
schemes of certain important modern Pedal Organs, must not be
understood to designate an independent stop. It is simply the
acoustical effect, or what is properly designated the differential tone,
produced naturally by the combination of tones of 32 ft. and 21%
ft. pitch, standing at the interval of a perfect fifth apart, thus:
GENERATING TONES. DIFFERENTIAL TONE.
CCCC32FT. GGGG2IMFT. ' " ! CCCCC 64 FT. TONE
1 6 24 8 vibrations
There are three ways of producing what is more expressively
termed the Vox GRAVISSIMA, varying in effectiveness, i. By
THEIR ARTISTIC REGISTRATION 151
associating with the DOUBLE DIAPASON, 32 FT., an independent
SUB-QUINT, 21 y$ FT., tonally the most effective method. 2. By
extending the compass of the DOUBLE DIAPASON seven notes, and
introducing a Quint coupler acting on the same stop and sounding
its notes in fifths. 3. By extending the DIAPASON, 1 6 FT. , downward
to GGGG, and coupling it to the DOUBLE DIAPASON, 32 FT. We
are in favor of the first method, because it allows the SUB-QUINT or
GROSSQUINTENBASS, 21 J^ FT. (q. .), to be used in combination with
any other labial or lingual stops of 32 ft. pitch. This arrangement ob-
tains in the Pedal of the Schulze Organ in the Cathedral of Bremen.
In the Organ erected by the Hutchins-Votey Organ Company, in
Woolsey Hall, Yale University, the GRAVISSIMA was produced by
the combination of the open wood DOUBLE DIAPASON, 32 FT., and a
QUINT, 21 J^ ft. tone, derived from the CONTRA-BOURDON, 32 FT.
There has not been sufficient experience gained as yet respecting
the acoustical effects of the extremely grave differential tones on the
proper voices of the Organ for any accurate conclusion to be reached
respecting their nature. It is quite evident, however, that they
can have no claim to be classed as musical sounds of a 64 ft. octave.
When, as Helmholtz states, determinate musical tones cease at that
produced by 41.25 vibrations, what can be said of those sounds
produced between 8 and 16 vibrations per second? The combina-
tion of the two generating -stops certainly produces a remarkable
acoustic effect, evidently due to the generation of a great series of
harmonic over-tones: hence the sole value of the Vox GRAVISSIMA.
GROBGEDECKT, GROSSGEDECKT, Ger. Lat., PILEATA
MAGNA. The terms signifying " great covered stop, " and used by
German organ-builders to designate a large-scaled and loud-voiced
covered stop of 16 ft. or 8 ft. pitch, commonly inserted in a manual
division. A GROSSGEDECKT, 16 FT., exists in the Great of the Organ
in the Church of St. Dominic, Prague; and a GROBGEDECKT, 8 FT., is
placed on the First Manual of the Walcker Organ in the Synagogue,
Berlin.
The word grobe has been used as a prefix, for the purpose of indi-
cating strength of tone, in the case of several other stop-names. We
accordingly find in old lists such terms as GROBCYMBEL, GROBMIX-
TUR, GROBPOSAUNE, GROBREGAL, etc.
GROSSDOPPELGEDECKT, Ger. A covered wood stop of
1 6 ft. pitch, the pipes of which are properly of large scale, deep in
proportion to their width, and have double mouths, after the fashion
of the DOPPELFLOTE (q. v.). The stop is suitable for the Pedal Or-
152 ORGAN-STOPS
gan or for a manual division in which there are powerful labial or
lingual stops of unison (8 ft.) pitch. Carefully made and artistically
voiced, this important stop, in combination, would be productive of
many fine and uncommon tonal effects in which gravity would be a
special element.
GROSSFLOTE, Ger. Fr., GROSSE FLUTE. An open wood
stop of medium scale and properly of 8 ft. pitch; usually placed in a
full-toned manual division of the Organ, while it forms a good OC-
TAVE, 8 FT, for the properly appointed Pedal Organ. In the general
form of its pipes it resembles the English CLARABELLA, but it is
voiced to yield a more powerful tone of an unimitative flute organ-
tone. For this purpose, the mouths are cut high and their upper
lips made thick, carefully rounded, and polished with black-lead:
this last is seldom done in ordinary trade practice. Leathered lips
should be avoided.
TONE AND REGISTRATION. When the GROSSFLOTE is properly
made and voiced to yield its characteristic tone, it is very valuable
in registration in which a background of full unimitative flute-tone
is required. It combines in the most effective manner with all the
more powerful lingual stops, imparting great body and firmness to
their tones without destroying their characteristic voices.
GROSSGEDECKT, Ger. The term appropriately applied to a
large-scaled covered stop, of 16 ft. pitch, yielding a pure covered-
tone similar to that of the true LIEBLICHGEDECKT, 16 FT., but of
much greater volume and assertiveness. Its place is in a manual
division of the Organ, in which there are powerful lingual and other
commanding stops, where it will play an important rdle in effective
registration.*
GROSSHOHLFLOTE, Ger. The Pedal Organ stop of 16 ft.
pitch, the open pipes of which are of wood and formed and voiced
in all respects similar to those of the manual HOHLFLOTE, 8 FT.
See HOHLFLOTE.
GROSSNASAT, Ger. Fr., GROS NASARD, GROSSE QUINTE.
This mutation stop, which is of 10% ft. pitch in the Pedal Organ
and 5 J^ ft. pitch in the manual divisions, is made of either wood or
metal and both in open and covered forms. In the first pitch it
belongs to the 32 ft. harmonic series, and in the higher pitch to
1 6 ft. harmonic series. As GROSSNASAT, 10% FT., it exists as an
*See " The Organ of the Twentieth Century," pp. 311, 312.
THEIR ARTISTIC REGISTRATION 153
open stop in the Pedal of the Organ in the Cathedral of Halberstadt,
As GROSSNASARD, 10% FT. of wood, it is inserted in the Pedal of
the Organ in the Cathedral of Merseburg. It appears as GROSSE
QUINTE, 10^3 FT., in the Pedale, and as GROSSE QUINTE, 5^ FT.,
in the Clavier des Bombardes, in the Grand Organ in the Cathedral
of Notre-Dame, Paris. The stop of 10^ ft. pitch was inserted in the
Pedal, and of both 10^ ft. and 5^ ft. pitch in the First or Great
division of the Concert Organ installed in the Festival Hall of the
Louisiana Purchase Exposition, 1904.
TONE AND REGISTRATION. The value of the harmonic-creating
or corroborating tones, such as those of 10% ft. and 5^ ft. pitch, in
registration is hardly realized at the present time, owing to the
rarity of their introduction in modern Organs. In a scheme lying
before us, for a much-divided Organ of 283 speaking stops, there is
only one Pedal Organ stop of 10% ft. pitch, and only one manual
stop of 5 % ft. pitch, showing how little attention is paid by the or-
gan-builder of to-day, even in his greatest essays, to the scientific
side of his art; and how little thought or care he bestows on provid-
ing the organist with proper material for varied and artistic regis-
tration.* How long are organists to remain content with the results
of incompetency which handicap them at every turn in their essays
in registration? Surely it is time that science and art are infused
into the tonal appointment of important Organs.
GROSSOCTAV, Ger. The name given by Walcker to an open
labial stop, of 8 ft. pitch, placed on the First Manual of the Organ
in Paulskirche, Frankurt a. M. The term is used because, strictly
considered, the manual is of 1 6 ft. pitch, containing, including the
PRINCIPAL, three stops of 16 ft. pitch. In the Pedal of the same
Organ is a stop of 16 ft. pitch labeled GROSSOCTAVBASS.
GROSSPOSAUNE, Ger. The term employed by German organ-
builders to designate the dominating lingual stop, of 32 ft. pitch,
belonging to the Pedal Organ. It is, as its name implies, similar in
all essentials to the CONTRA-TROMBONE, 32 FT. (q. v.). This grave
stop has not been a favorite with German organ-builders. In the
* If the reader will refer to our scheme for the Concert-room Organ, given in Chapter XI, of
"The Organ of the Twentieth Century," he will find we have provided in the Pedal Organ one
fifth-sounding labial stop of 21 M ft., one labial stop of 10 % ft., and one lingual stop of 10 % ft.
pitch. Properly distributed in the several manual Organs, are six labial stops of 5 % ft., and two
lingual stops of 5 K ft. pitch. In addition to these important stops, there are six fifth-sounding
stops of 2 % ft. pitch. In all, seventeen fifth-sounding stops out of the total number of 230
speaking stops in the entire tonal scheme: yet we know, on both scientific and artistic grounds,
there is not one fifth-sounding stop too many to meet the requirements of artistic registration
and scientific timbre-creation.
154 ORGAN-STOPS
Pedal of the Organ in the Nicolaikirche, Leipzig, the stop is labeled
POSAUNE, 32 FT. The stop is a free-reed.
GROSSPRINZIPAL, Gen Fr., GROS PRINCIPAL. The name
appropriately given to the open labial stops of 16 ft. pitch in the
manual divisions and 32 ft. pitch in the Pedal Organ, yielding pure
organ-tone, and, accordingly, belonging to the foundation-work of
the Organ. It exists, of 32 ft. pitch in the Pedal of the Organ in the
Marienkirche, Liibeck. It also exists, as a displayed stop of English
tin, of 1 6 ft. pitch, in the Organ in the Church of Waltershausen.
These stops are similar to the English DOUBLE DIAPASONS of
corresponding pitches.
GROSSQUINTENBASS, Ger. This name is given to a covered
stop, of 21 J^ ft. pitch, inserted in the Pedal of the Organ in the
Cathedral of Bremen. The stop strictly belongs to the 64 ft. har-
monic series, and was evidently introduced with the view of pro-
ducing the acoustic differential tone which we have termed Vox
GRAVISSIMA (see GRAVISSIMA). This fine Organ of 59 speaking
stops, was built by Schulze, of Paulinzelle.
GROSSREGAL, Ger. An old lingual stop of the REGAL family,
the only peculiarity of which existed in its grave pitch, which was of
1 6 ft. The stop is no longer made and its name is obsolete. See
REGAL.
GROSSTERZ, Ger. Fr., GROSSE TIERCE. This mutation
harmonic-corroborating stop is of 6 2 /5 ft. pitch in the Pedal and 3)^
ft. pitch in the manual Organs. In the former it belongs to the 32
ft. harmonic series, and in the latter to the 16 ft. harmonic series;
in each case corroborating the fourth upper parial tone of the prime
of the corresponding series. A GROSSE TIERCE, 6 2 /5 FT., is inserted
in the Pddale, and a GROSSE TIERCE, 3)^ FT., in the Bombardes, in
the Grand Organ in the Cathedral of Notre-Dame, Paris.
The stop may be formed of pipes of metal or wood and either
open or covered, according to the quality of the tone desired. In
any case, the tone should be somewhat subdued, for third-sounding
tones are liable to be undesirably assertive and penetrating, and,
accordingly, have to be confined to full-toned combinations. When
properly subordinated to the fifth-sounding stops, the TIERCES
become extremely valuable in artistic registration, imparting dis-
tinctive tonal coloring.
GROSSUNTERSATZ, Ger. A large-scaled covered stop of
THEIR ARTISTIC REGISTRATION 155
wood, and of 32 ft. pitch, belonging to the Pedal Organ. An ex-
ample, under the name, exists in the Pedal of the Organ in the
Church of Waltershausen. The Organ was built by Trost, of Alten-
burg, in 1730. When of medium scale, the stop has been commonly
labeled UNTERSATZ, 32 FT., as in the Pedals of the Organs in the
Cathedral of Merseburg and the Nicolaikirche, Leipzig. A stop of
the same form and pitch, but of smaller scale, occupies a place on
the First Manual of Walcker's Organ in Paulskirche, Frankfurt, a.
M., where it is labeled MANUAL-UNTERSATZ, 32 FT.
A covered stop of this grave pitch cannot be pronounced entirely
satisfactory, and utterly fails to take the place of an open stop of
32 ft. Cheapness and limitations of space are the only arguments
that can be advanced in favor of its introduction, either as a Pedal
Organ or a manual stop. If it is made, we strongly advise its being
voiced with the harmonic-bridge, so as to generate as great a series
of harmonic upper partials as possible: these partials alone give
tonal value to the lower octave and a half of the stop; the prime tones
of which are not per se musical sounds.
H
HALBPRINZIPAL, Ger. The term sometimes used by the old
German organ-builders to designate the half-length PRINCIPAL;
namely, the ordinary organ-toned OCTAVE, 4 FT. Schlimbach says:
"H ALB PRINCIPAL so viel als Principal 4 Fuss, weil das gewohnliche
Hauptprincipal 8 Fuss hat. Zuweilen bedeutet es auch Octav 8
Fuss." The old English organ-builders abbreviated the term, and
called the stop, simply and illogically, PRINCIPAL, 4 FT. The correct
term, now coming into general use, is OCTAVE, 4 FT.
HARFENPRINZIPAL, Ger. A manual, open labial stop of
8 ft. pitch, the pipes of which are cylindrical and of small scale.
The stop was commonly made of tin, and voiced to yield a delicate
compound tone, which, in quick arpeggio passages, bore a faint re-
semblance to those of the orchestral Harp. The effect was secured
by the presence of certain upper partial tones which are prominent
in the sounds produced by plucked gut strings. Seidel says : " HAR-
FENPRINZIPAL ist ein lieblich intoniertes Principal, welches einen
der Harfe ahnlichen schnarrenden Ton haben soil." The stop is
disused and the name is obsolete.
HARFENREGAL, Ger. A soft-toned lingual stop, the tones
156 ORGAN-STOPS
of which bore a remote resemblance to the twang of harp strings
when roughly plucked. The stop and its name are both obsolete.
HARMONIA ^ETHERIA, Grk A compound harmonic-cor-
roborating stop, composed of two or more ranks of very small-scaled
and delicately voiced metal pipes. In its most desirable form it is
composed of ^EOLINE labial pipes (See ^EOLINE); but in this ex-
tremely refined form it is at present practically unknown. The stop
has been made in different forms. It appears, under the unusual
name "HARMONICA ^ETHERICA, " of two ranks, in the Echo of the
Organ, built by Schulze, in the Parish Church of Doncaster, York-
shire. In Walcker's Organ in Riga Cathedral a HARMONIA J&THE-
RIA, of three complete ranks TWELFTH, 2^ FT., FIFTEENTH, 2 FT.,
and SEVENTEENTH, i% FT., is placed on the Third Manual.
Similar stops are inserted in the Organ in the Gewandhaus, Leipzig,
and several other Walcker Organs. The German organ-builders do
not seem to have carried the stop beyond three ranks, nor do they
appear to have made breaks in the ranks.
FORMATION. While a stop of through ranks, such as adopted by Walcker and
other builders, formed of small-scaled and softly voiced pipes, could not fail to be
very useful ; a HARMONIA ^STHERIA, in the form of a MIXTURE, composed of several
ranks of high-pitched JEouNE or Vox ANGELICA pipes would be extremely valu-
able as a harmonic-corroborating stop in the softest toned manual division of an
artistically appointed Organ. It certainly should find a place in the Organ of the
twentieth century. The following is the appropriate composition for a full stop
of six ranks :
HARMONIA ^ETHERIA VI. RANKS.
CC to F# . . . . 15 17* 19 22 26 -29.
G to f# .... 12 15 17* 19 22 26.
g 1 tO f# .... 8 12 15 17* 19 22.
g a to c* .... i 8 12 15 17* -19.
Should this composition be considered too full, the third-sounding break* in
each rank may be omitted, making a five-rank stop. If the SEVENTEENTH is re-
tained, it should be voiced softer than the octave- and fift;h-sounding ranks, for
that interval is, otherwise, very liable to be too assertive. This refined harmonic-
corroborating stop should speak on wind of ij^ inches to 2% inches, and be most
scientifically regulated. The pipes to be of tin.
TONE AND REGISTRATION. Properly made and artistically
voiced, the HARMONIA ^THERIA should yield a singing, silvery,
compound tone of great beauty; so delicate that, in registration, it
may be employed with any of the softest unison stops in the Organ,
string- or flute-toned. As a harmonic-corroborating or timbre-
creating stop it would be difficult to overrate its value a value
absolutely unrealized in the present noisy-organ epoch.
THEIR ARTISTIC REGISTRATION 157
HARMONICA, HARMONIKA, Ger. An open labial stop,
usually of 8 ft. pitch, the pipes of which are of wood, of small scale!
and voiced to yield a combination of flute-tone and string-tone of a
soft and beautiful quality. The stop has long been a favorite with
German organ-builders, while it is practically unknown by those of
this and other countries: indeed, both in the design and construc-
tion of wood stops, generally, the German organ-builders have
always taken the lead, surpassing all others. Both French and
English organ-builders have favored metal pipes, under the mis-
taken impression, perhaps, that a better tone could be obtained
from them than from wood pipes; but more likely from the fact that
metal-pipes are more easily made and voiced. This was the rock
Willis split upon in his tonal appointment of the Organ in the Royal
Albert Hall, South Kensington, London. In that large Organ there
are only two complete wood stops, and the tone of the instrument
suffers accordingly.
Fine examples of the HARMONIKA, 8 FT., exist on the Third
Manuals of the Organs, by Walcker, in the Cathedrals of Riga and
Ulm, and in the Concert Organ in the Gewandhaus, Leipzig. It
exists in the Echo of the Organ in the Cathedral of Lund, Sweden,
in which it is said to produce "a most beautiful effect." The stop
is of wood. Equally good examples, labeled HARMONICA, 8 FT.,
exist in the Echo of the Schulze Organ in the Parish Church of Don-
caster, and in the Choir of the Organ in the Church of St. Bartholo-
mew, Armley, Yorkshire.
FORMATION. The scale and form of the HARMONICA vary in different examples;
and while in the generality of cases its pipes are straight, in some examples they
are slightly pyramidal. For the stop of 8 ft. pitch, formed of straight, square
pipes, Haas, the distinguished German organ-builder, has recommended the
following small scale :
SCALE OF HARMONICA, 8 FT.
CC C c 1 c 2 c*
66 mm. 40 mm. 24 mm. 15 mm. 9 mm.
We are in favor of a slightly larger scale and the slower reduction secured by
the ratio 1 : 2.66, which halves on the eighteenth pipe. The following is our pro-
posed scale, in inches, for square pipes :
SCALE OF HARMONICA, 8 FT. RATIO 1:2.66.
CC C c 1 c 2 c* c4
2.84 1.74 i. 06 0.65 0.40 0.25
Some German organ-builders use scales which give the pipes of the HARMONICA
a greater depth than width. An example of this treatment obtains in the fine stop
158
ORGAN-STOPS
in the Armley Organ, in which the pipes from middle c r to the top note are about
twice their width in depth. Schulze seems to have found this extreme proportion
favorable to the production of very soft and refined qualities of tone. The chief
peculiarity of the HARMONICA pipe is the formation of its mouth and attendant
cap, shown in the accompanying illustration, Fig. 20, which presents a Front View
and Sections of the mouth portion of the c 1 pipe of the Armley stop. It will be
seen that the mouth is circular, hollowed on the inside of the pipe so as to present
FIG. 20
a thin lip to the wind-stream. The manner in which this is done is shown in the
Sections 2 and 3. The size of the circular mouth varies according to the quality
and strength of the tone desired, but it should not be less than one-half the internal
width of the pipe. It will be observed in Section 2 that the cap is compound,
being formed of an inner, wedge-shaped piece, which forms the lower lip of the
mouth ; and an outer piece, in which the wind- way is carefully cut in the manner
shown, in black, in the Transverse Section of the mouth, 3. In the formation of
the compound caps extreme care and accuracy must be observed in graduating
the thickness and slope of the wedge-piece, and smoothly rounding and polishing
with black-lead its upper edge or lip.* The chief difficulty in connection with the
voicing of this beautiful stop lies in the adjustment of both parts of the caps with
relation to the circular mouths, and the exact proportions of the wind-ways, so as
to obtain a perfectly even intonation throughout the compass. As a rule, the cap
is adjusted to cover about one-third of the diameter of the mouth, but different
* The dimensions given by Tdpf er-Allihn for the outer portion of the cap (Frosch) and the
inner wedge-piece (Platte) are of sufficient importance to be given here : " Die Lange der grOssten
Frdsche kann 50 mm betragen und nimmt ab bis zu 36 bis 40 mm. Die Dicke kann filr die
grdsseren Pfeifen 20 mrn und fur die kleineren zwischen 8 bis 10 mm betragen. Es versteht sich,
dass die vordere Dicke der Platte von den hdchsten T6nen bis zum tiefsten gleichmassig zuneh-
men muss, weil ausserdem die Gleichmassigkeit des Tones darunter leiden wurde. ' Man kann
diese Dicke fur C 8 Fuss 5 mm und fur c3 0,2 mm setzen. Bei diesen Annahmen erhalt man die
vordere Dicke der Platte fur c = 2.2 mm; fur c 1 = I mm; fur c 2 = 0.45 mm. Die hintere Dicke
nimmt von 8 mm bis zu 3 mm ab. " " Die Theorie und Praxis des Orgelbaues, " p. 232.
THEIR ARTISTIC REGISTRATION 159
positions are adopted according to the tonality desired. While pine may be used
for the sides and backs of the pipes, it is necessary, with such a form of mouth, for
their fronts and caps to be of some close-grained hard wood: preferably beech for
the caps.
TONE AND REGISTRATION. The tone of the HARMONICA in its
finest quality is extremely refined, being a combination of flute- and
string-tone, in some examples inclining to one and in some to the
other tonality. Locher says it "is a very tender 8-ft. string-tone
stop, of narrow scale, intonated between the MOLINA and SALI-
CIONAL." According to Topfer-Allihn, "Es ist eine Stimme von
atherischem Charakter, welche bei sanften Altargesangen zum
Gemtite des stillen Beters spricht." Although in the case of this
stop one cannot speak from actual experience, it is not difficult to
arrive at a fairly accurate estimation of its value in refined registra-
tion. It has hitherto been, and will continue to be inserted in the
softest-toned division of the Organ, where it will be associated with
stops labial and lingual of varied tonalities and, properly, of
equally refined voices, admitting of combinations of great variety
and beauty. If the HARMONICA inclines to flute-tone, the most
effective registrations will be those with labial stops of string-tone
and the lingual stops. It will be specially valuable with the Vox
HUMANA, should one be present. On the other hand, if the HAR-
MONICA inclines (as many German examples seem to do) to string-
tone, its most effective registrations will be with stops of flute-tone
and pure organ-tone, imparting to them a delicate brilliancy and
clearness. To the lingual stops it will add firmness and richness.
In the registration of stops of soft and refined tonalities the principle
of tonal contrast should always be observed. The HARMONICA
should speak on wind of i^ inches to 2^ inches pressure.
HARMONICABASS, Ger. A Pedal Organ labial stop, of 16
ft. pitch, formed of small-scaled open pipes of wood, voiced on a
low-pressure wind, and yielding (like the HARMONICA, 8 FT., to which
it is the true bass) a tone in which both refined flute-tone and string-
tone are combined. A representative example of this uncommon
stop exists in the Pedal of the Organ in the Catholic Church, at
Berne, Switzerland. Locher describes this stop as "specially suita-
ble for the accompaniment of soft passages. J>
The pipes should be constructed and voiced in the same manner
as those of the HARMONICA (q. y.) ; and the following scale will be
found suitable. As this stop would be required to furnish the bass
to combinations of soft stops, it should be voiced on wind of 2^
or 3 inches.
160 ORGAN-STOPS
SCALE OF HARMONICABASS, 16 FT. RATIO i: 2.66.
CCC CC C G
4-44 2.73 1.67 1.25
HARMONIC CLARIBEL. The name given by Thomas
Casson, organ-builder, of London, to a HARMONIC FLUTE of a large
scale and full intonation, constructed by him. Other English organ -
builders have used the name to designate harmonic stops of a similar
character. A fine example exists in the Organ constructed by Nor-
man & Beard, of Norwich, in 1905, for the Colston Hall, Bristol.
This stop is of 8 ft. pitch, and speaks on high pressure wind. Its
pipes have inverted mouths, and are of double standard length and
harmonic from c 1 to the top note. Such a treatment is favorable to
the augmentation of the treble.
HARMONIC CLARION. Fr., CLAIRON HARMONIQUE. This
lingual stop is the true Octave of the HARMONIC TRUMPET, 8 FT.,
and is, accordingly, of 4 ft. pitch. The pipes are formed in all re-
spects similar to those of the unison stop, being of about double the
normal speaking lengths, voiced on high-pressure wind to speak the
octave pitch. The CLAIRON HARMONIQUE was invented by Aristide
Cavaille-Coll, and introduced for the first time in the Rcit-Echo
Expressif of the Organ in the Royal Church, Saint-Denis, con-
structed in 1841.
HARMONIC DIAPASON. The name used by Bryceson
Brothers, organ-builders, of London, to designate a large-scaled
labial stop of DIAPASON formation, the pipes of which are, in the
more important examples, of double the standard speaking length
and harmonic from G or c 1 to the top note. The aim of the build-
ers was to produce a volume of powerful foundation-tone; but, as
might be expected on scientific grounds, the result was anything but
pure organ-tone. The step was in the wrong direction, and, natur-
ally, the result was a powerful and unpleasant flute-tone. Examples
exist in certain Organs by Bryceson Brothers, notably one in the
Organ built in 1882 for the Concert Hall, Paisley, Scotland. Such a
stop is not a desirable addition to the Organ of to-day, which is
over-furnished with unduly assertive voices.
HARMONIC FLUTE. Fr., FLUTE HARMONIQUE. The stop
formed of cylindrical metal pipes of 8 ft. and 4 ft. pitch. In the uni-
son stop the pipes from the middle octave to the top note are made
of double the normal speaking lengths; and are voiced and so blown
THEIR ARTISTIC REGISTRATION 161
as to speak the octaves of the tones properly belonging to the full
lengths of the pipes so treated. For full particulars respecting this
important stop, see FLUTE HARMONIQUE. This stop was invented
by Cavaille-Coll, and inserted, for the first time, in the Organ in the
Royal Church, Saint-Denis, in the year 1841.
The HARMONIC FLUTE has been very successfully constructed of
wood in several forms. For particulars respecting those yielding
imitative flute-tone, see ORCHESTRAL FLUTE.
HARMONIC PICCOLO. Fr., PICCOLO HARMONIQUE. A
cylindrical metal stop, of 2 ft. pitch ; the pipes of which are double
the standard speaking lengths, and are formed and voiced in all
respects similar to those of the harmonic portion of the HARMONIC
FLUTE, 8 FT. (q. v.). An example of this stop under the name Oc-
TAVIN HARMONIQUE, 2 FT., was inserted by Cavaille-Coll in the
Rcit-Echo Expressif of the Organ in the Royal Church, Saint-
Denis, in the year 1841. See FLUTE HARMONIQUE.
HARMONIC TRUMPET. Fr., TROMPETTE HARMONIQUE.
Ger., HARMONIETROMPETE. A lingual stop, of 8 ft. pitch, the reeds
of which are of special formation, and the resonators of inverted
conical form are, for the greater portion of its compass, of greater
lengths than those required for the ordinary TRUMPET, 8 FT. The
reeds are of the open class and of large scale, and the tongues are
thick and well curved, so as to respond properly to the high-pressure
wind on which the stop is voiced; producing powerful tone's, which
are the octaves of those which naturally belong to the lengths of the
resonators. Accordingly, the resonator of eight feet in length in the
HARMONIC TRUMPET yields a note of the same pitch as that pro-
duced by the resonator of four feet long in the ordinary TRUMPET,
both stops being of 8 ft. pitch. The double-length resonators usually
commence at tenor C.
The TROMPETTE HARMONIQUE was invented by Aristide Cavaille-
Coll, and used for the first time in the Organ erected, in 1841, in
the Royal Church, Saint-Denis. One TROMPETTE HARMONIQUE was
placed in the Positif ; two in the Grand Orgue; and two in the Recit-
ficho Expressif; all of 8 ft. pitch.* These stops, voiced on compara-
* " La trompette harmonique du clavier dere"cit est, par sa puissance et par Texcellence des sons
qu'elle produit, incomparablement superieure a tout ce que Ton connalt en ce genre. Le carac-
tere tout-a-fait particulier de ses basses, lorsqu'on 1'emploie comme partie chantante accom-
pagn6e des jeux de fond, est d'un effet admirable. On en peut dire autant des series de jeux de
fldtes harmoniques qui donnent a 1'ensemble de i'orgue tant de rondeur et de puissance." J.
Adrian de La Page, in Report to the Societe des Beaux- Arts, on the Organ in the Royal Church
Saint-Denis, 1844.
162 ORGAN-STOPS
tively low-pressure wind, were widely different in tone from the
corresponding stops subsequently developed by Willis and other
leading English organ-builders, who adopted suitable high wind
pressures to produce the rich tone desired.
TONE AND REGISTRATION. The tone of the HARMONIC TRUM-
PET varies considerably, according to the method in which it is
voiced and the pressure of the wind on which it speaks. The tones
of the French stops are, like the generality of the French striking-
reed stops, somewhat hard and brassy, due to the thinness and
peculiar curvature of their tongues, combined with wind of moder-
ate pressure. On the other hand, the HARMONIC TRUMPETS made
by the Willis school of reed-voicers, in which thick tongues are used,
beautifully curved so as to strike the echalote with a smooth uncurv-
ing motion, yield, under high-pressure wind, full rich tones, which,
practically free from objectionable brassy clang, are imitative of the
tones of the orchestral Slide Trumpet played by a master.
The office of so powerfully- voiced a stop as the HARMONIC TRUM-
PET is necessarily limited in artistic registration. Its chief value is
in solo passages of an orchestral character, and when suitably
accompanied. It is also suitable in very full combinations in which
it assumes the dominating tonality, assisted, perhaps, by its true
Octave the HARMONIC CLARION, 4 FT., or the TUBA CLARION, 4 FT.
Powerful as its voice is and naturally rich in harmonics, it can be
effectively colored by the addition of loud-toned harmonic-cor-
roborating stops, simple and compound. The HARMONIC TRUMPET
should never be placed in an unexpressive division of an Organ.
HARMONIC TUBA. The most powerful lingual stop, of 8 ft.
pitch, inserted in the Organ. Its pipes are of large scale, having
resonators of inverted conical shape and double the normal standard
length. The reeds are formed similar to, but slightly larger in scale
than, those of the HARMONIC TRUMPET (q. .). The stop is voiced
on wind of fifteen inches upward, according to the volume and power
of the tone required. Reeds with double tongues have been sug-
gested for this impressive stop; but they are too troublesome to
make, too uncertain in speech, and too difficult to tune, ever to
favor their adoption: and it is questionable if they are desirable; for
sufficient musical noise can be produced by the single-tongued reeds
for all legitimate effects in dignified music.
TONE AND REGISTRATION. The proper tone of the HARMONIC
TUBA is full, sonorous, and commanding; dominating all the voices
of the 8 ft. stops, labial and lingual, in the Organ. Its use is, accord-
THEIR ARTISTIC REGISTRATION 163
ingly, limited to rare and very special effects, chiefly of an orchestral
character, and to grand climaxes in which a "full organ" burst of
sound is called for. The stop is, therefore, only necessary and appro-
priate in Concert-room Organs of the first magnitude. Although
stops of this dominating assertiveness have in several noteworthy
cases been thoughtlessly not to use a stronger expression planted
in exposed and uncontrollable positions, it is surely unnecessary to
insist, on both artistic and common-sense grounds, that the HAR-
MONIC TUBA be placed under control and in an expressive division
of the Organ. Its place is either in the Solo Organ or, better still, in
the Brass-wind division of the properly schemed Concert-room
Organ. Generally, the remarks anent the registration of the HAR-
MONIC TRUMPET apply in the case of the stop under review, although
the HARMONIC TRUMPET must be considered to be more generally
useful than the more powerful HARMONIC TUBA.
HARMONIC TWELFTH. A covered labial stop, of 2^ ft.
pitch, the pipes of which are of metal, of medium scale, and stopped
so as to have speaking lengths equal to one and one-half times the
lengths of the standard open pipes of the corresponding pitch. Thus,
the CC pipe of the HARMONIC TWELFTH, having a speaking length
of 4 ft., yields, on being properly voiced and overblown, a tone of
the same pitch as that of an open pipe of 2% ft. speaking length.
The HARMONIC TWELFTH was introduced by Thomas Casson,
and inserted in the Organ installed in the London Organ School and
certain other instruments constructed, under his supervision, by
the Positive Organ Company, of London. It has been made by some
other builders.
TONE AND REGISTRATION. The tone of the stop is properly full,
smooth, and clear; and owing to its having comparatively few upper
partial tones of an assertive character it mixes well, and is more
agreeable in combination than the open TWELFTH, 2^3 FT., which
belongs to the foundation harmonic series. It is on this account less
prominent, while it is more effective as a timbre-creating stop in
combination with both labial and lingual stops of the softer class.
It may, accordingly, be considered more valuable in artistic regis-
tration than any open-toned TWELFTH. In fact, it can be used in
such registration ten times for once that the TWELFTH of the foun-
dation-work can be used. Its proper position is in a soft-toned divi-
sion of the Organ ; and it is specially to be recommended for insertion
in the chief accompanimental division of the properly appointed
Church Organ.
1 64 xjKGAN-STOPS
HARMONIEFLOTE, Ger. The name given to an open labial
stop, of 8 ft. pitch, the pipes of which are of wood and of small scale,
narrow and deep, voiced on wind of low pressure. In the best
examples the tone is soft and singing, in character between the
voices of the HARMONICA and the MELODIA, while it is less assertive
than either. Desirable as a stop of this tonality would be when re-
finement is aimed at, it is not likely to be favored during the present
prevailing craze for loud and crude intonation. Such a flute-toned
stop would be ideal for a properly appointed Chamber Organ. We
are not aware of a single example obtaining in an English or Ameri-
can Organ.
HARP. Commonly, in modern Organs, a percussion stop, of 8
ft. pitch, formed of metal or wood plates or bars, suspended over
tuned resonators, and struck by a hammer-action, electrically com-
manded by a manual clavier. The compass of the stop is usually of
49 notes CC to c 3 but is properly made of the full manual com-
pass, 6 1 notes. Different opinions obtain respecting the most desir-
able material for the resonant bars; some maintaining that the Harp
tone is better imitated by sonorous wood than metal. The latter,
however, is generally used. To produce the desired tone, the
hammers have to be very carefully padded, and the action artis-
tically adjusted.
In 1733, Thomas Schwarbrook inserted in his Organ in the
Church of St. Michael, Coventry, three remarkable string stops
HARP, LUTE, and DULCIMER. But owing to the difficulty of keeping
the strings in tune, the stops were removed in 1763. Some attempts
have been made to furnish the modern Organ with a string HARP,
sounded by plectra.
HAUTBOIS D'AMOUR, Fr. Ital., OBOE D'AMORE. A lingual
stop of the OBOE species, the pipes of which have resonators of small
scale, partly covered so as to impart to the tone a singular softness
and refinement which, in all likelihood, suggested its name.* This
stop seems to be no longer made, being displaced, along with other
softly-toned stops, by those of more powerful and less beautiful
voices. The stop should be revived for insertion in the true Cham-
ber Organ, and in the softest and accompanimental divisions of the
Church and Concert-room Organs.
* ' ' HAUTBOIS D'AMOUR. Hat mit cler vorhergehenden Stimme viel Aehrvliches, doch muss sie
wie die Vox humana halb gedeckt sein. Man findet sei zu 8 Fusston, und kleiner darf sie ihrer
Natur nach auch nicht sein." Sclilimbach, 1845.
THEIR ARTISTIC REGISTRATION 165
HAUTBOY.- Fr., HAUTBOIS. Ital. and Ger., OBOE. A lingual
stop, of 8 ft. pitch, the tone of which imitates, more or less closely,
that of the orchestral instrument of the same name. The resonators
of the ordinary HAUTBOY are slender and of standard length, and
are surmounted by long inverted conical bells, which are sometimes
shaded. The variety of the stop designated the ORCHESTRAL OBOE
has been made with resonators of different forms and proportions,
with the view of obtaining a close imitation of the tone of the or-
chestral instrument. In England and America the HAUTBOY is
invariably a striking-reed; but in France and Germany the free-
reed stop seems to have been preferred. An example exists on the
Second Manual of Ladegast's fine Organ in Schwerin Cathedral.
For description and illustration of the striking-reed stops, see OBOE
and ORCHESTRAL OBOE.
HELLFLOTE, HELLPFEIFE, Ger. This name, which signi-
fies Clear-toned Flute, has been given by German organ-builders to
an open labial stop, of 8 ft. pitch, the pipes of which are of wood and
of small scale, voiced to yield a clear semi-imitative flute-tone, of
medium strength, and of good mixing quality. Schlimbach does not
mention the stop; and Seidel dismisses it in these words: " HELL-
PFEIFE, ein offenes Flotenwerk zu 8' von besonders helium Ton."
For description of the corresponding English stop, see CLEAR FLUTE.
HOHLFLOTE, HOHLPFEIFE, Ger. Pr., FLUTE-CREUSE.
Dtch. HOLFLUIT, HOLPIJP. This name, which means Hollow-toned
Flute, is used to designate an open labial stop, of 8 ft., 4 ft., and,
sometimes, 2 ft. pitch, the pipes of which are of large scale, made, in
the most characteristic examples, of wood, and voiced to yield a
full, somewhat dull, and hollow tone, which has suggested its name.
Examples exist in several of the more important Walcker Organs,
including those in Riga Cathedral and the Gewandhaus, Leipzig.
In both these Organs they are of 8 ft. pitch. In the Organ in Pauls-
kirche, Frankfurt a. M., the HOHLFLOTE is of 4 FT. pitch. They are
all placed on the First Manual the proper division (the Great) in a
Church Organ. In the properly stop-apportioned Concert-room
Organ it should find a place in the Wood-wind division, and in the
subdivision devoted specially to the FLUTES.*
* "La FLOTE-CREUSE (die Hohlflcete) est une flute de grosse taille en gtoffe e' mieux en bois
pour repondre & son nom et sonner le creux du sapin par ces levres 6troitement pmcees. On la
peutfaire detoutesles hauteurs; & seize-pieds, elle est fort originate, mais aussi rare qu' deux; a
huit-pieds, elle chante avec beaucoup de melancolie; gen6ralement, on ne la melange qu'avecles
jeux effiles et tranchants." Regnier.
1 66
ORGAN-STOPS
FORMATION. The HOHLFLOTE, of wood, is made in different forms; all with the
aim of obtaining from pipes of moderate scales the maximum volume of the tone
peculiar to the stop, while speaking on a copious supply of wind at a moderate
pressure. We cannot do better than describe the formation of the pipes of two
representative stops, made and voiced by the distinguished artist, Edmund
Schulze, of Paulinzelle. The first stop, from the Organ formerly in the Town Hall
of Northampton, was formed of quadrangular pipes of greater width than depth,
with the mouth cut on a wide side. The lower portion of a pipe of this stop is
shown, in Front View and Longitudinal and Transverse Sections, in Fig. 21. The
II
FIG. 21
mouth is of the German form, cut up equal to one-half its width, having a straight
upper lip, and its side-pieces and the upper portion of the cap sloped toward the
lower lip. The second stop, from the Organ in the Church of St. Peter, Hindley,
is formed of triangular pipes of greater depth than width, with the mouth cut on
the narrow side. The lower portion of the middle c 1 pipe of this stop is shown in
Front View and Longitudinal and Transverse Sections, in Fig. 22. The peculiar
formation of the pipe is clearly indicated by the Sections. The internal dimen-
sions are I Y% inches in width at the mouth by 2 J^ inches in depth.* These meas-
urements give an internal transverse area equal to that of a quadrangular pipe of
1 5^8 inches in width by i & inches in depth. The adoption of the triangular form
is simply for the purpose of obtaining a mouth large in proportion to the transverse
area of the pipe, as in the case of the preceding example. The mouth is % inch in
height and arched as shown. HOHLFLOTE pipes have also been made of greater
depth than width, as in the stop in the Great of the Organ in the Public Halls,
Glasgow, constructed by Lewis, of London. The tenor C pipe of this stop meas-
ures 2 J/6 inches in width by 2$ inches in depth. The mouth, formed on the wide
way of the pipe, is cut up i/% inches in height, and its upper lip is thick and care-
fully rounded, as is generally the case in HOHLFLOTE pipes. The bass octave of
this stop is, as usual, of large-scaled covered pipes. Provision for a copious supply
* In speaking of the width of a pipe, allusion is invariably made to the internal dimension of
the side in which the mouth is cut. A pipe may, therefore, be described as having a greater depth
than width, or greater width than depth, according to the location of its mouth.
THEIR ARTISTIC REGISTRATION 167
of wind is necessary in the pipes of the HOHLFLOTE. The stop has in some in-
stances been made of metal ; but, in the matter of tone, it cannot be considered the
true stop when of that material.
TONE AND REGISTRATION. The proper full unimitative flute-
tone of the HOHLFLOTE, 8 FT., somewhat dull and hollow in its char-
acter, places the stop almost in a class by itself. Its recognized
place is in the Great Organ, where its voice is of the most value in
combination, imparting great firmness and tonal solidity to the uni-
FIG. 22
son foundation tone, and fullness to the voices of the lingual stops
belonging to that fundamental division. The tone being deficient
in harmonic upper partials of high pitch, and free from any of an
assertive character, prevents its assuming the office of a timbre-
creator, while it admits of its being freely used in registration as a
body-builder. To be of maximum use, the stop should be placed
under control and rendered flexible and expressive. By affording
tones of different degrees of assertiveness, its value in registration
with exposed stops will be greatly increased. In full effects it will
largely help in binding the various tonalities and the several pitches
together, admitting of a free use of the compound harmonic-cor-
roborating stops at their full tones. In the Second Pedal of the Organ
in the Cathedral of Ulm there is a HOHLFLOTE of two feet pitch.
Speaking of this stop, Locher remarks: "As a particularly rare
specimen, I found this stop in Ulm Minister, as a 2 ft. pedal stop,
where combined with other stops on the Upper Pedal, it gives, with-
out need of any coupler, a power of expression belonging almost ex-
168 ORGAN-STOPS
clusively to the manuals. To explain the term ' Upper Pedal, ' I
must say that at Ulm, as well as in the Church of St. Paul, Frank-
furt-on-the-Main, in the Marienkirche, Lubeck, and in the Stifts-
kirche, Stuttgart, there aru two pedal claviers placed one above the
other (like the manuals) instead of the customary single clavier."*
HOHLFLOTENBASS, Ger. An open labial stop, of 16 ft.
pitch, the pipes of which are of wood, quadrangular in form, and
mouthed and voiced similar to those of the HOHLFLOTE (q. #.). It
furnishes the proper bass to the manual unison stop, and is, accord-
ingly, a Pedal Organ stop; but rarely introduced in modern Organs.
HOHLQUINTE, Ger. An open labial stop, of 5^ ft., 2% ft.,
and i J^ ft. pitch, the pipes of which, in the two lower pitches, are
commonly made of wood, and, in the ij^ ft. pitch, of metal. In
formation and tone, the stop resembles the unison HOHLFLOTE. As
a tone-building harmonic-corroborating stop it would prove valu-
able in either the Great or the Solo of a large Church or Concert-
room Organ. Its tone not being highly distinctive, the stop lends
itself very freely to combinations of both labial and lingual stops.
In the Pedal Organ the HOHLQUINTE, 5% FT., finds its most use-
ful place; while those of 2% ft. and i}-^ ft. belong to the manual
department, f
HOHLSCHELLE, Ger. The term that has been employed by
German organ-builders to designate a QUINTATEN. Schlimbach
says: "HOHLSCHELLE. 1st ein veralteter Beiname der Quintaton."
From the name, which means Hollow-Bell, it might be supposed that
the tone of the stop resembled that of the HOHLFLOTE, strongly in-
clining to that of the HOHLQUINTE a compound tone that might
prove very valuable in artistic registration. Perhaps it is to be re-
* In our work, "The Art of Organ-Building," Vol. II., pp. 145, 146, we give a Section and
Plan of the double pedal claviers of the Ulm Organ, made from drawings furnished by Messrs.
Walcker, of Ludwigsburg, the builders of the Organ.
f ' ' Mais ce qu'il faut bien retenir, comme nous 1'avons de* ja fait remarquer et comme le prou-
vent encore les quintes tirdes des r6gistres etroits, c'est qu'& peu pr&s tous les jeux de f onds alle-
mands engendrentleur quinte, et que sa ge'nfc'ration par un r6gistre special donne & la quinte cous
lescaracteresdecergistre,la montre, par exemple, a sa quinte ouverte, et de mme mesure
qu'elle et de mSmes proportions.
" Cette affection des Allemands pour la quinte est logique; ils ne veulent pas accompagner
leursf onds avec des fonds seulement, parce que c'est un accompagnement sans vigueur; ils nele
feront pas non plus avec des timbres 6clatants comme les jeux d'anches; la quinte leur offreun
renforttout simple et se marie convenablement avec le chant, qui occupe une part siimportante
dans leur culte exte"rieur. Quand la quinte est & chemine'e, comme en France, c'est souventle
Nasard; elle se nomme Rohrnasad, Rohrquinl; quand elle ne s'accorde qu'avec les fonds d'un
degre'su.pe'rieur, onl'appelle souvent Quinta ex octavo.; quand elle est a la douzieme, on la nomme
en Italic Dodecima, & la dix-neuvieme, Deci ma-Nona." Regnier.
THEIR ARTISTIC REGISTRATION 169
gretted that some of the tones due to the skill of the old voicers are
no longer heard in the Organ.
HORN, ORCHESTRAL HORN. Fr., COR D'ORCHESTRE.
ItaL, CORNO. Ger., HORN. Stops named HORNS were introduced
by some of the old English organ-builders. Renatus Harris, Junr.,
erected an Organ, in 1724, in St. Dionis Backchurch, in which he
inserted a stop labeled FRENCH HORN. In 1730, Richard Bridge
inserted a similar stop in the Organ he built for Christ Church,
Spitalfields; and another, in 1741, in the instrument he erected in
the Church of St. Anne, Limehouse, London. All these stops were
doubtless designed to imitate in their tones those of the Horn of the
time, which was first used in England, in 1720, at the performance,
in London, of Handel's opera <{ Radamisto." What success attended
the organ-builders' imitations can only be surmised.
The orchestral Horn of to-day produces tones open and hand-
closed which are sui generis, and particularly difficult to imitate
by organ-pipes, lingual or labial. Yet essays have been made by
skillful pipe-makers and voicers, chiefly with lingual pipes, which
have been attended with such good results that there are hopes of a
satisfactory imitation some day rewarding their labors. The tones
of the orchestral Horn are so peculiarly smooth and tender, that we
have long held the opinion, supported by effects that have come
before our notice, that there is a likelihood of a good imitation of the
Horn tones being produced by metal labial pipes. At present, how-
ever, no serious attempt has been made in this direction.
Neither the French nor German organ-builders seem to have
considered the beautiful tones of the orchestral Horn either capable
or worthy of imitation; and, accordingly, have made a great mistake.
In all the lists of stops of Walcker Organs we have examined, we
have only found HORN, 8 FT., given once, in that of the Concert-
room Organ in the Philharmonie, Warschau. In this instrument it
is inserted, along with the OBOE, 8 FT., in the soft-toned Third
Manual division, where it is expressive. Two HORNS, of 8 ft. pitch,
were inserted by the celebrated German organ-builder, Edmund
Schulze, in the Organ in the Parish Church of Doncaster, but neither
of these can be considered imitative. In the stop-appointments of
French Organs we have been unable to find a single instance of the
introduction of the imitative HORN or COR D'ORCHESTRE. It,
accordingly, seems to have been left to English and American artists
in reed- voicing to develop the imitative HORN to a state approach-
ing all that can reasonably be expected of lingual organ-pipes.
170 ORGAN-STOPS
FORMATION. In certain attempts made, during the latter part of the last
century, by English organ-builders to produce tones resembling those of the or-
chestral Horn, lingual pipes were used having resonators of inverted conical form,
like those of the TRUMPET, but usually of a larger scale; in which some device of a
muting character, such as a perforated disc of metal, was inserted so as to obtain
tones having the peculiar tonality of the closed or "hand notes " of the orchestral
instrument, which are absolutely free from brassiness or reediness. In other stops,
the resonators were shaded to obtain the desired subdued tones. Such expedients
proved only partly successful, and merely pointed the way to more effective
methods, involving the adoption of resonators of widely different forms and pro-
portions : of these two representative examples will be sufficient, both due to Ameri-
can ingenuity and skill.
The most successful HORN which has come under our notice is that recently
produced by the Hook and Hastings Company, organ-builders, of Kendal Green,
Mass., through whose courtesy we are able to describe and illustrate the formation
of the pipes of the stop. It will be seen, on referring to the illustration, Pig. 23,
Plate III., that the resonator employed is mainly of the inverted conical form, but
differing from all other resonators of the form, in being closed at top with a conical
cap, soldered on, and having a slot, with double adjustment for regulating and
tuning, cut close to its upper end, as shown. It can be readily understood that the
peculiar cap exerts a considerable mellowing effect on the tone in conjunction with
the adjustable slot through which alone the subdued sound finds free egress. The
slot is properly placed in a line with the tuning-wire, but it is so placed in the illus-
tration that both it and the wire can be properly shown. The measurements of
the pipe are as follows: The length of the resonator, exclusive of the conical cap, is
2 feet 3% inches; diameter at top 3% inches; height of cap internally %inch.
Length of slot (subject to alteration) 2 rg- inches; width & inch; distance from top
edge of resonator i J^ inches. The measurements of the sound-producing portions
are : The length of the reed (echalote) from the under side of the block is i % inches,
the width of the tongue (languette) at its free end is &k inch, and where it enters
the block & inch. The reed is of the closed form, its perforation commencing &
inch from the lower end.
The second noteworthy example is the HORN which Mr. E. M. Skinner, organ-
builder of Boston, Mass., added some years ago to the list of imitative stops.
The formation and proportions of the pipes of this stop forcibly illustrate by
what very dissimilar treatments similar results are obtained. In this case it may
be stated that the resonator, instead of being of large scale and after the TRUM-
PET model, is extremely slender and, accordingly, very slightly tapered. Also
unlike that of the other HORN, it is open at top, and has a short cylindrical por-
tion carrying an adjustable slide shaded by a partially attached disc of metal.
The measurements of the tenor C pipe are as follows: The length of the res-
onator is 3 feet 3}^ inches (subject to slight alteration), and its internal diameter
at top is 1.63 inches. The measurements of the sound-producing portions are:
The length of the reed from the under side of the block is 2 3^- inches; the width of
the tongue at its free end is ^ inch, and where it enters the block <& inch. The
reed is of the closed form, its perforation being about i & inches long, commencing
about y% inch from the lower end. Of examples of less importance it is unnecessary
to speak. As a specially refined tone, entirely free from clang or brassiness, is
required, it is necessary that a suitable wind-pressure be adopted for the imitative
HORN. This may vary under certain conditions; but, in our opinion, it should
THEIR ARTISTIC REGISTRATION 171
never exceed twelve inches. The stop represented by the pipe illustrated in Fig.
23 was wisely voiced on wind of eight inches, which should never be exceeded in a
Church Organ HORN.
TONE AND REGISTRATION. The closer the tone of the organ
HORN approaches the tones of the orchestral instrument the greater
will its value be in solo work and in registration. But, it may reason-
ably be asked, to which of the characteristic tones of the orchestral
Horn is it most desirable for the tone of the organ stop to approach?
The two tones of the Horn are the open and dosed; the latter being
also designated the hand-tones, because they are formed by inserting
the hand, in a special manner, and more or less, into the bell of the
instrument while it is being played. These tones are widely different
in sonorousness, and call for great skill on the part of the performer
to reconcile them.* It would seem most desirable for the voicer of
the HORN to aim at obtaining a tone as free from brassy clang as
possible in striking-reed pipes, and approaching closely the tender
sympathetic tonality of what are known as the " half -stopped notes"
of the orchestral Horn. Such a tonality would give the HORN, 8 FT.,
an individuality of peculiar charm and value both in solo effects and
in refined registration; separating its voice from the voices of all the
other lingual stops in the Organ; and placing itself, as it were, mid-
way between them and the voices of the organ-toned labial stops.
With such an intermediate tonality it is not difficult to realize the
unique position the imitative HORN would occupy in registration
with the more refined and contrasting stops of the Organ. To even
such a distinctive stop as the Vox HUMANA it would impart volume
and dignity of peculiar value in a tenor solo. In combination with
the FLUTES it would create tones of beautiful colorings. In the
properly-appointed Concert-room Organ its correct place is in the
Brass-wind division ; while in the Church Organ its appropriate place
is in an expressive accompanimental division.
HORNBASSLEIN, Ger. The name given, according to Schlim-
bach (1843), to a Pedal Organ stop, of 2 ft. pitch, formed of open
metal pipes of medium scale. The stop is represented in the Pedal
Organ of Walcker's instrument in the Cathedral of St. Stephen,
Vienna, under the name OCTAVBASS, 2 FT. While stops of this high
* Dr. W. H. Stone remarks: "Between the stopped or 'hand notes ' and the open notes there
is an obvious difference in character and quality which it is impossible wholly to suppress, but
which may be sufficiently modified so as not to offend the ear. This object is attained by blowing
the open notes softly, so as to reduce the contrast between their sonorousness, and the closed or
'stuffed' (&ouff) character of those modified by means of the hand." Grove's " Dictionary of
Music."
172 ORGAN-STOPS
pitch are common in German Pedal Organs, they are practically
unknown as independent stops in French, English, and American
Pedal Organs. They should, however, appear both in the full COM-
PENSATING MIXTURE, and the Pedal GRAND CORNET.
HORN DIAPASON. The name that has been used to designate
a metal stop, of 8 ft. pitch, the pipes of which resemble in formation
and scaling those of the true DIAPASON, but which are modified in
tone by being boldly slotted and, necessarily, slightly increased in
length. The effect of the slotting is the introduction of certain har-
monic upper partials into the pure organ-tone belonging to the
normal DIAPASON, changing it into a horny and somewhat stringy
quality which fails to satisfy the sensitive musical ear. The value of
the tone is, in our opinion, not sufficient to warrant its insertion in
any save a Concert-room Organ of the first magnitude, in which it
may be valuable in building up a full volume of foundation tone
through the absence of tonal sympathy with the pure DIAPASON
voices. With this view, it may occupy a place in the First or Great
Organ. Messrs. Walker & Sons, organ-builders, of London, appear
to favor the stop, having introduced it in several of their important
instruments. They inserted a HORN DIAPASON, 8 FT., in the Greats
of the Organs in St. Martin's Church, Leicester, and the Church of
St. Mary-le-Bow, London; and in the Swells of the Organs in York
Minster; and Holy Trinity Church, Chelsea, London. We inserted
the stop in the Third Organ, or Wood-wind division, in our scheme
for the Organ installed in the Festival Hall of the Louisiana Purchase
Exposition, 1904, chiefly on account of its value in contrasting
combinations.
HORNLEIN, Ger. The name employed, according to Schlim-
bach (1843), to designate a manual stop, of 2 ft. pitch, which may
be either of the GEMSHORN or the NACHTHORN tonality. Such a stop,
both as a harmonic-corroborator and timbre-creator, would be
found valuable in artistic registration, furnishing a vivid contrast
to the PICCOLO, 2 FT. The term HORNLEIN was applied, in the orig-
inal Organ in the Cathedral of Lucerne, to a soft-toned lingual stop
of 8 ft. pitch.
HUMANGEDECKT, Ger. A covered stop, of 8 ft. pitch, the
soft and compound tone of which seemed to imitate a refined human
voice.* The pipes were of similar formation to those of the LIEB-
* "Human heifst so viel als lieblich, dar HUMANGEDECKT: ein liebliches, angenehmes, ge-
decktes Register. Man findet es gewohnlich zu 8 Pusston." Seidel.
THEIR ARTISTIC REGISTRATION 173
LICHGEDECKT, 8 FT., but of somewhat smaller scale, and voiced on
wind of lower pressure.
A stop of this description would be a valuable addition to the
tonal forces of the true Chamber Organ, and also to the tonal appor-
tionment of the Choir or chief accompaniment al division of the
Church Organ, in which great refinement of tone is essential.
J
JEU ERARD, Fr. The name given to a free-reed stop, invent-
ed by rard, of Paris, and inserted by him in the Organ erected in the
Tuileries, now destroyed. Like all properly-constructed free-reed
stops, the JEU ERARD was furnished with very large boots. Its res-
onators were in the form of a short inverted cone, surmounted by a
hemispherical capping, having a perforation, where the two forms
joined, for the escape of the condensed air and the emission of the
sound. The tone of this stop is stated to have been agreeable but
rather muffled, as might be expected from a free-reed under such a
resonator. The stop is now obsolete. See REGAL.
JUBALFLOTE, Ger. An open labial stop of 8 ft. and 4 ft.
pitch, the pipes of which have double mouths, and yield a full un-
imitative flute-tone.* A JUBALFLOTE, 8 FT., exists in Walcker's
Organ in St. Paul's Church, Frankfurt. It also exists, of both 8 ft.
and 4 ft. pitch, in the Pedal of the Organ in the Church of SS. Peter
and Paul, Goerlitz. Hopkins gives, in his specification, a third
JUBALFLOTE, of 2 ft. pitch, but we question his accuracy in this
small matter.
JULA, Ger. The name given, according to Schlimbach and
Seidel, to the SPITZFLOTE, 8 FT., but for what reason neither au-
thority gives any information. The term JULAQUINTE has, after the
same fashion, been used to designate a similar stop of either 5^ ft.
or 2^ ft. pitch.
JUNFERNREGAL, Ger. An old lingual stop, which was
usually made of 8 ft. and 4 ft. pitch; but an example of 1 6 ft. pitch
exists, or did exist about fifty years ago, in the Choir of the Organ
in the Church of St. Dominick, Prague. Like all old REGALS, the
stop is one of the tonal curiosities of the art of organ-building. See
REGAL.
* "JUBALFLOTE, ein Bezeichnung fur ein offenes Flotenwerk 8' oder 4', auch mit doppelter
Labiierung, kommt im deutschen und amerikanischen Orgelbau vor. Der Name ist entnommert
aus I, Mos. 4, 21." Topfer-Allihn.
174 ORGAN-STOPS
JUNGFERNSTIME, Ger. Lat., Vox VIRGINE. The term
which has occasionally been employed by German organ-builders
to designate an open labial stop , the pipes of which were cylindrical,
of small scale, and properly made of tin in the best examples. The
stop, voiced on wind of low pressure, yielded a tone of an extremely
refined and delicate character, probably closely resembling that of
the Vox ANGELICA, 8 FT. , of to-day. It was usually and properly
made of unison (8 ft.) pitch, but an OCTAVE seems to have sometimes
been made.
K
KALBERREGAL, Ger. An obsolete lingual stop of the old
REGAL family, which received its remarkable name from the subdued
and lowing character of its voice, which somewhat resembled that
of a calf (KalV). See REGAL.
KERAULOPHONE, Grk. This name, compounded from the
Greek words ^epag a horn, auXos a pipe or flute, and cpcovq voice
or sound, is used to designate an open metal labial stop, of 8 ft.
pitch. The stop was originally made by Gray & Davison, organ-
builders, of London; and was introduced by them, for the first time,
in the Organ they constructed, in 1843, for the Church of St. Paul,
Knightsbridge, London. The stop is common in English Organs,
but has very seldom been made by Continental builders : an example
appears in the Positif of Merklin's Organ in the Church of Saint-
Eustache, Paris. We have been unable to find an example in any
Organ built by Walcker, Ladegast, or any other great German organ-
builder: but Locher states that a KERAULOPHONE was inserted by
Steinmeyer, in 1880, in the Organ in the Frauenkirche, Munich.
Roosevelt, of New York, showed his wise appreciation of the value
of this stop by inserting it in some of his important Organs. He
placed it in the Great of his Organ in Grace Church, and in the Solo
of the Organ in the Church of St. Thomas, New York City; in the
Echo of the Organs in the Auditorium, Chicago; the Cathedral of
The Incarnation, Garden City, L. L, and the First Congregational
Church, Great Barrington, Mass. English organ-builders are, to-
day, very unwisely abandoning the stop.
Although the name of the stop under consideration has com-
monly been rendered KERAULOPHON, it seems clear to us that, as its
last syllable is derived from the word cpwv-iq, it should be rendered
KERAULOPHONE ; agreeing with many other words which derive their
THEIR ARTISTIC REGISTRATION 175
terminations from the same Greek word such as EUPHONE, STEN-
TORPHONE, DIAPHONE, graphophone, microphone, telephone, etc.
The French term is KERAULOPHONE. Roosevelt invariably used the
final E in his stop names.
FORMATION. The pipes of the KERAULOPHONE are cylindrical in form and of
medium scale. The desirable maximum scale in the ratio i : 2.66 halving on the
eighteenth pipe gives the CC pipe a diameter of 3.94 inches; the C pipe a di-
ameter of 2.41 inches; the c 1 pipe a diameter of 1.47 inches, and the c 4 pipe a diame-
ter of 0.34 inches. Smaller scales have been adopted by different builders, and
are desirable when the stop is destined for an Echo Organ or a Chamber Organ:
but too small a scale will destroy the characteristic tone of the stop, which should
be round and rich. The mouth of the KERAULOPHONE pipe should be one-fifth
the circumference of the pipe in width, and about one-fourth its width in height.
This latter proportion, however, depends upon the wind-pressure which should
not exceed 3% inches and the quality of the tone desired. The upper lip is
straight and not cut sharp, being smoothly rounded, and the nicking of the
languid is moderately fine. The mouth has ears of small projection and without
any harmonic attachment. The characteristic feature of the pipe, and that which
is the principal factor in the production of its special tone, is its perforated tuning-
slide, as shown at A in Fig. 24, Plate III. The length of the slide is about two and a
half times its own diameter ; and its perforation is made the distance of one diame-
ter from its top edge, as indicated. In the CC pipe the diameter of the perforation
should be 0.79 inch; in the C pipe 0.56; and regularly diminishing to 0.14 inch in
the c 4 pipe. The slides must be so accurately fitted to the bodies of the pipes as to
firmly retain their position, while they can be easily tapped up or down in
the process of tuning. Metal of good substance must be used for this stop; and
Hoyt's Two-ply Pipe-Metal is highly suitable, being specially firm at moderate
thicknesses.
TONE AND REGISTRATION. The tone of the KERAULOPHONE
when in its true form and artistically voiced is full, smooth, and in
the principal portion of its compass so strongly resembling the tone
of the "hand notes " of the orchestral Horn as to have suggested its
name.* In this respect the KERAULOPHONE may be said to stand
alone among labial stops : and its present neglect by English organ-
builders goes far to show how little they appreciate refinement of
tone, and the great value of timbre-creating voices in the Organ.
There is not sufficient roar or scream in the KERAULOPHONE to please
the present prevailing want of taste in the organ-building world.
We can speak, from long experience, of the great value of the
true English KERAULOPHONE in refined and artistic registration.
Its building-up and vivid-coloring properties are remarkable; and
in this direction it is specially valuable in association with both open
and covered, unimitative, flute-work. It combines admirably with
* In the KERAULOPHONE in our own Chamber Organ this horn-like tone was so pronounced
that we labeled it CORNO DI CACCIA. It spoke on wind of 2 % inches.
1 76 ORGAN-STOPS
all the softer lingual stops; and would form an effective helper to the
ORCHESTRAL HORN. The proper situation of the stop is in the true
accompanimental divisions of the Church and Concert-room Organs.
It would also be very valuable in the Wood-wind division of the
properly-apportioned Concert-room instrument.
KINURA. The name, derived from the Greek word /avupa
Harp; and given by Hope- Jones to a lingual stop, of 8 ft. pitch,
somewhat resembling a poor OBOE in tone, and, accordingly, having
nothing to recommend it for adoption in any class of Organ. It was,
however, introduced by him in the Organ installed in McEwan Hall,
Edinburgh; and in a few other instruments.
KLEINGEDECKT, Ger. The name given to a covered stop
of 8 ft. and 4 ft. pitch, of small scale, and yielding a delicate unimita-
tive flute-tone. This stop, according to Regnier, is of metal and of
4 ft. pitch only, and of small scale; but this is not altogether correct,
for in the Choir of the Organ in the Church of St. Michael, Hamburg,
there is a KLEINGEDECKT, 8 FT., of wood probably a diminutive in
scale and voice of the usual LIEBLICHGEDECKT, 8 FT. Examples, of
4 ft. pitch, exist in the Organs in the Cathedral of Ulm and the Town
Church of Fulda.
The prefix klein small, has been employed by German organ-
builders in the names of other organ-stops; we, accordingly, find the
following: KLEINCIMBEL, KLEINFLOTE, KLEINFLOTENBASS, KLEIN-
REGAL, and KLEINTERZ (i % FT.). The prefix is employed to indicate
either octave pitch or small scale.
KLEINPRINZIPAL, Ger. Literally Small Principal. The
term which has occasionally been employed to designate the OCTAVE,
4 FT., of pure organ -tone, the scale of which is adjusted from that of
the MAJORPRINZIPAL, 8 FT. This octave stop is identical with the
old English PRINCIPAL, 4 FT., and the HALBPRINZIPAL of the old
German organ-builders. See OCTAVE.
KLEINREGAL, Ger. An obsolete stop of the old REGAL
family, which derived its name from the small size of its pipes of 4 ft.
pitch. There seems to have been nothing special about its tone.
See REGAL.
KNOPFREGAL, Ger. Literally Knob-Regal An obsolete
lingual stop, of 8 ft. pitch, which, as in the case of several other old
REGALS, derived its name from the peculiar form of the resonators
of its pipes. The resonator consisted of a short cylindrical body or
tube surmounted by a globular head, across which a narrow slit or
THEIR ARTISTIC REGISTRATION 177
opening was cut for the escape of the pipe-wind and the emission
of the sound. The head in some examples assumed a pear-shape or
some other bulbous form. See REGAL.
KOPFREGAL, Ger. One of the curious and now obsolete
lingual stops, which, like several others of the REGAL family, de-
rived its name from the form of the resonators of its pipes. The
usual form of the resonator was that of a short body surmounted by
a headpiece in the shape of two truncated cones joined together at
their bases ; a form which, considerably modified, has been followed
in the resonators of the free-reed COR ANGLAIS. This form, however,
was not invariably adopted in the old KOPFREGAL of the early Ger-
man organ-builders. See REGAL.
KOPPEL, Ger. The term employed by old German organ-
builders to designate a labial stop, usually of the flute-toned family,
and of different pitches; peculiarly suitable, on account of its un-
pronounced tone and good mixing quality, for. combination or cou-
pling with almost any other stop in the Organ. Another somewhat
useful office of the KOPPEL was that of a "helper, "commonly em-
ployed to help or cover the slow speech of certain stops, notably the
GERMAN GAMBA (g. v.). This valuable stop in the old Organs was
sometimes labeled KOPPELFLOTE when of . ft. pitch, and KOPPEL-
OCTAVE or KOPPELOKTAV when of 4 ft. pitch. Seidel remarks: " It
is a common labial stop, covered, of 4 ft., 8 ft., or 16 ft. pitch, and in
some very few cases it is open, like the HOHLFLOTE. It belongs to
the manual department. The 16 ft. stop is termed GROSS-KOPPEL;
the 8 ft. stop is found under the name COPULA MAIOR; and the 4 ft.
stop under the name COPULA MINOR. Some call the GEMSHORN a
KOPPELFLOTE. The KOPPEL is sometimes a description of QUINT,
5> ft., 2% ft, and i^ ft., and in this case it is open, like the HOHL-
FLOTE. KOPPEL means also a variety of MIXTURE, of two or three
ranks. There is a KOPPEL of three ranks in the Pedal of the Organ
in the Church of St. Dominick, Prague, composed of a TWELFTH,
2^ FT., a FIFTEENTH, 2 FT., and a SEVENTEENTH, i% FT."
TONE AND REGISTRATION. Although such a stop as the German
KOPPEL is not required in its office of helper in the Organ of to-day,
it is just worth while considering the use of a unison stop, preferably
of wood, having a pure organ-tone of medium strength, for the pur-
pose of coupling with, and imparting desirable smoothness and body
to certain labial and lingual stops especially the latter. Such a
stop, which might appropriately be called a body -builder, would be
extremely valuable in artistic registration, in combination with the
1 78 ORGAN-STOPS
more cutting string-toned stops and such lingual stops as the CLARI-
NET, CORNO DI BASSETTO, FAGOTTO, and Vox HUMANA. With the
last-named it would go far to remove the prominence of the objec-
tionable nasal twang so commonly and unfortunately characteristic
of the stop.*
KURZFLOTE, Ger. Pr., FLUTE COURTE. Literally Short
Flute. A cylindrical metal labial stop, of medium scale, and 4 ft.
pitch, yielding an unimitative flute-tone of an agreeable quality.
An example exists in the Echo of the Organ in the Cathedral of
Lund, Sweden.
KUTZIALFLOTE, Ger. This, according to Wolfram (1815), is
a small-scaled, flute-toned stop, of 4 ft. and 2 ft. pitch. According
to both Seidel and Hamel, it is an open stop of 4 ft,, 2 ft., and I ft.
pitch; while it is sometimes met with of 1^3 ft. pitch. The pipes of
the KUTZIALFLOTE, 4 FT., have been properly made of wood, while
those of the stops of higher pitch are of metal. A KUTZIALFLOTE,
i FT., exists in the Great of the Church of St. Dominick, Prague;
and one, of ij^ ft. pitch, in the Organ in Kreuzkirche, Dresden.
Beyond its affording examples of the employment of complete stops
of such high pitches, there is nothing further calling for special
comment.
LARIGOT, Fr. Eng., NINETEENTH. Ital, DECIMA NONA.
An open cylindrical metal stop, of ij^ ft. pitch in the manual de-
partment and 2%j ft. in the Pedal Organ, f The pipes are properly of
small scale; but, when forming the independent stop, are invariably
voiced too loud and piercing. The fact seems to be ignored that the
* While performing on the Organ in St. George's Hall, Liverpool, the great master of the
Organ, the late W. T. Best, invariably used a body-giving stop along with the Vox HUMANA,
f "LARIGOT (from an old French word Varigot, for a small flute or flageolet, now obsolete),
the old name for a rank of small open metal pipes, the longest of which is only i ) ft. speaking
length. . . . It is first met with, in English Organs, in those made by Harris, who passed many
years in France, and who placed one in his instrument in St. Sepulchre's, Snow Hill (London),
erected in 1670. " E. J. H., in "A Dictionary of Music and Musicians."
' ' Le LARIGOT est la quinte de la Doublelte,p&r consequent 1'octave sup6rieure du Nasard,$t la
superoctave du Gros-Nasard. On le fait, ou plutdt on le faisait de grosse taille et d'<$toffe: seize
pouces, ou Quarante-trois centimetres, & son premier tuyau. On le placait d'ordinaire au positif ,
It cause de son peu de hauteur; mais il est tombS en de"sue"tude dans la facture franchise oft Ton
ne s'est pas suffisamment pntr6 de la ne"cessit6 d'assembler toujours les trois degr6s de Thar-
roonie pour 1'avoir complete, quatre, huit et seize-pieds, et dans ceux-ci, les deux et demi, cinq
et dix-pieds. Quand les trois degre"s de quinte 6taient tir6s avec les trois, quatre et cinq degree de
sons toniques, 1'effet ce qu'il parait en etait si pergant qu'on a encore gard6 dans la conversation
1'expression vulgaire de jouer & Tire-Larigot, pour signifier un vacarme solennel." Regnier,
THEIR ARTISTIC REGISTRATION 179
LARIGOT is a mutation stop belonging, in its different pitches, to an
8 ft. or 1 6 ft. foundation harmonic series; and, to be of true value as
a harmonic-corroborating stop should be subordinated in tone to the
prime unison tone of the series to which it belongs. The LARIGOT,
i% FT., seldom appears as a complete and independent stop, but
examples exist in the Grand Chceur of the Organ of the Cathedral of
Notre-Dame, and in the Positif of the Organ in the Church of Saint-
Sulpice, Paris. There is much to be said in favor of this practice of
introducing the LARIGOT as an independent stop. For further
particulars see NINETEENTH.
LIEBLICHFLOTE, Ger. The name sometimes used to desig-
nate a small-scaled labial stop, formed of either wood or metal.
The name refers to the tone only, which, freely rendered, signifies
Lovely-toned Flute. The pipes are invariably covered, and form the
correct OCTAVE LIEBLICHGEDECKT. If properly made and artis-
tically voiced, the wood stop is to be preferred to that of metal,
although fine examples in the latter material have been produced.
So long as zinc or inferior pipe-metal can be used, the wood stop is
not likely to be favored by organ-builders.
TONE AND REGISTRATION. The true LIEBLICHFLOTE, 4 FT.,
yields a soft unimitative flute-tone free from any prominent har-
monics: accordingly, as an octave stop it is extremely valuable in
registration, brightening and enriching almost all classes of unison
(8 ft.) tone, without impairing their individual colorings. As a rule,
there are too few stops of 4 ft. pitch introduced in modern Organs;
for, as they corroborate the first and principal upper partial tone of
all open labial stops of 8 ft. pitch, their value is unquestionable and
their importance only second to that of the unison stops. The LIEB-
LICHFLOTE should be introduced in the manual division in which the
LIEBLICHGEDECKTS of 1 6 ft. and 8 ft. pitch are placed, practically
completing that valuable family of covered stops. It may be re-
marked, that as the stops of 16 ft. and 8 ft. pitch will properly be
made of wood, it may be desirable to secure a somewhat lighter and
brighter tone by having the LIEBLICHFLOTE made of a high-grade
alloy or Hoyt's hard-rolled Two-ply Pipe-Metal.
LIEBLICHBORDUN, Ger. A covered wood stop, of 16 ft.
pitch, and small scale; the pipes of which are constructed on the
BOURDON model, and voiced, on wind of moderate pressure, to
produce a soft and good mixing quality of tone, in which the second
upper partial is slightly in evidence. In this respect differing from
the purer-toned LIEBLICHGEDECKT. A good example exists in the
i8o ORGAN-STOPS
Great of the Schulze Organ in the Church of St. Peter, Hindley,
Lancashire. The CCC pipe of this stop measures, internally, 3%
inches by 5 inches, with a mouth 3^ inches in height.
LIEBLICHGEDECKT, Ger. The name appropriately given
by the German organ-builders to a covered labial stop, of small
scale, and 16 ft. and 8 ft. pitch. The larger stop is usually and
properly made of wood throughout; while the stop of 8 ft. pitch,
which should, preferably, be made of wood throughout, has in some
examples its two or three higher octaves made of metal. On the
Fourth Manual of the Ladegast Organ in the Cathedral of Schwerin,
the LIEBLTCHGEDECKT is formed of rnetal throughout.* Both the
stops belong to the manual divisions of the Organ. Instances of the
insertion of the stops of both pitches in the same division obtain on
the Second Manuals of the Organs in Christuskirche, Aachen, and
Predigerkirche, Erfurt. The stop of 16 ft. pitch is inserted on the
Second Manual and that of 8 ft. pitch on the Third Manual of the
Organ in St. Petrikirche, Lubeck. We have not been able to find a
single instance of the insertion of the LIEBLICHGEDECKT, 16 FT., in
the Pedal of a German or other Continental Organ.
It is strange that the value of a soft- toned unison stop in the
Pedal Organ has been so systematically overlooked by all Continen-
tal organ-builders and organists. It would seem that the desirability
for refinement in Pedal Organ tone never entered their brains; yet
on artistic grounds alone its necessity must be obvious to everyone
endowed with musical sense and taste. It is not too much to say
that the presence of such a stop as the LIEBLICHGEDECKT, 16 FT., is
imperative in all well-appointed Pedal Organs; unless its insertion
is not absolutely necessary through the presence of a DULCIANA,
1 6 FT. In the Pedal of a Concert-room Organ, or a Church Organ of
any pretensions to proper tonal appointment, both stops should
certainly be present, and from these, by extension, may be derived
valuable OCTAVES, 8 FT.
FORMATION AND SCALE. In general formation the pipes of the LIEBLICHGE-
DECKT differ in no essential from those of the BOURDON, the only distinction lying
in scale, proportions of mouth, and treatment in voicing. The LIEBLICHGEDECKTS
of different German organ-builders vary in scale and in the proportions of depth
to width, the latter, carrying the mouth, being of the most importance. In the
softer toned stops the pipes are narrow in proportion to their depth ; while in the
louder voiced stops the pipes approach the square. The most satisfactory propor-
tion being as two to three. This is practically the proportion adopted by Edmund
* It is described thus: "LIEBLICHGEDBCKT 8', aus I2l6th, Zinn; Mensur 4^; cbenfalls von
zarter Intonation, das schwachste Gedeckt in der Orgel."
THEIR ARTISTIC REGISTRATION 181
Schulze, of Paulinzelle, for the LIEBLICHGEDECKT, 8 FT., in the Choir of his Organ
in the Church. of St. Peter, Hindley, England, the CC pipe of which measures,
internally, 2}^ inches in width by aJi inches in depth. The following scale, in
inches, to the ratio i : 2.66 -halving on the eighteenth pipe is practically that
adopted by Schulze:
SCALE OF LIEBLICHGEDECKT, 8 FT. RATIO 1:2.66
PIPES CC C c 1 c a cs c
WIDTH 2.13 1.3! O.8O 0.49 0.30 O.2O
'DEPTH 3.08 1.89 I.I5 0.71 0.43 0.27
As the pipes in the two high octaves are very small, it will be generally deemed
desirable to form them of metal in the usual shape. It is not in these octaves that
the individuality of the stop obtains. For insertion in a large Concert-room Organ,
the following slightly larger scale may be found desirable:
SCALE OF LIEBLICHGEDECKT, 8 FT. RATIO 1:2.66.
PIPES CC C c 1 c a cs c<
WIDTH 2.95 1.81 i. ii 0.68 0.45 0.26
DEPTH 3.94 2-41 1.47 0.91 0.55 0,34
Regarding the LIEBLICHGEDECKT, 16 FT., suitable for insertion in a manual
division, different opinions obtain among organ-builders on the question of scale.
It is certain, however, that a small scale is desirable: and when it is inserted along
with the LIEBLICHGEDECKT, 8 FT., in the same division, it should be of a somewhat
smaller scale than the unison stop, so as to be voiced slightly subordinate in tone.
Dr. Hopkins, in his work, "The Organ," gives a scale for a Choir or Swell LIEB-
LICHGEDECKT, 1 6 FT., apparently of German origin. The CCC pipe is 3-^ inches
in width by 5 inches in depth; the CC pipe is 2 ^ inches in width by 3 inches in
depth ; and the C pipe is i % inches in width by 1 7 /% inches in depth. The scale
is irregular, not being developed on any standard ratio. It approaches most closely
to the ratio i : 2.519, halving on the nineteenth pipe. The following scale will be
suitable for a Pedal Organ stop of 16 ft. pitch:
SCALE OF PEDAL LIEBLICHGEDECKT, 16 FT., RATIO 1:2.519.
PIPES CCC CC C G
WIDTH 3.79 2.39 1.50 1.15
DEPTH 5.36 3.38 2.13 1.62
Of equal importance to the scale is the form of the mouth; and that, in its pro-
portions of height to width, differs considerably from the mouths of the generality
of wood pipes belonging to other stops of the flute-toned family. The height of
the mouth is an important factor in the production of the pure organ flute-tone of
the true LIEBLICHGEDECKT. In no case should it be less than half its width in
height; while it may with advantage exceed its width in height, as in the stop in
the Swell of the Organ in St. Peter's, Hindley, the CC pipe of which has a mouth
2% inches wide and 2}^ inches high. The lower portion of this pipe, in correct
proportions, is shown in Front View and Section in the accompanying illustration,
Fig. 25. The mouth of the CC pipe of the Choir stop is 2}^ inches wide and iH
inches high. In the smaller scales, mouths ranging from three-quarters to their
1 82
ORGAN-STOPS
entire width in height may be used with advantage tonally. The thickness of the
upper lip is another factor in the production of satisfactory tone. This may vary
in the CC pipe, from a quarter to half an inch (the thicker lip producing the
smoother tone), and be cut square or have carefully rounded edges: and the lip
may be straight, as in the illustration, or arched. Pipes having mouths of so
great a height in proportion to width require a copious supply of wind, desirably
of moderate pressure, for their proper speech. The manual stop should speak on
wind of 33/2 inches and the pedal stop on wind of from 4 inches to 5 inches.
PIG. 25
TONE AND REGISTRATION The tone of the properly made and
artistically voiced LIEBLICHGEDECKT is very beautiful, being of a
singularly pure organ flute quality almost free from its first upper
partial tone. It may be accepted as the most refined tone produced
by the covered stops of the Organ; while it is equal to the finest
tones of the half-covered stops. The purity of its tone renders it
extremely valuable in simple combination and artistic registration,,
chiefly on account of its perfect mixing quality, and its forming so
fine a background for the display, so to speak, of pronounced tone-
colors furnished by stops having imitative voices. The value, from a
registration point of view, of the entire family of the LIEBLICH-
GEDECKTS, 1 6 ft., 8 ft., sJ^ ft., and 4 ft., has not been properly, if at
all, recognized by organ experts and designers; and their special
value when grouped in the same division has been altogether un-
realized. The LIEBLICHGEDECKT, 5^ FT., may be omitted should a
stop of the same pitch and of the flute-toned class be considered
THEIR ARTISTIC REGISTRATION 183
more desirable in the stop apportionment of the division.* But we
strongly recommend retaining the complete family.
LIEBLICHGESCHALLT. The name, derived from the
German, given by Hope-Jones to a small-scaled metal LIEBLICH-
GEDECKT, 8 FT., voiced to yield a very soft tone, as if it was an echo
of that of the normal full-toned LIEBLICHGEDECKT. Otherwise of no
special interest.
LIEBLICHQUINTE, Ger. The covered labial stop, of 5^ ft.
pitch, belonging tc the LIEBLICHGEDECKT family; the pipes of which
are formed in all respects similar to those of the LIEBLICHGEDECKT
(q. .)
LITICE. Lat., LITUUS. "A curved brass trumpet, clarion,
used by cavalry " (Nail). The name that has been used by old
German organ-builders to designate a lingual stop yielding a piercing
tone. It is described by Seidel: " ' Litice oder Lituus ist einerlei mit
Zink, Krummhorn oder Cornett." Schlimbach agrees with this
definition. Now obsolete.
LLENO, Span. The name used by Spanish organ-builders, as a
general appellation for all compound harmonic-corroborating stops
or MIXTURES.
LUTE. The name given to an organ-stop formed of strings, in
all probability sounded by plectora or "jacks, 77 after the method
followed in the Harpsichord. The stop was introduced by Thomas
Schwarbrook, organ-builder, of Warwick, in his masterpiece, the
Organ in the Church of St. Michael, Coventry, constructed in 1733.
Dr. Rimbault says: "This noble instrument cost 1400. It origi-
nally contained three remarkable stops the HARP, LUTE, and DUL-
CIMER ; but, in consequence of the 4 difficulty of keeping the strings
in tune,' they were removed in 1763." History is repeating itself:
after the lapse of more than a century and a half the HARP stop is
again appearing in the Organ, and, perhaps, the Lute and the Dulci-
mer will find their imitations in the Organ of the Twentieth Century.
M
MAJORBASS, Ger. The name that has sometimes been used
to designate the Pedal Organ covered labial stop, of 32 ft. pitch, as
* The stop apportionment here alluded to will be found fully developed in the First Expres-
sive Subdivision of the Second Organ, in our scheme for the tonal appointment of the Concert-
room Organ, given in "The Organ of the Twentieth Century," page 303.
1 84 ORGAN-STOPS
in the Organ in the Church of St. Elizabeth, Breslau. The stop has
been better known as the UNTERSATZ, 32 FT., as in the Organ in the
Cathedral of Schwerin. Walcker has employed the term GRAND
BOURDON, 32 FT., in the Organs in the Cathedrals of Ulm and Frank-
furt a. M. : and this term is appropriate for the stop in our own
Organs to-day. We find the stop under the name CONTRA-BOURDON
in certain English and American Organs.
MANUALUNTERSATZ, Ger. The name to be found in cer-
tain important Organs constructed by Walcker, of Ludwigsburg,
designating a manual covered labial stop, of 32 ft. pitch. This very
grave stop is never carried below tenor C, chiefly on account of the
great size of the pipes composing its bottom octave, but also to avoid
giving undesirable gravity and dullness to the manual bass. Ex-
amples of the MANUALUNTERSATZ, 16 FT., under that name, exist in
the First Manual divisions of the Organs in the Cathedrals of Ulm
and Vienna. This grave stop has been seldom introduced in English
Organs; examples exist in the Greats of the Organs in the Parish
Churches of Leeds and Doncaster, labeled SUB-BOURDON, 32 FT.
Under the name CONTRA-BOURDON, 32 FT., Hill has inserted it in
the Great of the Organ in the Centennial Hall, Sydney, N. S. W.
All these stops commence at tenor C. We have not been able to
find a single example in a French Organ.
TONE AND REGISTRATION. It is very questionable if such a stop
as the MANUALUNTERSATZ is necessary in any Church Organ, es-
pecially if it is of the ordinary BOURDON quality of tone. If, however,
it should be deemed desirable in a very large instrument, it should
certainly be of the LIEBLICHGEDECKT family. The heavy droning
tone of the ordinary BOURDON should be avoided. In the Concert-
room Organ of the first magnitude such a stop, in its most desirable
form, may be introduced in the foundation division First or Great
Organ but there an' open stop of pure organ-tone is greatly to be
preferred. To meet this demand, without entailing the necessity of
cumbering the division with a rank of such large pipes, a soft-unison
Pedal stop, preferably a DULCIANA, 16 FT., may be borrowed and
made to speak on the manual clavier from tenor C. In our scheme
for the Concert-room Organ of the Twentieth Century, we have
suggested adding an Auxiliary Stop, of full compass, under the
name DOLCIANO PROFUNDO, 32 FT., derived from the CONTRA-
DULCIANA, 32 FT., of the Pedal Organ, necessarily extended to 61
notes.* It is quite obvious that a stop of this soft and pure organ-
* See "The Organ of the Twentieth Century, " pp. 297 and 319.
THEIR ARTISTIC REGISTRATION 185
tone would be productive, in registration with the double and unison
foundation stops of the First Organ, of many tonal effects of a dig-
nity and grandeur never yet heard in organ music.
MEERFLOTE, Ger. The name that has been given to the
stop commonly designated UNDA MARTS (q. z>.).
MELODIA. An open labial stop, of 8 ft. pitch, having an un-
imitative flute-tone of a smooth and singing quality, in certain
examples inclining slightly to a horn-like intonation. This fine stop,
in its proper form, may be said to be unknown in English organ-
building; its closest representative being the WALDFLOTE (q. v.},
many fine examples of which exist in English Organs. In the soft-
toned Fourth Manual of the Walcker Organ in the Cathedral of
Riga, the stop exists under the name MELODICA, 8 FT. An example
exists in the Choir of the Hook and Hastings Organ in the Music Hall,
Cincinnati, 0. ; the tone of which is described as "round, rich, and
mellow. " A MELODIA, 8 FT., of a pure and beautiful tone, existed in
the Choir of the Organ constructed by the Hutchings-Votey Com-
pany for Woolsey Hall, Yale University. The organ-builders who
seem at the present time to realize most clearly the value of the
MELODIA are MM. Casavant Frres, of St. Hyacinthe, P. Q. Ex-
amples exist in the Choirs of their Organs in Emmanuel Church,
Boston, and the Second Church of West Newton, Mass.; and in
the Positif of the Organ in the Church of Notre-Dame, Montreal,
Canada; where it bears the French name MELODIE.
FORMATION AND SCALE. The pipes forming the true MELODIA are of wood
and quadrangular: white pine being used for their sides and backs, and some close-
grained hard wood, such as mahogany or maple, for their fronts from tenor C to
the top note. The pipes of the bass octave may have fronts of pine with hard
wood mouth pieces. The mouths are of the inverted form, and are cut up about
one-third their width in height. The caps to be of hard wood, hollowed, and the
wind-way formed in them, and to be set below the under lip of the mouth just
sufficiently to produce perfect intonation. The block is set about half the internal
width of the pipe below the under lip. The scale of the stop varies in different
examples, but the following may be accepted as suitable:
SCALE OP MELODIA, 8 FT., IN INCHES RATIO i : 2.66.
PIPES CC C c 1 c* cs c*
WIDTH 3.34 2,05 1.25 0.77 0.47 0.29
DEPTH 4.44 2.73 1.67 1.02 0.63 0.38
It is a common practice to insert the bass octave in covered pipes, so as to save
money; but this practice is contrary to the canons of artistic and correct organ-
building, one of which is: Each stop in the Organ must be carried throughout its
compass in pipes of its own class and tonality.
1 86 ORGAN-STOPS
TONE AND REGISTRATION. When artistically voiced, on wind of
low pressure, the tone of the true MELODIA is very beautiful: as
before stated, it is of an unimitative flute quality, smooth and sing-
ing, inclining in the tenor and middle octaves to a horn tonality.
This latter is largely due to the smooth speech of the pipes; just as
in the case of the KERAULOPHONE. The appropriate place for the
MELODIA is in the Choir or accompanimental division of the Church
Organ, and in the softest-toned division of the Concert-room Organ.*
It is a perfect stop for the true Chamber Organ.
It is hardly necessary to point out that a stop of this beautiful
and refined tonality is invaluable in artistic registration. Perfect in
mixing quality and capable of receiving tonal coloring of any class,
it furnishes an admirable foundation for the most delicate labial and
lingual combinations. It is a perfect body-giver to the Vox HU-
MANA, imparting roundness and fullness where it is so greatly required.
MENSCHENSTIMME, Ger. The term employed by the old
German organ-builders to designate a lingual stop which, in its
peculiar tone, imitated in some respects the human voice while
singing. See Vox HUMANA.
MESSINGREGAL, Ger. An obsolete lingual stop, the tone of
which was strongly suggestive of the clang of brass wind instruments.
See REGAL.
MITTELFLOTE, Ger. This term, which has neither reference
to formation nor tonality, has been employed to designate a FLUTE,
of 4 ft. pitch, simply because it occupied a middle position between
FLUTES of 8 ft. and 2 ft. pitch in the same division of the Organ.
This name is a good example of the vagueness and senselessness one
too frequently meets with in organ-stop nomenclature.
MITTELGEDECKT, Ger. The term that has been used by
old German organ-builders to indicate a GEDECKT which occupies a
middle position between two other stops of the same family, placed
in the same division of an Organ. Schlimbach says: "Wenn z. B.
auf einen Clavier Gedact 8, 4, und 2 Fusston sich zugleich befindet,
so ist das von 4 Fusston das Mittelgedact."
MIXTURE. Lat., MISCELLA. Ger., MIXTUR. Dtch., Mix-
TUUR. ItaL, RIPIENO. Span. LLENO. The generic name for all
*In our tonal scheme for the Concert-room Organ, the MELODIA, 8 FT., is placed in the
Ancillary Aerial Organ, where it speaks on wind of ij^ inches. It is, accordingly, available
for registration purposes on any manual clavier. See " The Organ of the Twentieth Century.' '
Pages 329 and 331.
THEIR ARTISTIC REGISTRATION 187
compound harmonic-corroborating stops, applied generally to such
stops as are composed of octave- and fifth-sounding ranks of ordi-
nary open metal labial pipes, and to such compound stops as do not
call for any special name indicative of peculiar composition, forma-
tion, or intonation. The MIXTURE may be composed of two or
more ranks of pipes, according to the tonal structure or appoint-
ment of the division of the Organ in which it is placed. The different
pitches of the separate ranks and their breaks are dictated by that
tonal structure, and the number and nature of the single harmonic-
corroborating stops that are included in the divisional appointment.
The presence of a MIXTURE, of suitable size and appropriate
composition and tonality, is essential in every important division of
an Organ pedal and manual in which a proper scientific and ar-
tistic tonal structure or effective stop apportionment is essayed.
This fact has, of late years, been systematically neglected or ignored
by unscientific and inartistic organ-builders: and almost completely
unrealized or misunderstood by organists, who have been content to
accept the opinion of self-interested organ-builders on the subject.
It is quite easy to see on what grounds organ-builders hate all com-
pound harmonic-corroborating stops; and it is proper that some
light be let in upon the matter.
In the first place, the willingness of musicians to allow the elimi-
nation of MIXTURES in the tonal appointment of Organs has been
largely, if not entirely, due to the very crude, unscientific, and
highly objectionable form in which organ-builders have commonly
constructed, voiced, and regulated them. Such stops, producing
loud screaming voices, more like the noise of smashing glass than
musical sounds, entirely at variance with their scientific and tonal
office in the economy of the Temple of Tone, are certainly better
absent than present in any Organ. It seems questionable if or-
ganists have given sufficient study to the subject to realize the great
value, both from a scientific and musical point of view, of a correctly
proportioned and artistically voiced and regulated MIXTURE, in
effective registration, and in the production of the richest tones the
Organ can furnish.
In the second place, organ-builders have, during recent times,
strongly advocated the elimination of compound harmonic-cor-
roborating stops, not on any contention based on scientific or musical
grounds, which they were in no form able to advance; but because
the formation of such complex stops are undesirably troublesome;
involving a considerable amount of scientific knowledge, and what to
them seems an unprofitable expenditure of high-class skilled labor
188 ORGAN-STOPS
in their formation, artistic voicing, and scientific tonal graduation.
In short, the organ-builder fails to see sufficient remuneration, in
dollars, for such stops; and relying on the apathy, if not the want of
knowledge, on the part of the organist, he cries, "Away with all
MIXTURES!"
In old Organs the introduction of compound stops was carried
to an extent altogether unwarranted by either scientific or artistic
demands. Two instances may be given : in the Great of the Organ
in the Old Church, Amsterdam, finished in 1686, there are eighteen
ranks, and also in the Choir eighteen ranks of MIXTURE work ; and
in the Organ in the Monastery Church, at Weingarten, finished in
1750, there are no fewer than ninety-five ranks of MIXTURE work,
two stops alone having twenty and twenty-one ranks, respectively.
Such excessive apportionment out of all reason on any grounds
has naturally led to a swing in the opposite direction, which has
ended, as we now see, in a too sparing introduction, or frequently in
the entire omission, of MIXTURES ; creating a starvation in the true
and the most characteristic tones of the Organ, a fact that is greatly
to be regretted.
The largest compound harmonic-corroborating stop we have
examined and know of in a modern Organ is the HARMONIC MIX-
TURE, of fourteen ranks (784 pipes), in the Organ in the Edinburgh
University. Turning to examples in modern German organ-build-
ing, and selecting two representative instruments by Walcker, of
Ludwigsburg, we find in the Organ in the Cathedral of Riga an
instrument of 124 speaking stops on the First Manual, fifteen
ranks of MIXTURE work; on the Second Manual, ten ranks; on the
Third Manual, four ranks; on the Fourth Manual, three ranks; and
in the Pedal Organ, twelve ranks, including the GRAND BOURDON,
V. RANKS. In all, forty-four ranks. In the Organ in the Cathedral
of Ulm an instrument of 107 speaking stops on the First Manual,
twenty-one ranks of MIXTURE work; on the Second Manual, eleven
ranks; on the Third Manual, five ranks; and in the Pedal Organ,
five ranks. In all, forty-two ranks. No such apportionments of
MIXTURE work are to be found in modern Organs built in any other
country. In the Organs constructed by Cavaille-Coll we find an
adequate amount of MIXTURE work provided: for instance, in the
Organ in the Church of Saint-Sulpice, Paris, there are in the Grand-
Choeur (portion of the Grand-Orgue) , nineteen ranks; in the Positif,
six ranks; in the Recit, fourteen ranks; and in the Solo, five ranks.
This is the largest apportionment of MIXTURE work we have been
able to find in a French Organ.
THEIR ARTISTIC REGISTRATION 189
FORMATION. The formation of a MIXTURE, properly proportioned to the stop-
appointment of the division of the Organ in which it is placed, calls for a certain
amount of scientific knowledge and a keen appreciation of the value of musical
sounds. The prevailing "rule-of-thumb" methods are invariably attended by
failure.
A properly formed MIXTURE comprises different ranks of pipes of high pitches ;
which, as they cannot be carried complete throughout the compass of the division
in which the stop is placed, have to be divided into two or more portions, tech-
nically termed breaks; each of which, in each rank, commences on a pipe of a lower
pitch, so as to allow the stop to be carried throughout the compass of the division.
This arrangement is shown in the following example a MIXTURE, of five ranks,
composed of octaves and quints, breaking on each octave of the compass ;
MIXTURE V. RANKS.
RANKS I. II. III. IV. V.
BREAK I. CC tO BB . . 19 22 26 29 33.
2. C to B . . 15 19 22 26 29.
3. C 1 tO b r . . 12 15 19 22 26.
4. c 2 to b 2 . . 8 12 15 19 22.
5- C3 tO C4 . . j 8 12 15 1 9 .
All the sounds produced by the pipes, in the several breaks and ranks, serve to
corroborate the harmonic upper partial tones of the unison sounds, with which
they combine, throughout the compass of the clavier. Accordingly, to everyone
conversant with the phenomena of compound musical sounds, the value of, and,
indeed, the necessity for the introduction of compound harmonic-corroborating
stops represented by MIXTURES of proper formation in the Organ must be
evident and incontestable.
It is well known to all who have devoted any serious study to the phenomena
of musical sounds, that in those produced by the cultivated human voice and the
string instruments of the orchestra, the prime tones are accompanied by a great
number of upper partials or harmonic over-tones; and that these sensibly decrease
in strength as they rise in pitch, until they become inaudible. This important
natural law has to be recognized in the proper voicing and regulating of the MIX-
TURES and all the harmonic-corroborating stops in the Organ. It is not too much to
say, and our extended observation supports the assertion, that this acoustical
phenomenon has been seriously, if not altogether, ignored by all organ-builders
who have hitherto constructed MIXTURES: and, accordingly, stops so unscien-
tifically constructed have outraged the natural laws of musical sounds, and have
destroyed, instead of enriching and beautifying, the voices of the Organ with which
they were combined.
TONE AND REGISTRATION. The tone of the MIXTURE varies ac-
cording to the nature of the pipes of which it is composed. It may be
of pure organ-tone, belonging to the foundation-work, corroborat-
ing the harmonic upper partial tones of the prime tones of the DIA-
PASONS; and infusing into them rich compound tones, so satisfying
to the cultivated musical ear. ^Or the MIXTURE may be so construct-
ed of pipes of special tonalities as to become both harmonic-cor-
190 ORGAN-STOPS
roborating and timbre-creating; finding its proper place in different
tonal divisions, according to their stop-apportionments.
In artistic registration, a properly constructed, voiced, and re-
gulated MIXTURE is of the greatest value; entering into combina-
tions of infinite variety, either in its office of harmonic corroborator
or timbre-creator, and imparting to them elements of singular
brilliancy and beauty all but unknown in organ music to-day. We
have practically proved that a five-rank MIXTURE can be scaled,
voiced, and regulated, so as to become one of the most valuable and
generally useful stops in refined and effective registration. We have
heard beautiful passages rendered on such a MIXTURE while in com-
bination with a PICCOLO, 2 FT., only; and equally beautiful solo
passages when combined with the Vox HUMANA alone. The MIX-
TURE, in whatever division of the Organ it is placed, should be in-
closed and rendered flexible and expressive.
For further particulars respecting MIXTURES, see ACUTA, CARIL-
LON, CLARION MIXTURE, COMPENSATIONSMIXTUR, CORNET, CYMBAL,
DOLCE CORNET, DULCIANA CORNET, POURNITURE, FULL MIXTURE,
HARMONIA ^ETHERIA, PLEIN-JEU, SESQUIALTERA, and TERTIAN.
MONTRE, Fr. The term commonly used by French organ-
builders to designate such foundation and organ-toned and highly-
finished metal stops as may be mounted and displayed in the buffet
or case of an Organ. For such a purpose, the MONTRE s are usually
made of tin, highly burnished, and may be of 32 ft., 16 ft., or 8 ft.
speaking lengths, as in the Organ in the Royal Church of Saint-Denis.
Sometimes the term is applied to the PRESTANT, 4 FT., when its pipes
are mounted and displayed. All the MONTRES are most carefully
fashioned, having the boldly-formed French mouth, and being of
tin brightly burnished, produce a fine effect in combination with the
dark woodwork of the case.*
MOUNTED CORNET. The name given by the old English
* "LA MONTRE. Jeu labial ouvert, le plus ordinairement de moyenne taille, dont tous les
tuyaux ou la plupart sont en Evidence, en montre, comme les plus brillants et les plus parfaits.
Le metal, qui est ou doit etre d'etain fin, revet un poll digne de rivaliser avec les m6taux d'Ulra
ou de 1'Escurial. Les bouches de montre sont ordinairement 6cussonnees, c'est-a-dire que leurs
levies, au lieu d'etre simplement pliees comme dans le tuyau commun, sont formees de deux
pieces rapporte"es, courbees et terminees en forme d'ecusson. . .
" Nos montres frangaises ont les trois degr6s de ton, seize, hu.it, et quatre-pieds: cette der-
niere dimension, elles prennent le nom de prestant; elles vont rarement a trente-deux ou meme
vingt-quatre-pieds, a cause de la quantite" de vent, de metal, et d'espace necessaire; souvent
meme la montre de seize n'exhibe pas ses plus gros tuyaux, qui, simplement en bois, cachent leur
humble costume a 1'interieur du buffet. D'autres fois, on apergoit en montre de magnifiques
viugt-quatre et trente-deux-pieds qui n'appartiennent qu'a la pedale de 1'orgue; mais la montre
s'entend toujours des tuyaux correspondent aux claviers manuels, " Regnier.
THEIR ARTISTIC REGISTRATION 191
organ-builders to a compound stop, of short compass, which, to
economize room inside the Organ, was mounted, above the main
wind-chest, on a small wind-chest designed for its reception alone;
the wind to its pipes being conveyed from the grooves of the main
chest by conveyances one for each note of its compass. The
MOUNTED CORNET consisted usually of five ranks of large-scaled
open metal pipes, yielding a full dominating tone, though sometimes
the unison rank was inserted in covered pipes. Its ranks comprised
a DIAPASON, or a FLUTE A CHEMINEE, and an OCTAVE, TWELFTH,
'FIFTEENTH, and SEVENTEENTH. In old English Organs its compass
never extended below middle c 1 ; but in German Organs, in which
the stop held a more important office, it usually was carried down to
tenor C. Hopkins says the MOUNTED CORNET "was chiefly used for
playing out the melody of the chorals upon, and for the performance
of an obsolete kind of voluntary : but it is of great use in large Organs
in hiding the breaks in the several compound stops, as it proceeds
itself without any 'repetitions.' In Father Smith's Organs the
CORNET was never 'mounted,' but stood on the sound-board."
Schulze introduced a MOUNTED CORNET, of four ranks and tenor C
compass, in both the Great and Swell divisions of the Organ in the
Parish Church of Doncaster. There is no call for the CORNET of
short compass, in any form, in the Organ of the Twentieth Century.
MUNDFLOTE, Ger. Literally Mouth-Flute. An open labial
stop of metal, and of 2 ft. pitch; similar in all essentials to the
French FLUTE A BEC (q. v.). The stop has no distinctive flute
tonality. A stop under the name exists in the Organ in the Cathe-
dral of Konigsberg.
MUSETTE, Fr. Ger., SACKPFEIFE. The Musette was a small
Bagpipe much used in olden times by the people of different Euro-
pean countries ; and the MUSETTE of the Organ is supposed to imi-
tate the characteristic tone of that old instrument. The MUSETTE,
as made in France, is a free-reed stop, having resonators of small
scale and inverted conical form, made of tin, and yielding a bright
and pleasing tone, as might be produced by a small and weak species
of Bagpipe.* Examples of the stop have been made with striking-
* " LA MUSETTE, le CHALUMEAU (Sackpfcife, Schalmey). Jeu d'anche de forme pyramidale,
en Stain fin comme la plupart des timbres effile~s. La Musette a quatre-pieds de hauteur et huit de
ton. Le timbre, plus faible que celui du Cromorne, imite assez bien le chetif instrument dont on
lui a donne" le nom. Dom Bedos disait de la Musette: *Ce jeu est encore peu connu, dans le
royaume.' II y est tous les jours moins connu, et si dans certaines melodies populaires et rares,
to!6res par 1'feglise qui se fait toute a tous, il peut paraitre utile, il n'est jamais n4cessaire."
Regnier.
192 ORGAN-STOPS
reeds and slender cylindrical, covered, and perforated resonators,
producting refined and characteristic tones. We introduced a free-
reed MUSETTE, 4 FT., to serve as an OCTAVE to the free-reed CORNO
INGLESE, 8 FT., in the Second Organ of our scheme for the Organ
installed in the Festival Hall of the Louisiana Purchase Exposition,
1904. The MUSETTE when of its true and characteristic tonality is a
valuable voice in the properly-appointed Concert-room Organ. It
mixes well and, accordingly, enters freely into combinations of a soft
and distinctive character. It is peculiarly useful in music of a pas-
toral description.
MUSICIRGEDECKT, Ger. A name sometimes given by the
old German organ-builders to a covered stop, of 8 ft. or 4 ft. pitch,
yielding a peculiarly soft and sweet unimitative flute-tone. Schlim-
bach, a great authority on old stops, describes it thus: "MusiciR-
GEDACKT. Die Bestimmung dieses Gedacts ist: zur Begleitung der
Musik gebraucht zu werden. Man findet es zu 8 und 4 Fusston,
still intonirt. Man vergleiche Kammerflote. " This stop was prob-
ably identical with the later GELINDGEDECKT (q. v.}.
N
NACHTHORN, Ger. Fr., COR DE NUIT. Ital, PASTORITA.
In addition to the name first given, German organ -builders have
occasionally employed the term NACHTSCHALL, which signifies
Night-sound. The stops bearing these names in German and French
Organs are formed of covered or open labial pipes, usually of metal
but in some examples of wood. It is found of 8 ft. and 4 ft. pitch,
and very rarely of 2 ft. pitch. According to Wolfram (1815), the
covered NACHTHORN resembles the QUINTATEN both in formation
and tone, but its pipes are of larger scale. The open stop has pipes
resembling those of the HOHLFLOTE. The tone of the NACHTHORN
is round, soft, and agreeable; and in fine examples has, as its name
implies, a combination of flute and horn tones, which imparts in-
dividuality to the stop.* Wolfram alludes to this horn-like tonality.
*"La PASTORITA ou le NACHTHORN, cor de nuil. Cette flute de bois ou d*6toffe, tantdt
bouche'e. tant6t ouverte, mais toujours largeet trapue,se fait souvent dequatre-pieds etnedes-
cend jamais a plus de profondeur que huit; son timbre est agreable, son harmonie delicate.
Ouverte, elle se rapprodhe de la flute creuse, mais on tient alors sa taille plus e"troite et ses levres
plus senses. Bouch6e> elle imite le son du cor dans le lointain et dans le silence des nuits, d'ou
lui vient son nom; a cette fin, on lui donne une plus large taille qu'& tin des plus curieux bourdons
allemands, le Quintaton, si Ton ne veut pas qu'elle lui emprunte son principal effet. Bien des
facteurs promettent un nachthorn de quatre-pieds, qui ne tiennent pas autre chose que 1* "
Regnier.
THEIR ARTISTIC REGISTRATION 193
A representative example of the covered NACHTHORN, 8 FT., of large
scale, exists in the Positif Expressif of the Cavaille-Coll Organ in the
Albert Hall, Sheffield. One of 4 ft. pitch appears in the Pedal of the
Organ in the Garrison Church, Berlin; and one of 2 ft. pitch in the
Echo of the celebrated Haarlem Organ. Seidel says: "Im Pedal
heisst diese Stimme NACHTHORNBASS, ist sie 2 fussig, so kommt sie
zuweilen unter dem Nameh NACHTHORNCHEN vor."
NASARD, FT. Ger., NASAT, NASSART. Ital. and Span.,
NASARDO. The name given to a manual labial stop, of 2^ ft. pitch,
the pipes of which are of metal, and either open or covered. It is
the equivalent of the TWELFTH, 2% FT. in English and American
Organs, when in its open form, although it differs in tone according
to its position in the stop-appointment of the Organ. In all cases it
is a harmonic-corroborating stop, representing the nth upper par-
tial tone of the 32 ft., the 5th upper partial tone of the 16 ft., and the
2nd upper partial tone of the 8 ft. harmonic series. When inserted
in the foundation division of the Organ, it is properly formed of open
metal pipes, scaled, voiced, and regulated in strength of tone with
respect to the MAJOR PRINCIPAL or DIAPASON, 8 FT. (See TWELFTH).
When inserted in a soft-toned division of an Organ, it sometimes
assumes the form of a FLUTE A CHEMINEE. A NASARD, 2.% FT.,
exists in the Recit Expressif of the Organ in the Church of Saint-
Sulpice; and others are introduced in the Grand Orgue and the
Bombarde of the Organ in the Royal Church, Saint-Denis. The stop
frequently appears in French Organs, named QXJINTE, 2% FT.
NASARDS are to be found on the Third Manual of the Organ in the
Cathedral of Ulm, and on the First Manual of the Protestant Church
at Mulhausen. A NASAT, 2% FT., exists in the Echo of the Grand
Organ, and another in the Choir Organ in the Cathedral of Breslau.
The corresponding stop, of 5^ ft., called GROS NASARD, belongs
to the harmonic structure of the Pedal Organ. An example exists
in the Pedal of the Organ in the Royal Church, Saint-Denis. See
GROSSNASAT.
NASARD FLtTTE. The name given by its introducer, George
W. Till, of Philadelphia, Pa., to a labial dual stop, formed of a
metal FLXJTE HARMONIQUE, 8 FT., and a metal NASARD HARMONI-
QUE, 2 % FT. ; and introduced, for the first time, in the Concert Organ
in the Wanamaker Store, in Philadelphia, Pa., where it occupies a
place in the Great Organ, speaking on wind of 5 inches. This com-
pound stop is essentially timbre-creating, yielding a tone of a re-
13
194 ORGAN-STOPS
markable quality and volume, impossible of production by a stop,
however voiced, having only a single rank of pipes. As we have said
elsewhere, the existence of such a dual stop multiplies to an un-
dreamt of extent the resources of the flute-toned forces of the Organ,
when used in combination. The tone might well earn the stop the
name GRAND QUINTATEN, 8 FT.*
NASARD GAME A. The name given by its introducer, George
W. Till, of Philadelphia, Pa., to a labial dual stop, formed of a metal
GAMBA, 8 FT., having the compass of CC to c s , and a tin NASARD
GAME A, 2^ FT., of the same compass. The unison stop is voiced to
yield a powerful string-tone, so as to firmly establish the unison pitch.
The NASARD is voiced to yield a clear but a decidedly subordinate
string-tone. The combined ranks, which speak on wind of 1 5 inches,
produce a tone of wonderful richness and color-value; unlike any-
thing known to be yielded by a single-ranked string-stop. The
NASARD GAMBA was designed for, and introduced for the first time
in, the Swell of the Concert Organ in the Wanamaker Store, in Phila-
delphia, Pa.f
NASON. The name used by the old English organ-builders to
designate a covered stop of wood, of 4 ft. pitch, yielding a soft flute-
tone, inclining to that of a QUINTATEN. The stop was practically
an OCTAVE STOPPED DIAPASON. A NASON, of ij^ ft. pitch, was
inserted by Casparini in the Echo of the Organ he erected, in
1703, in the Church of SS. Peter and Paul, Gorlitz. It is probably
still in existence. The name has entirely disappeared from modern
nomenclature.
NINETEENTH. Fr., LARIGOT. Ital, DECIMA NONA. A
mutation, harmonic-corroborating stop, of ij^ ft. pitch in the
manual department, and 2^ ft. pitch in the Pedal Organ. In its
proper form, it corroborates the fifth upper partial tone of the unison
tone of the department of the Organ in which it is placed. The
NINETEENTH is invariably formed of open metal pipes, which should
be scaled in proportion to the scale of the principal unison stop with
which it is associated. When it appears as a separate stop it is
almost invariably voiced much too loud, seriously impairing its
tonal value in combination and registration. It, however, seldom
appears as a complete and separate stop : but certain old English and
* Full particulars of the formation and scales of this stop are given in "The Organ of the
Twentieth Century," pp. 104-5.
f Pull particulars of the formation and scales of the NASARD GAMBA are given in "The Organ
of the Twentieth Century.*' pp. 105-6.
THEIR ARTISTIC REGISTRATION 195
modern French organ-builders have favored it in that form. Under
the name LARIGOT, i J^ FT., it appeared as an independent stop in
the Great of the Organ built by Renatus Harris, in 1670, for the
Church of St. Sepulchre, Snow Hill, London. He also introduced
one in the Organ he built for the Church of St. Peter, Mancroft,
Norwich. The stop is inserted, in its complete form, in the Grand-
Chceur of the Organ in the Cathedral of Notre-Dame; and in the
Positif of the Organ in the Church of Saint Sulpice, Paris: both by
Cavaille-Coll.
The NINETEENTH, i^ FT., may form a complete rank in a COR-
NET; but it commonly appears, in a broken form, in the ranks of
MIXTURES. As a harmonic-corroborating stop, it should, in what-
ever form it appears, be graduated softer in tone as it rises in pitch,
in accordance with the natural laws of musical sounds.
o
OBOE, Ital. Eng., HAUTBOY. Fr., HAUTBOIS. Ger., HOBOE.
A lingual stop, of 8 ft. pitch, which belongs to the manual depart-
ment of the Organ. Two varieties of the stop are introduced in the
Organ to-day. The older and still prevailing form cannot be con-
sidered as imitative in tone; but in the finest examples of the strictly
modern forms the tones produced closely resemble those of the
orchestral Oboe. The former stop is labeled, simply, OBOE ; while
the imitative stop is properly labeled ORCHESTRAL OBOE; under
which name it will be found fully described. The OBOE in its com-
mon and unimitative form is best suited for a Church Organ in
which only one is inserted; because, unlike the ORCHESTRAL OBOE
with its somewhat thin and characteristic voice, it is an all-round
useful stop when a soft-toned lingual voice is desirable. It is gen-
erally found in Organs of any pretensions toward completeness; and,
indeed, in most small instruments.
FORMATION. The resonators of the unimitative OBOE, 8 FT., are approxi-
mately of the standard speaking lengths; and are in the form of slender tapered
tubes surmounted by long inverted conical bells, after the fashion of the conical
and belled tube of the orchestral Oboe. In its best treatment, the resonator has
its bell shaded by a disc of spotted-metal, soldered to its edge so as to allow about
one-half being bent up, to any required extent, in the process of regulating the
tone of the pipe. Shaded resonators are not invariably used but they are unques-
tionably the most desirable, for without them artistic toning is practically im-
possible. The form of the OBOE pipe just described is shown in Fig. 26, Plate III.,
which represents a pipe of a lower octave of the stop. As the pipes become shorter,
the relative proportions of their tubes and bells are largely altered; until in the top
196 ORGAN-STOPS
octave there is little difference in their respective lengths. The reeds or echalotes
of this stop are of medium scale and are invariably of the closed class, having
small triangular openings which extend from the thick discs which close their
lower ends. The tongues or languettes are correspondingly slender in form and of
medium thickness, and have a slight and finely-formed curve, preventing coarse-
ness or brassiness of intonation.
The formation just described is that of the striking-reed stop, as constructed
in this country and in England; but the OBOES of the French and German organ-
builders have frequently been made with free-reeds. An example exists on the
Second Manual of the Organ, built by Friederich Ladegast, in 1 87 1 , for the Cathe-
dral of Schwerin. Of late years, however, the French have adopted the striking-
reed. French examples are rare.
TONE AND REGISTRATION. The tone of the unimitative OBOE
varies according to the proportions of its pipes, the pressure of the
wind on which it speaks, and the method and musical sense of its
voicer : but its most desirable tone is smooth and of medium strength,
inclining, in the finer examples, to a plaintive quality, which imparts
a certain charm and impressive character to it. With such a voice,
the stop becomes extremely valuable both in solo passages and in
combination; occupying in a softly-toned division about the same
position as that held by the TRUMPET in the Great or fundamental
division of the Organ. Under such conditions the OBOE is extremely
valuable in artistic registration. It combines perfectly with both
open and covered flute-toned stops, imparting to their voices rich-
ness and dignity. It also imparts to the keen voices of the string-
toned stops, imitative and unimitative, a fullness and impressiveness
highly desirable, without seriously affecting their characteristic
voices. It is questionable if there is another lingual stop in the
modern Organ more generally useful in refined registration: of course
this last remark applies only to the OBOE produced by a master-
hand. It is remarkable that so valuable a stop should have been
systematically neglected by Cavaille-Coll. In not one of his Organs
have we been able to find an OBOE.
OBOE D'AMORE, ItaL The orchestral instrument of this
name the French, Hautbois d'Amour now almost obsolete, has a
"tone more veiled and pathetic than that of the ordinary orchestral
Oboe.'* This tone is due to the different form of its bell, which is
globular with a somewhat contracted opening, contrasting with the
flaring bell of the ordinary Oboe. This fact leads to the construction
of an organ-stop, having resonators formed with the slender tubes of
the OBOE surmounted by small bells, in the shape of two truncated
cones soldered together at their bases, after the fashion of those of
the CORNO Di BASSETTO, Fig. 8, Plate II. The addition of a stop
THEIR ARTISTIC REGISTRATION 197
producing a quality of tone between that of the unimitative OBOE
and the imitative ORCHESTRAL OBOE would be a valuable addition
to the tonal forces of the Concert-room Organ.
OBOE-FLUTE. The name given by William Hill, organ-build-
er, of London, to an open labial stop, of 4 ft. pitch, and small scale,
which yielded a soft flute-tone combined with a slight reedy intona-
tion. He introduced it in the Organ he installed in Worcester Cathe-
dral. It was probably not considered sufficiently distinctive to
recommend its general adoption, for it was disused. There is room,
however, in the Organ of the present century for a timbre-creating
OCTAVE, in the voice of which flute- and reed-tones are in full com-
bination. Ample proof obtains that such a desirable stop could be
made, wood pipes being used.
OBOE-HORN. The name given by Hope-Jones to a stop de-
veloped by him from the common unimitative OBOE, by imparting
to it a broader and somewhat hornlike intonation. As a compromise
between the OBOE and HORN, without the advantage of either, such
a stop has very little to recommend it. An example exists in the
Organ in Llandaff Cathedral, Wales.
OCARINA. An open metal labial stop, of 4 ft. pitch, which
yields a hollow fluty tone, resembling that of the instruments of the
same name.* The special pipes of this purely modern and uncom-
mon stop commence at tenor C and extend to the top note. The
pipes have cylindrical bodies surmounted by long and slightly
spreading bells, the proportions of which may be realized from the
following dimensions of the c 2 pipe. The speaking length of this
pipe is 63/s inches; the cylindrical body being 3 j/ inches long and
1.13 inches in diameter; while the bell is 3 inches long and the diame-
ter of its open end is 1 .69 inches. The mouth is 7 /s inch in width, and
about one-fourth its width in height : its upper lip is straight, and the
languid is finely nicked. The pipe is tuned by means of a strip cut
from a small slot in the belL We obtained these particulars from the
stop in the Choir of the Organ in the Church of St. Mary, Bradford,
Yorkshire, constructed by M. C. Annessens, of Gramont, Belgium.
There is a stop of the same name in the Positif of the Organ, built
by A. Amezua, in 1903, for Seville Cathedral.
* " OCARINE (It.}. A series of seven musical instruments [of different pitches] made of terra
cotta pierced with small holes [and having whistle-like mouthpieces], invented by a company of
performers calling themselves the Mountaineers of the Apennines. With these instruments,
which are of a soft and sweet, yet 'traveling' quality of tone, operatic melodies with simply-
harmonized accompaniments were given." "A Dictionary of Musical Terms. " Stainer and
Barrett.
I 9 8 ORGAN-STOPS
OCTAVE. Get., OCTAV. Ital., OTTAVA. Span., OCTAVA.
Dtch., OCTAAF. The name properly used by German, Italian,
Spanish, Dutch, American, and to some extent by English and
French organ-builders, to distinguish the chief stop of octave or
4 ft. pitch in a manual division, and of 8 ft. pitch in the Pedal Organ.
The French organ-builders commonly use the term PRESTANT for
the stop, of 4 ft. pitch, which belongs to the foundation-work; occa-
sionally applying the term OCTAVE, of the same pitch, to some sub-
ordinate stop. In this relationship we find both terms in the stop-
apportionment of the Grand-Orgue in Cavaille-Coll's fine instru-
ment in the Cathedral of Notre-Dame, Paris. In the P6dale of the
same Organ, we find the term OCTAVE illogically applied to a stop of
4 ft. pitch. In this department such a stop is strictly a SUPER-
OCTAVE or FIFTEENTH. But organ-builders are not distinguished
for their regard to correct stop nomenclature.
The old English organ-builders invariably used the term PRINCI-
PAL, 4 FT.; and the practice has been largely followed up to the
present time. Proof of this may be readily found. In a list of speci-
fications of thirty-three Organs, constructed by J. W. Walker &
Sons, of London, between the years 1858 and 1904, we find the term
PRINCIPAL, 4 FT., used fifty-nine times in the manual divisions, and
PRINCIPAL, 8 FT., twelve times in the Pedal Organ: whereas the
term OCTAVE occurs only eight times, chiefly in the Organ in York
Minster, constructed in 1903. At the present time American organ-
builders wisely and almost invariably use the term OCTAVE.
As the OCTAVE, 4 FT., is the leading stop of medium pitch in the
manual foundation-work, it is commonly and properly the one first
correctly pitched and tuned, the other stops below and above it in
pitch being tuned from it. It was probably on this account that the
old English builders gave the name PRINCIPAL to the stop. There
seems to be no other apology possible for so illogical an appellation.
FORMATION. The OCTAVE, 4 FT., which strictly belongs to the foundation-
work of the Organ, is composed of open cylindrical metal pipes, formed in all re-
spects similar to the pipes of the DIAPASON, to which they are tonally related. In
old English work the scale of the CC (4 ft.) pipe was commonly made one pipe
smaller than the tenor C (4 ft.) pipe of the DIAPASON, 8 FT. In modern work it has
frequently been made two pipes smaller in scale. Unscientifically voiced as the
OCTAVE commonly is, a smaller scale would seem desirable. When the DIAPASON
scale is of the ratio i : 2.66 (See DIAPASON), halving on the eighteenth pipe,
it may be desirable to have the scale of the OCTAVE of the ratio i : I/ 87 halving
on the seventeenth pipe.
TONE AND REGISTRATION. The OCTAVE is essentially and prop-
THEIR ARTISTIC REGISTRATION 199
erly a harmonic-corroborating stop, and its chief office is to establish
the first and most important upper partial of the tone produced by
the fundamental unison. Accordingly, while its voice should be of
pure organ-tone, similar to that of the DIAPASON, it must be suffi-
ciently soft to intimately combine with and enrich the unison tone
without disturbing its pitch. A properly proportioned OCTAVE is
of great importance and value in registration ; combining with other
labial stops than the DIAPASONS, and also with lingual stops, im-
parting to their voices clearness and brilliancy.
In a division of the Organ in which there is no DIAPASON, the
OCTAVE should partake of the character of the principal labial unison
or be suitable for combination with it in artistic registration. In this
case the OCTAVE will generally serve as a solo stop, a timbre-creating,
and a harmonic-corroborating one.
The term OCTAVE has been frequently applied as a prefix to the
names of stops, to indicate their pitch with relation to the unison
pitch of the divisions in which they are placed, or with relation to
some musical interval calculated from that unison. Accordingly,
we find stops bearing the compound names: OCTAVE DULCIANA,
OCTAVE FIFTEENTH, i FT., OCTAVE FLUTE, OCTAVE GAMBA, OCTAVE
QUINT, 2% FT., OCTAVE TWELFTH, i J^ FT., etc.
OCTAVE OBOE. The lingual stop, of 4 ft. pitch, bearing
this name has been inserted in certain important modern Organs.
It is the proper OCTAVE of the unimitative OBOE, 8 FT. An example
exists on the Second Manual of the Organ in the Cathedral of Riga;
and examples exist in the Choir and Solo of the Organ in the Cen-
tennial Hall, Sydney, N. S. W. Hutchings inserted one, under the
name OBOE CLARION, 4 FT., in the Chancel Swell of the Organ he
installed in the Church of St. Bartholomew, New York City. We
inserted one in the First Subdivision of the Third Organ repre-
senting the wood-wind forces of the orchestra in our tonal scheme
of the Organ installed in the Festival Hall of the Louisiana Purchase
Exposition, 1904.
Although the introduction of such a soft-toned stop as the
OCTAVE OBOE is, or should be, is extremely rare, there can be no
question regarding the value of its voice in expressive registration.
The present prevailing craze for musical noise is not favorable to
the introduction of refined voices in the Organ, or to the develop-
ment of artistic registration.
OCTAVE VIOLA. An open metal labial stop, of 4 ft. pitch,
the tone of which is intended to imitate that of the orchestral Viola.
200 ORGAN-STOPS
An example exists in the Swell of the Willis Organ in St. George's
Hall, Liverpool. Another, labeled VIOLA, 4-FT., exists in the Great
of the same instrument, where it serves as the OCTAVE to the VIO-
LONCELLO, 8 FT. Neither stop is highly imitative.
Imitative string-toned stops, of 4 ft. pitch, are extremely rare;
but their value in tonal-coloring and artistic registration must be
evident to everyone who has studied the combinations of organ-
stops scientifically and practically. The imitative stop is more
appropriately termed VIOLETTA (q. v.). As a harmonic-corroborat-
ing and timbre-creating stop it should find a place in every Concert-
room Organ; preferably in the String division, ancillary or otherwise,
where it forms the most perfect OCTAVE.
OCTAVIN, Fr. The term sometimes employed by French
organ-builders to designate an open metal labial stop, of 2 ft. pitch;
equivalent, in all essentials, to the SUPER-OCTAVE or FIFTEENTH in
English and American Organs. Examples of the OCTAVIN exist in
the Recit Expressif of the Organs in the Cathedral of Notre-Dame
and the Church of Saint-Sulpice, Paris.
OCTAVIN HARMONIQUE, Fr. The name given to an open
labial stop, of 2 ft. pitch, all the pipes of which are of double the
standard speaking length, and voiced harmonically; examples of
which exist in the Recit-ficho Expressif of the Organ in the Royal
Church, Saint-Denis, and in the Organ in the Church of Saint-
Vincent de Paul, Paris. The stop is flute-toned, and is practically a
HARMONIC PICCOLO.
OFFENBASS, Ger. The name that has been used by certain
German organ-builders to designate an open bass stop simply in
contradistinction to the covered stop GEDECKTBASS. It is an open
appellation, conveying no idea of pitch or tonality. Without the
addition of some expressive qualification such a term must be con-
sidered practically valueless.
OFFENFLOTE, Ger. The name given by German organ-
builders to an open wood stop, of 8 ft. or 4 ft. pitch, yielding an
unimitative flute-tone closely resembling that of the English CLARA-
BELLA. The stop obtains, under the simple name FLOTE, in the
Organ in the Cathedral of Ulm. We find the German word off en
used in other compound stop-names, as OFFENQUINTFLOTE and
OFFENFLQTENQUINTE, meaning Open flute-toned, Quint.
OPEN DIAPASON. The term employed by English organ-
THEIR ARTISTIC REGISTRATION 201
builders from the middle of the seventeenth century to designate the
principal foundation stops both in the manual and pedal depart-
ments of the Organ. In the former it is invariably of 8 ft., and in
the latter of 16 ft. pitch; these being the standard unison pitches of
the respective departments. The prefix open was applied to dis-
tinguish the stop from the one, of the same pitch, illogically named
STOPPED DIAPASON, which is in no sense a DIAPASON, being a
covered wood stop, strictly belonging to the Flute-work of the
Organ. For full particulars see DIAPASON.
OPHICLEIDE. Fr., OPHICLEIDE. Ger., OPHICLEID. Ital.,
OFFICLEIDE. The name derived from o<c<; a serpent, and xXsfq
a key, and given to a large brass instrument of extensive compass
and powerful voice. The organ-stop to which the name is applied
is supposed to imitate the tone of the orchestral instrument. It is a
striking-reed of large scale and powerful intonation, of 8 ft. pitch
in the manual divisions, and 16 ft. pitch in the Pedal Organ. The
OPHICLEIDE was first introduced by W. Hill, of London, in the Great
of the Organ he erected in the Town Hall of Birmingham. It is
labeled GREAT OPHICLEIDE, 8 FT., and speaks on high-pressure wind.
In the Organ in St. George's Hall, Liverpool, there are three OPHI-
CLEIDES, of 8 ft. pitch, inserted in the Great, Swell, and Solo divi-
sions, and one, of 16 ft. pitch, in the Pedal Organ. As a proof of the
old-fashioned system of tonal appointment pervading this instru-
ment, the only OPHICLEIDE endowed with flexibility and power of
expression is that in the Swell Organ.* The stop in the Solo Organ
speaks on a wind of 22 inches pressure : it is a superb stop of its class,
but its value is unfortunately circumscribed through its voice being
beyond artistic control.
FORMATION. The pipes of the OPHICLEIDE have resonators of inverted conical
form, of full speaking length and large scale, made of thick spotted-metal or zinc.
Their reeds or echalotes are also of large scale and of the open variety. These have
tongues or languettes of hard rolled brass, thick, and carefully curved, so as to
produce a tone full and commanding, but without brassy clang; so as to resemble
as closely as practicable the full round tones of the orchestral Ophicleide.
* "Before the recent renovation [1898], the Swell Organ was the only expressive division in
the instrument; for in the year 1855 the introduction of more than a single swell-box was neither
appreciated nor understood. It is, indeed, remarkable that for upwards of forty years the Solo
Organ of fifteen stops, including four high-pressure reeds, remained uninclosed and entirely
devoid of flexibility and powers of expression; and, to our mind, it is still more remarkable that
when the swell-box was applied to the Solo Organ in 1898, the four high-pressure and very power-
ful reed stops were left uninclosed. It is not too much to say that had these noisy stops been
placed under control their utility and effectiveness would have been increased tenfold. In our
opinion, while it was right to inclose the eleven stops of the Solo Organ, it was positively bar-
barous to leave the very stops which call most loudly for tonal control in the whole Organ ab-
solutely uncontrollable. ""The Art of Organ-Building. " Vol. II., p. 730.
202 ORGAN-STOPS
REGISTRATION. Unless used under perfect control and more or
less subdued, the OPHICLEIDE is practically valueless in general
artistic registration, and even under such control it is only good in
exceptional effects. It is occasionally employed as a solo stop when
it is expressive; but its principal use and value is in impressive cres-
cendo passages and grand climaxes. Under all circumstances, a
lingual stop of so powerful a voice has to be very carefully used; and
will always be so when commanded by a true musician and artist.
It will, however, be beloved by the lover of musical noise.
ORCHESTRAL BASSOON. The name properly given to the
lingual stop, of 8 ft. pitch, the tones of which successfully imitate
those of the orchestral Bassoon. The pipes of the most satisfactory
examples of the stop have resonators of wood or metal, conical or
pyramidal in form and of very small scale. The true compass of the
Bassoon of the orchestra does not extend beyond eb 1 ; accordingly,
above this note the stop is completed in what are practically OR-
CHESTRAL OBOE pipes (See ORCHESTRAL OBOE). The tone of the
ORCHESTRAL BASSOON should throughout its compass be of greater
body than that which characterizes the ORCHESTRAL OBOE, so as to
give a marked individuality to the stop. For further and full par-
ticulars, see BASSOON.
ORCHESTRAL CLARINET. The name properly employed
to designate the lingual solo stop formed and voiced to yield a tone
imitative of that of the orchestral Clarinets. For general particulars
respecting the formation, tonality, etc., of the CLARINET, 8 FT., of
the Organ, see CLARINET. As it has been found that the CLARINET,
in its usual form, is rendered much more closely imitative.of the tone
of the orchestral instruments, by having a soft-toned labial stop of
perfect blending quality combined with it, we would suggest that
the term ORCHESTRAL CLARINET be confined to a dual stop so
formed. In our experience we have found that a covered wood flute-
toned stop preferably a small-scaled and softly-voiced DOPPEL-
FLOTE, 8 FT. combines perfectly with the lingual stop, and pro-
duces a compound tone of remarkable imitative quality. It is in
the imitation of the chalumeau register of the Clarinet of the orches-
tra that the voicer will meet with the principal difficulty; but this
can be largely overcome by skillful voicing, without causing any
perceptible break in the tone of the stop.
ORCHESTRAL FLUTE. Fr., FLUTE TRAVERSIERE. Ital,
FLAUTO TRA VERSO. Ger. CONCERTFLOTE. The stop to which these
THEIR ARTISTIC REGISTRATION 203
names are properly applied yields tones as closely imitative of those
of the Flute of the orchestra as are practicable in a rank of organ-
pipes, blown by wind of uniform pressure. There are only two
Flutes employed in the orchestra, the Flauto Traverso and the
Flauto Piccolo; accordingly, the range of tone to be imitated is very
limited. Several forms of both metal and wood pipes have been
devised, by organ-builders of different nationalities, for the produc-
tion of satisfactory ORCHESTRAL FLUTES. The most effective
examples which have come to our observation are those labeled
FLAIJTO TRAVERSO, of both 8 ft. and 4 ft. pitch, in the Organ in the
Parish Church of Doncaster, Yorkshire, made by Edmund Schulze
of Paulinzelle. In these we find cylindrical pipes of wood employed
yielding tones almost identical with those of the orchestral instru-
ment. Another beautiful example, by the same artist, of 8 ft. pitch
and tenor C compass, exists in the Echo of the Organ in the Parish
Church of Leeds, where it speaks on wind of ij/ inches. The stop
appears under the names CONCERTFLOTE, 8 FT., and TRAVERSFLOTE,
4 FT., on the Second Manual of theOrgan in the Cathedral of Liibeck.
In its most expressive name, ORCHESTRAL FLUTE, 8 FT., it exists in
the Solo of the Organ in the Music Hall of the Carnegie Institute,
Pittsburgh : and, of 4 ft. pitch, in the Solo of the Organ in St. George V
Hall, Liverpool. Under the name FLAUTO TRAVERSO, of 8 ft., 4 ft.,
and 2 ft. pitch, it is introduced in the Solo of the Organ in the Cen-
. tennial Hall, Sydney, N. S. W. The stop of 2 ft. pitch should have
been labeled ORCHESTRAL PICCOLO. The stop exists in French
Organs under the name FLUTE TRAVERSIERE, 8 FT., as in the Organ
in the Church of Saint-Sulpice, Paris; and under the uncommon
name FLUTE D'ORCHESTRE, 8 FT., in the Recit of the Organ in the
Cathedral of Albi.
FORMATION. Three forms have been adopted for the wood pipes of the
ORCHESTRAL FLUTE; namely, cylindrical, quadrangular, and triangular; all of
which are, in the best examples, harmonic in the principal portion of their com-
pass. In the 8 ft. stop the harmonic pipes should commence on c x , and in the 4 ft.
stop on tenor C.
The pipes which produce tones most closely resembling those of the orchestral
Flute are formed of circular tubes of hard wood, having small mouths, placed in
relation to their blocked ends just as is the embouchure of the orchestral Flute.
Their caps are formed so as to direct the wind-stream across their mouths, just as
the wind from the human mouth is directed across the embouchure of the orches-
tral instrument. The cylindrical pipes are invariably harmonic, and, accordingly,
are about double the standard speaking length.* Flutes of this imitative form are
* A full description, accompanied by accurately detailed drawings, of the fine cylindrical
stops in the Organs in the Parish Churches of Doncaster and Leeds, made by Schulze, is given in
204 ORGAN-STOPS
met with in high-class German work; but we are not aware of one ever having
been made by a French, English, or American organ-builder; their cylindrical
FLUTES being invariably of metal, and, when imitative, harmonic.
Very successful ORCHESTRAL FLUTES have been made with quadrangular
harmonic pipes, usually of 4 ft. pitch. These are usually square, though in some
cases slightly deeper than wide. Their mouths are circular and inverted: and
their caps, which cover a small portion of the mouths, are formed so as to direct
the wind-stream in the manner described above.* A good example of the ORCHES-
TRAL FLUTE, 4 FT., formed of harmonic triangular pipes, is furnished by the stop
in the Echo of the Concert-room Organ in the Town Hall of Leeds. The harmonic
pipes commence on the tenor C key and are carried to the top. The bottom octave
is of nonharmonic pipes. The form and construction of the triangular pipes are
shown in Fig. 27, in which are given a Front View and Longitudinal and Trans-
verse Sections of the largest one. The following are the measurements of the three
pipes yielding the 2 ft., I ft., and M ft- notes, taken in the Organ. C, 2 ft. tone,
width 2 M ins. ; depth along sides 2 ins. ; length from lower lip 3 ft. 10 ins. ; distance
of perforation from lower lip I ft. 9 ins. ; diameter of perforation J4 in. ; width of
mouth i % ins. ; height of mouth $ in. ; and block sunk below lower lip of mouth
% inch. Pipe c 1 , I ft. tone, width i^ ins.; depth along sides i^ ins.; length
from lower lip I ft. 10 ins. ; distance of perforation from lower lip 1 1 ins. ; diameter
of perforation xV in.; width of mouth J^ in.; height of mouth ^ in.; and block
sunk below lower lip 5^ in. Pipe c 2 , J^ ft. tone width i y$ ins. ; depth along sides
Y| in. ; length from lower lip 10 % ins. ; distance of perforation from lower lip $$
ins. ; diameter of perforation J^ in. ; width of mouth 5^ in. ; height of mouth -^ in. ;
and block sunk below lower lip fg- in. The wind pressure is 2 J^ inches; and the
tone is pure and of medium strength. The FLUTE TRAVERSIERE and FLUTE
D'ORCHESTRE, of the French organ-builders appear to be formed invariably of
metal harmonic pipes of usual shape and treatment.
TONE AND REGISTRATION. While the strictly imitative voice of
the ORCHESTRAL FLUTE renders the stop essentially a solo one, and
most effective in that capacity, it is extremely valuable in combina-
tion and artistic registration. Its pure and singularly liquid intona-
tion, when not undesirably prominent, creates beautiful qualities of
compound tone of an orchestral character, through its decided con-
trast with the voices of the imitative string-toned and reed-toned
stops, with which it combines perfectly. In both its solo and com-
binational offices, the ORCHESTRAL FLUTE is valuable in both its
8 ft. and 4 ft. pitches.
ORCHESTRAL HORN. The name given to the important
lingual stop, the voice of which imitates that of the Horn of the
orchestra. For full particulars respecting formation, tone, etc., see
HORN.
"The Art of Organ-Building," Vol. II., pp. 463-4; and in '* The Organ of the Twentieth Century,"
PP- 438-9.
* A fine quadrangular example is described and illustrated in "The Art of Organ-Building, "
Vol. II., pp. 465-7 : and in "The Organ of the Twentieth Century, " pp. 439-40.
THEIR ARTISTIC REGISTRATION
205
^ ORCHESTRAL OBOE. The lingual stop, of 8 ft. pitch, voiced
to imitate the tones of the Oboe of the orchestra. Both in formation
and tonality it differs from the ordinary unimitative OBOE of the
Organ. The ORCHESTRAL OBOE, 8 FT., is of comparatively rare
introduction in even the large Organs of to-day; the difficulty of
producing the peculiar voice of
the orchestral instrument being
the principal cause. Speaking of
the Oboe, Berlioz correctly re-
marks: "It is especially a me-
lodial instrument, having a pas-
toral character, full of tenderness
nay, I would even say, of tim-
idity," From such a description
one can readily realize the prob-
lem before the pipe-maker and
reed-voicer. The most success-
ful examples of the ORCHESTRAL
OBOE, 8 FT., which have come
under our direct observation are
those constructed by Willis, and
inserted in several of his import-
ant Organs, including those in
St. George's Hall, Liverpool; the
Town Hall, Huddersfield; and
the Cathedrals of Durham and
Glasgow. The ORCHESTRAL
OBOE was invented and first
made by George Willis, brother
of Henry Willis, England's most
distinguished organ-builder.
George voiced all the superb
series of lingual stops in the St.
George's Hall Organ (1855)
founding thereby the unequalled
Willis school of reed-voicing.
FIG. 27
The stop has been, so far as our observation extends, completely
ignored by French organ builders : we have been unable to find an
example even in their most important Organs. The stop labeled
BASSON-HAUTBOIS exists in many instruments, but it cannot be con-
sidered orchestral in character. Stops labeled ORCHESTRAL OBOE,
8 FT., exist on the First Manual of the Organ in the Cathedral of
206 ORGAN-STOPS
Ulm ; and on the Third Manual of the Organ in the Synagogue, Berlin.
FORMATION. In briefly describing the formation of the pipes of the ORCHES-
TRAL OBOE, one cannot do better than follow that adopted by Willis. The re-
sonators are of an extremely slender inverted conical form, devoid of bells, and
having closed upper ends and long and narrow slots adjoining them. This form of
resonator is shown in Fig. 28, Plate III., which is drawn from the Willis ORCHES-
TRAL OBOE in the Organ in the Town Hall, Huddersfield. The reeds or echalotes
are of very small scale, and have their stopped ends formed at an acute angle up-
ward from the lower edge of their faces. The tongues or languettes are very narrow,
of good substance, and finely curved, as usual in all Willis reeds. Other makers
have adopted resonators of the ordinary OBOE form but of very slender propor-
tions. Others have been content to use plain inverted conical resonators, of ex-
tremely small scales, and open at top : these have only cheapness to recommend
them a very strong recommendation in the mind of the pipe-maker.
Up to this point lingual stops have been alluded to, but we have now to briefly
describe a labial stop, formed of quadrangular wood pipes, invented by Mr.
William E. Haskell, which yields a tone closely imitative of that of the Oboe of the
orchestra. It will be more convenient to describe a single pipe, taking the middle
c 1 pipe as representative. The stop is of 8 ft. pitch. The pipe measures ^ inch
square, internally, at its mouth line; and, while its width remains the same, its
depth is gradually reduced to JJ- inch at its open top end, the reduction com-
mencing about 12 inches above the mouth and taking a curve inward from that
point in the front of the pipe. The block is slightly sunk below the mouth, which
is inverted, carefully formed, and furnished with a sloping cap and cylindrical
harmonic-bridge. The length of the pipe from the mouth is i foot 1 1 J^j inches.
The stop speaks on wind of 3^ inches pressure.
TONE AND REGISTRATION. It is somewhat difficult to describe
the peculiar tone of the Oboe of the orchestra. In addition to his
description, quoted above, Berlioz speaks of its "small acid-sweet
voice" a very happy expression. Judging from the number of
compositions written for it, and the place it has always held in or-
chestral music, it is obvious that the voice of the Oboe was highly
esteemed by all the great composers. As the Oboe of the orchestra
does not go below tenor Bb, it is obvious that the imitative tones of
the ORCHESTRAL OBOE, 8 FT., of the Organ must commence on that
note. Below that note the pipes should strictly be of the FAGOTTO
class, because the Fagotto provides the proper bass to the Oboe in
the orchestra: it is this fact that led the French organ-builders to use
the term BASSON-HAUTBOIS. But it is neither necessary nor desir-
able to break the organ-stop in either its tone or the form of its pipes.
In correct Oboe solos, the performer will never be compelled to go
below the compass of the instrument; while in general organ playing
the complete stop of uniform tone will be extremely valuable in
combination and registration. While the ORCHESTRAL OBOE will
not be so generally useful in registration as the unimitative OBOE
THEIR ARTISTIC REGISTRATION 207
(q. 0.), it will lend itself to the production of many beautiful com-
pound tones having very marked tonal colorings. In this direction,
the stop and its effects are well worthy of the organist's careful study.
ORCHESTRAL PICCOLO. An open labial stop of metal or
wood, sometimes of both; formed, in the finest examples, of harmonic
pipes. It is of 2 ft. pitch; and is voiced to yield bright and piercing
flute-tones, imitative of those of the orchestral Plauto Piccolo. See
PICCOLO.
ORLO, CRO ORLO, Span. A lingual stop, of 8 ft. pitch, of the
MUSETTE or CHALUMEAU character. It exists in certain large Span-
ish Organs, including those in the Cathedrals of Burgos and Valla-
dolid. In the North Organ in the former Cathedral there are two
ORLOS on the First Clavier, probably of different pitches; and on the
Second Clavier there are two stops labeled CRO ORLO. The Spanish
organ-builders seem to have had a great love for lingual stops, judg-
ing from the number introduced in their Organs, and the manner
they disposed them, en chamade, in their sumptuous cases.
P
PANFLOTE, Ger. Ital, FLAUTO DI PAN. The name that has
been given to a very rare Pedal Organ labial stop, of I ft. pitch,
formed of open metal pipes, yielding acute tones resembling those
of the Pan-pipes. An example exists in the Pedal of the Grand
Organ in the Cathedral of Lund, Sweden. See FLAUTO DI PAN.
PASTORITA, Ital. The name that has occasionally been used
to designate the labial stop now commonly known as the French
COR DE NUIT or the German NACHTHORN (q. v.).
PAUKE, Ger. The stop so called by German organ-builders
was one of the curiosities to be found in old Organs. It consisted of
two large-scaled and loudly-voiced covered wood pipes, sounding
the notes of the proper pitches of the Pauken or Kettle-drums of
the orchestra, placed in the Pedal Organ. The loud and thumping
sounds produced by these special pipes, when played staccato, were
supposed to imitate those of the Kettle-drums.
Another curiosity, called PAUKERENGEL, was a mechanical
accessory, which took the form of an angel playing the Drum. Two
or more of these were mounted on the case, chiefly as ornaments
and were actuated by pedal mechanism. Examples are said to have
208 ORGAN-STOPS
existed in the Organ built by Joachim Wagner in 1725, for the Garri-
son Church, Berlin.
PERDUNA. The name which has sometimes been employed
by old German organ-builders to designate a covered wood stop, of
1 6 ft. pitch, and similar in all essentials to the BOURDON.
PPEIFE, Ger. This term when used alone properly designates
an open metal labial stop, of 2 ft. pitch, yielding a bright flute-tone.
It is, in the generality of examples, practically identical with the
stop named FIFE (q. v.). In certain compound names, used by
German organ-builders, the terms PFEIFE and FLOTE seem to be
synonymous. An instance of both terms being used in combina-
tion is given by Seidel-Kothe : " PFEIFERFLOTE wird zuweilen NASAT
2%' genannt."
PHILOMELA. Ital., FILOMELA. The name, which means
Nightingale (Daulias philomela), has been used to designate stops of
widely different tonalities so different, indeed, that one is disposed
to ask, so far as organ-builders' terminology is concerned, "What's
in a name? 1 ' In the first place, it has been applied, appropriately,
to a small-scaled wood stop voiced to yield an extremely refined and
soft flute-like tone, suggestive of the voice of the Nightingale.
Clarke, in his " Structure of the Pipe Organ," describes the PHILO-
MELA, 8 FT., as a flute-toned stop formed of "Small scale stopped
wood pipes, voiced with the sweetest and most delicate quality." In
the second place, and with a widely different signification, the name
is given to wood stops of large scale and powerful intonation. The
PHILOMELA, 8 FT., in the unexpressive Solo of the Organ in the Cin-
cinnati Music Hall, built by Hook & Hastings in 1878, is thus de-
scribed by the builders: "Open pipes of wood, having two mouths.
Tone full, rich, and mellow. ' ' This stop speaks on wind of ten inches
pressure. As a large-scaled, open, double-mouthed stop, the PHILO-
MELA has been styled a wood STENTORPHONE. If the name is to be
retained in modern stop nomenclature, let it be confined to the stop
which, however imperfectly, yields a tone having some resemblance
to that of the Nightingale's song.
PHOCINX. An uncommon term that has been applied, in a
few instances, to the German lingual stop usually known as the
KRUMMHORN. Fr. CROMORNE (q. v.).
PHYSHARMONIKA. A soft-vdiced lingual stop, of 16 ft.
and 8 ft. pitch, to be found in several German and Swiss Organs.
THEIR ARTISTIC REGISTRATION
20Q
It is placed on the Fourth Manual of the Organ in the Cathedral of
Riga; and on the Third Manual of the Organ in the Cathedral of
Ulm. Both stops are of 8 ft. pitch. In the well-known Organ in the
Cathedral of Lucerne, the PHYSHARMONIKA, 8 FT., is placed in the
Swell. In the Organ in the Cathedral of Fribourg, in Switzerland,
there are PHYSHARMONIKAS, of 16 ft. and 8 ft. pitch. The stops in
both the Lucerne and Fribourg Organs we personally examined,
accompanied, in Lucerne, by F. Haas, the organ-builder who in-
serted them: accordingly, we are able to describe the stops. In the
Organ in the Cathedral of Magdeburg there is an 8 ft. stop of similar
description, labeled HARMONIUM.
FORMATION. The best examples of the PHYSHARMONIKA to be found in
modern European Organs have been made by J. & P. Schiedmayer of Stuttgart,
through whose courtesy we are able to give an illustration showing the formation
of the stop. Fig. 29 is a Transverse Section of the complete appliance in its most
improved form. A is the chamber into which the compressed pipe-wind is con-
FIG. 29
ducted by a suitable wind-trunk or conveyance. This chamber is connected
through the opening D, with the bellows B, which, together with the chamber A,
forms the compressed-air reservoir. The bellows B is acted on by the spiral spring
C, which properly regulates the wind-pressure at all times on the many tongues.
E is a small escape-valve, held against its port by a light spring. A Longitudinal
Section of a free tongue or vibrator with its brass frame is shown at F; and its
2io ORGAN-STOPS
tuning-clip and wire are shown at G. The wire passes air-tight through the side
of the chamber A, to enable the tuning to be done from the outside. H is the reed-
groove special to the vibrator F, furnished with the pallet-hole I. J is the pallet,
covering the hole I, and commanded by the key-action of the Organ through the
agency of the rocking-lever K and the pull-wire L. The pallet is held against its
seat by the spring M, which is strong enough to resist the downward pressure of
the wind on the surface of the pallet. N is the sound-chamber, general to all the
reed-grooves: this is properly made much deeper than is shown. It is furnished at
its end, or ends, with a pivoted or sliding appliance for crescendo and diminuendo
effects. The free-reeds of the PHYSHARMONIKA are similar in all respects to those
used for the ordinary free-reed stops; the only difference lies in the manner in
which they are mounted.
The PHYSHARMONIKA, 8 FT., in the Lucerne Organ is, in its sounding portion,
similar to that just described, but is inclosed in a special swell-box, in one side of
which several heart-shaped openings are cut for the egress of sound ; and these are
commanded by a sliding shutter in which corresponding perforations are made.
The to-and-fro motions of the shutter, under the action of a small expression lever,
produce a perfect and gradual crescendo and diminuendo. In the Fribourg Organ
the PHYSHARMONIKAS have their reeds furnished with short resonators. They are
inclosed in the same manner as in the Lucerne instrument.
TONE AND REGISTRATION. The tones of the PHYSHARMONIKA
are somewhat indeterminate but generally pleasing, combining in a
satisfactory manner with the tones of the labial stops of all tonalities,
chiefly in the capacity of timbre-creator. In this direction one tonal
difficulty obtains; for, like all free-reed stops, the PHYSHARMONIKA
is not affected by changes of temperature in the same manner as are
the labial stops ; and, accordingly, is not always in perfect tune with
them. It, however, can be easily tuned and that difficulty is readily
overcome.
Free-reed stops devoid of resonators have recently been intro-
duced by certain English organ-builders, with favorable results in
tonal combination and registration. In the Organ in Colston Hall,
Bristol, built by Norman & Beard in 1905, there are three such free-
reed stops, labeled HARP ^EOLONE, KEROPHONE, and SAXOPHONE.
PICCOLO, FLAUTO PICCOLO, Ital A labial stop, of 2 ft.
pitch, formed of metal, or of wood and metal, the pipes of which are of
small scale, and voiced to yield a clear flute-tone. While 2 ft. pitch
is obviously the correct and most desirable one, the PICCOLO some-
times appears as a stop of i ft. pitch, as in the Positif of the Organ
in the Church of St. Sulpice, Paris. Under the name HARMONIC
PICCOLO, 4 FT., it is inserted in the Solo of the Organ in the Town
Hall of Leeds; while in the Great there is a PICCOLO, 2 FT. In the
case of a stop like this, which is, or should be, imitative of the orches-
tral Piccolo, it is most desirable that organ-builders should adopt a
rflEIR ARTISTIC REGISTRATION 211
uniform pitch, that of 2 ft. When voiced to closely imitate the voice
of the orchestral instrument, it is appropriately called ORCHESTRAL
PICCOLO. The French organ-builders usually fashion this imitative
stop with harmonic pipes, and label it PICCOLO HARMONIQUE.
TONE AND REGISTRATION. The PICCOLO, 2 FT., in its proper
and most desirable form, may be considered the FIFTEENTH or
SUPER-OCTAVE of the Flute-work of the Organ: but, being highly
imitative in its voice, it holds a decided position in the orchestral
forces; its proper place being in the Wood-wind division, or in the
Solo, of the properly-appointed Concert-room Organ. It is very
questionable if its voice is called for in the legitimate Church Organ.
The tone of this imitative stop should be carefully proportioned to
those of the ORCHESTRAL FLUTES, 8 FT., and 4 FT., so that whilst it
will prove amply sufficient for any solo work, it will not dominate,
by its high pitched and necessarily piercing voice, the graver and
richer voices of those more important stops. In artistic registration
its chief office will be in the production of bright qualities of orches-
tral tone, in which it will prove invaluable, never, however, losing
its identity ; in this respect resembling the Piccolo of the orchestra.
PICCOLO D'AMORE, ItaL The true OCTAVE of the FLAUTO
D'AMORE, 4 FT. It is a softly-voiced flute-toned stop of 2 ft. pitch,
properly formed of half -covered wood pipes of small scale, from CC
to c 2 , and thence to top note in small-scale open metal pipes, prefer-
ably of SPITZFLOTE form. The PICCOLO D'AMORE, 2 FT., is an ideal
SUPER-OCTAVE for the true Chamber Organ, or for an Ancillary
Aerial Organ.
PIFFERO, ItaL Literally Fife. The name used by Italian
organ-builders to designate a stop which is in all essentials similar to
the French CHALUMEAU (q. .).
PILEATA, Lat. Literally Hooded. A generic name for labial
covered stops, having no direct reference to their tone.* When used
alone it simply means a GEDECKT. Combined with other terms,
we find PILEATA MAXIMA a GROSSUNTERSATZ, 32 FT.; PILEATA
MAGNA a GROBGEDECKT, 16 FT.; PILEATA MAJOR a MITTEL-
GEDECKT, 8 FT. ; and PILEATA MINOR a KLEINGEDECKT, 4 FT.
* " VOX-PILEATA, ou simplement PILEA.TA. C'est encore un de ces noms gene"riques de jeu
bouche, dontla facture allemande abonde mme par tradition, puisqueles premiers trait6sde
facture furent 6crits en latin. Cela veut dire voix couverte, ou, mot a mot, coijfee; la coiff e est ce
quel'on nommeaujourd'nuicaZoWedansles bourdons de metal, et tampon oubouchon dans ceux
de bois. La qualification de Pileata se modifie selon la grandeur du registre ; de tous petits r<S-
gistresbouches, tels quela Bauer-Fldte d'un pied s'appelleront Pileata-Minima. Un bourdon de
212 ORGAN-STOPS
PLEIN-JEU, Fr. The name used by French organ-builders to
designate a compound harmonic-corroborating stop, of several
ranks of medium-scaled open metal pipes, yielding pure organ-tone.
A fine example of the stop exists in the Organ in the Town Hall of
Manchester, constructed by Cavaille-Coll. The composition of this
stop is here given ;
PLEIN-JEU VII. RANKS.
CC to E ..... 15 19 22 26 29 33 36.
F to e 1 .... 812151922 2629.
f 1 to e 3 .... i 8121519 2226.
f 3 to b 2 . . . . 15 812151922.
C3 to f3 . . . . DOUBLE I 5 8 12 15 19
fit' to C< ... DOUBLE-;-!- 5 - 8-I2-I5.
This PLEIN-JEU has no special feature to distinguish it from an
ordinary MIXTURE in which octave- and fifth-sounding ranks only
are employed, except in its extreme richness of structure, and the
introduction of breaks lower in pitch than the unison of the manual
division (Grand-Orgue) in which it is placed. It may be remarked
that this division, of fourteen speaking stops, contains three stops
of 1 6 ft. pitch, which fact accounts for the introduction of the
DOUBLES and DOUBLE QUINT in the upper breaks of the PLEIN-
JEU, and also the fifth-sounding ranks in the three higher breaks
which strictly belong to the 16 ft. harmonic series. A PLEIN-JEU
of ten ranks is the only compound stop in the Grand-Orgue of the
Organ in the Madeleine, Paris. In the Positif of the Organ in
Saint Sulpice there is a PLEIN-JEU HARMONIQUE of three and
four ranks. Further particulars are given in the appended note.*
seize, fagonne" de maniere a prendre un nom quelconque de flute, ajoutera a ce nom de fantaisie
la qualification de Pileata-Magna qui empe'che 1'organiste de 1'employer comme jeu ouvert.
Enfin Pileala-Maxima de"signera les bourdons gigantesques de la p6dale. " Regnier.
*" LE PLEIN-JEU. Les modernes ont souvent donn6 le mme nom a toute espece de mix-
tures; c'est une de ces petites erreurs de detail qui entrainent apres elles 1'oubli des principesen
facture. Ainsi ai-je rencontre souvent dans de petites orgues un rdgistre appelS Plein-jeu, qui
n'avait rien de plein, et qui n'etait un jeu que dans le sens ridicule du mot. II y avait de quoi
faire prendre en horreur toute espece de mixture; et j 'attribue k ce vol, fait aux vraies conditions
du Plein-jeu par les f acteurs charlatans, la reaction qu'on voit se prononcer centre la venerable
antiquite" de cette harmonic. ' Dans un seize-pieds, ' dit Dom Bedos, ' le moindre Plein-jeu est de
neuf tuyaux sur marche (ou par note). ... Si c'est un huit-pieds, le Plein-jeu est (au moins)
de sept tuyaux sur marche; si 1'orgue est un trente-deux-pieds ouvert avec bourdon de trente-
deux, on doit mettre Fourniture entiere et Cymbale entiere (c'est-a-dire que chacun de ces rSgis-
tresdoitavoirsa plus grande force connue). Pour un positif, si c'est un huit-pieds enmontre.on
met le Plein-jeu de sept tuyaux sur marche. S'il n'y a point de huit-pieds ouverts, le Plein-jeu
nesera que de cinq tuyaux sur marche, c'est-a-dire compos6 des trois dernieres rangees dela
Fourniture et des deux dernieres de la Cymbale.' (Parce qu'alors un Plein-jeu plus fort serait
trop dur, n'etant pas soutenu par la force des huit-pieds ouverts.) Dom Bedos donne encore les
THEIR ARTISTIC REGISTRATION 213
PORTUNAL, PORTUNALFLC^; Ger.-An open wood stop,
of medium scale, and of 8 ft. and 4 ft. pitch. Its pipes are, in the
finer examples, of inverted pyramidal form, after the fashion of the
DOLCAN, yielding a very pleasing fluty tone inclining, in good rep-
resentative examples, to a Clarinet quality.
POSAUNE, Ger. A lingual stop, of 8 ft. pitch on the manuals
and 1 6 ft. pitch in the Pedal Organ; the pipes of which have resona-
tors of large scale and inverted conical form. In the manual stop,
metal resonators are invariably employed; while in the Pedal stop
they are made of either metal or wood; and sometimes of both,
wood being used for the lower octave. The tone of the POSAUNE is
intended to imitate as closely as possible that of the orchestral Trom-
bone when played forte and firmly; it, accordingly, should have more
brassiness than the TROMBONE or TRUMPET.
Examples of the POSAUNE, 8 FT., exist in the Greats of the Organs
in Westminster Abbey and the Royal Albert Hall, London; and in
the Pedal of the Organ in the Cathedral of Riga." Examples of the
POSAUNE, 16 FT., are to be found in the Pedals of the Organs in the
above-named churches; and in those of the Organs in the Cathedral
of Ulm, and the Centennial Hall, Sydney, N. S. W. Under the name
POSAUNENBASS, 1 6 FT., it exists in the Pedals of the Organs in the
Cathedral of Lubeck, and the Gewandhaus, Leipzig. See CONTRA-
POSAUNE.
PRESTANT, Fr. The name commonly employed by French
organ-builders to designate the principal OCTAVE, 4 FT., in a manual
division; yielding pure organ-tone. .Its scale is properly derived
from, and proportioned to, that of the MONTRE, 8 FT., or chief uni-
son of the division. The PRESTANT occupies the same tonal position
and fulfils the same office in a French Organ, as the so-called PRIN-
regles pour un Plein-jeu de quatre tuyaux sur marche, mais dans ce cas la maigreur du re*gistre
doit le faire rejeter, car il est deja trop grincant, mime avec cinq tuyaux seulement. * Si le Plein-
jeu est de fruit ou de six tuyaux sur marche, on prend la moiti6 dans la Fourniture, et 1'autre
moitie dans la Cymbale: voila les regies ordinaires. . . . ' Dom Bedos ajoute : ' Je ne f erai point
remarquer ici toutes les variations de quelques facteurs dans la composition et I'arrangement du
Plein-jeu (il paralt que des ce temps-la on cherchait 6conomiser sur la peine et la depense qu'oc-
casionne la facture du vrai Plein-jeu) ; mais tous s'accordant a ne mettre que des quintes et desoc-
taves, et jamais de tierces.' Je ne puis omettre ici la citation de l'e"loge du Plein-jeu par le grand
artiste b6n6dictin; il donne trop d'autoritS a ce que nous avons d6ja dit: 'Tout ce qu'il y a de
plus harmonieux dans 1'orgue, au jugement des connaisseurs et de ceux qui ont du gout pour la
vraie harmonic, c' est le Plein-jeu t lorsqu'il est m61angavec tous lesf onds qui le nourrissent dans
une juste proportion; et la raison pour laquelle on met toujours ensemble les fonds de 1'orgue
avec la Fournilure et la Cymbale, est que si Ton employait celles-ci seules dans les diffSrentes
combinaisons d'accords que fait un organiste, elles f ormeraient des sons d6sagr6ables. qui dis-
paraissent a I'oreille, lorsque le melange des sons fondamentaux les mettent au rang des sons
harmoniques.' ' ' Regnier.
214 ORGAN-STOPS
CIPAL does in an English instrument, and the OCTAVE, 4 FT., in an
American and German Organ.* The name PRESTANT, derived from
the Latin Pr&stare to stand in front, was given to the stop because,
like the MONTRE, it was commonly displayed in the case-work.
See OCTAVE.
PRINZIPAL, Ger. Ital., PRINCIPALS. The name appropri-
ately and logically employed by German and Italian organ-builders
to designate the principal unison stop in both the manual and pedal
departments of the Organ that which is commonly named DIA-
PASON in English and American Organs, and MONTRE in French
Organs. French organ-builders have in some instances used the
borrowed terms PRINCIPAL and DIAPASON, as in the Recit and Solo
of the Organ in the Church of Saint-Sulpice, Paris. In a manual
division the PRINCIPAL is correctly of 8 ft. pitch, and in the Pedal
Organ of 16 f t. pitch : but in many German Organs, as in those of the
Cathedral of Bremen and the Marienkirche, Lubeck, we find PRINZI-
PALS of both 8 ft. and 16 ft. pitch in the chief manual division
(Hauptwerk), while the foundation unison in the Pedal Organ is
labeled PRINZIPALBASS, 16 FT. German organ-builders have applied
the term PRINZIPAL alike to the organ-toned foundation stops of
32 ft., 1 6 ft., 8 ft., and 4 ft. pitch.
English organ-builders have from old times continued to apply
the term PRINCIPAL to the open metal stop, of large scale, and 4 ft.
pitch, which is the true OCTAVE of the DIAPASON, 8 FT. The term,
so applied, is both illogical and undesirable. See OCTAVE.
PRINZIPALDISKANT, DISKANTPRINZIPAL, Ger.
A PRINCIPAL or DIAPASON, 8 FT., of large scale and short compass,
which extends throughout the treble octaves of the manual compass,
* " Le PRESTANT, c'est la suite de la Montre, dont il a toutes les proportions. Nous avons vu
qu'en Allemagne, le prestant allait du seize-pieds au quatre-pieds: en France, le prestant a tou-
jours quatre-pieds a son plus grand tuyau, tellement que souvent les facteurs ne Tappellent que
le quatre-pieds. Et lorsq'ils disent tout court aussi le huit-pieds, ils ne de"signent pourtant pas
1'octave du prestant, c'est-a-dire la montre de huit, mais bien une flute qu'on adjoint a la montre
etqui, au besoin, la supple"e I'interieur de 1'orgue, quand on n'a paslemoyend'avoir d'autre
montre qu'un prestant. Faisant sinte a la montre, le prestant est done, comme elle, ouvert, de
moyenne taille, et construit en etain fin. Son harmonie, en rapport avec cette taille elance"e
et ce m6tal de choix, est fine et brillante; elle donne un tel eclat, un tel tranchant auxfondsde
taille ordinaire et de grosse taille, que son adjonction ou son silence se font vivement sentir. Le
prestant parle rarement seul ; marie" aux flutes de quatre, il prend un corps dont il n'a que
rombre dans 1'isolement, et devient une flute 6clatante si les timbres sont bien d 'accord.^ Lie aux
huit-pieds .illeur donne du brillant; il ne supporte la grave adjonction des seize-pieds que
moyennant celledes huit, qui serventde transition. . . . Quelques facteurs intitulent prestant
le quatre-pieds dela pSdale; ce n'est qu'une flute de grosse taille en e"toffe, qui domine ses deux
compagnes ordinaires, flutes de huit et de seize, surtout quand ces flutes sont en bois, et
meme bouch&es. " Regnier.
THEIR ARTISTIC REGISTRATION 215
or from middle c 1 to the top note. The stop has very rarely been
carried to tenor C. This short stop was considered of considerable
value by the old German organ-builders, on account of its power of
reinforcing the weaker octaves of the unison foundation-work. This
weakness was so forcibly realized by Christian Muller, that he in-
serted two pipes to each note in the treble of several of the more
important stops in his noble Organ in the Cathedral of Haarlem.
This matter might, with advantage, receive serious consideration
at the present time; for the natural weakness of the treble in the
Organ still remains one of its tonal short-comings.*
PRINZIPALFLOTE, Ger. A powerfully-toned, open metal,
labial stop, of large scale, and 8 ft. and 4 ft. pitch; the voice of which
is a compromise between pure organ-tone and flute-tone. It has
been introduced as a Solo stop; as, 'by Walcker, in the Organ in St.
Peterskirche, Frankfurt. Such a stop, of unison pitch, would be
valuable as a timbre-creator; and might, with advantage, take the
place of a DIAPASON, 8 FT., when several are introduced. It would
form an admirable foundation stop in the wood- wind division of the
true Concert-room Organ.
PROGRESSIO HARMONICA. A compound harmonic-cor-
roborating and treble-enriching stop, the ranks of which have no
breaks, but increase in number as the stop progresses upward
through the manual compass. According to Seidel the stop was
invented by Musikdirektor Wilke, of Neu-Ruppin, and recom-
mended by him especially for small Organs. The stop, as originally
devised, begins at CC with two ranks of i^ ft. and i ft. pitch, and
at tenor C a third rank of 2 ft. pitch is added. Seidel adds: "The
scale of the stop is between those of the PRINZIPAL and the CORNETT.
The intonation is strong and the effect very fine. The tone of the
Organ becomes by this stop distinct, full, and bright." The PRO-
GRESSIO HARMONICA in the Brustwerk of the Organ in the Cathedral
of Merseburg, in Saxony, built by Ladegast in 1855, commences with
two ranks in the bass and finishes with four ranks in the treble. A
similar stop exists on the Third Manual of his Organ in the Cathedral
of Schwerin. The stop in the Unterwerk of the Organ in St. Peters-
kirche, Berlin, commences with three ranks and finishes with five
ranks in the treble.
To any one who has given serious consideration to the subject
of compound-tone production in the Organ, the advantages attend-
* See " Weakness and Augmentation of the Treble, " Chap. XV., Vol. II., pp. 1-12, of **The
Art of Organ-Building."
2i6 ORGAN-STOPS
ing the introduction of stops of the PROGRESSIO HARMONICA class
must be obvious and deserving of careful study. The inventor of
the stop was also the inventor of the COMPENSATIONSMIXTUR (q. v.),
applied to the Pedal Organ, in which the opposite treatment is
adopted; the purpose being to impart distinctness and richness to
the lower and somewhat indeterminate notes of that grave depart-
ment.
PYRAMIDFLOTE, Ger. Eng., PYRAMIDAL FLUTE. A wood
stop, of 8 ft. pitch, the pipes of which are square and, as the name
implies, are smaller at the top than at the mouth line. The tone
differs in stops made by different organ-builders, but in the best
examples it is light and clear, resembling a combination of the tones
of the MELODIA and the GEMSHORN. A good example exists in the
Unterwerk of the Organ in the Church of SS. Peter and Paul, Lieg-
nitz, in Silesia, built by Buckow in 1839. A stop of this class is more
statable for a Chamber Organ than for either a Church or a Concert-
room Organ: its tone is refined but without marked individuality.
PYRAMIDON. The name given to an open wood stop, of 16
ft. pitch, the pipes of which were of inverted pyramidal form and
remarkable proportions. It was invented by the Rev. Sir Frederick
A. Gore Ouseley, and made by Flight. The peculiarity of the stop,
which was applied to the Pedal Organ, lay chiefly in the form and
proportions of its pipes, which were covered, the CCC pipe measur-
ing 2 feet 3 inches square at top, 8 inches square at the mouth line,
and only 2 feet 6 inches speaking length. The 16 ft. tone produced
by this pipe resembled that of a BOURDON, but had nothing special
to recommend it. - The chief interest of the stop lay in the acoustical
problem it presented : its disadvantages were several. It now takes
its place among the discarded curiosities of organ-building. -
Q
QUARTE DE NASARD, Fr. The term sometimes employed
by French organ-builders to designate the open metal stop, of 2 ft.
pitch, belonging to the 8 ft. harmonic series. It is practically iden-
tical with the English FIFTEENTH or SUPER-OCTAVE, 2 FT. It derives
its name from the fact that its pitch is at the interval of a fourth
above that of the NASARD, 2% FT.* An example exists in the Positif
of the Organ in the Church of Saint-Remi, Amiens.
*"QUARTE. Jeu del'orgue. Quoique ce jeu. soit a 1'unisson de la doublette onluiadonn6
le nom de quarte, parce qu'en suivant la progression ascendante des jeux du cornet dont ilest
THEIR ARTISTIC REGISTRATION 217
, Ger. The name, which is formed of the words
quer cross or athwart and Fldte Flute, has been frequently used
by German organ-builders to designate the stop which, in its voice,
imitates, as closely as practicable in organ-pipes, the tone of the
Flute of the orchestra. Under the heading QUERFLOTE, QUER-
PFEIFE, Seidel gives the following particulars:
" QUERFLOTE is a labial stop of a particularly fine tone, imitative
of that of the real Flute. Organ-builders, in their endeavor to make
this imitation as striking as possible, have essayed with this stop all
sorts of shapes and proportions. The pipes are usually made of oak,
pear-tree, or maple; and they are either cylindrical or quadrangular,
open or stopped. Some organ-builders make the pipes twice as long
as they usually appear, and overblow them so as to make them
sound the octave higher. Other organ-builders bore out the bodies
of the pipes, and provide them with mouths of an oval form, like the
embouchure of the real Flute. The QUERFLOTE made by Muller, of
Breslau, for his Organ in the Cathedral of his city, has oval mouths
in its pipes, against which the wind is directed sidewise, imitating the
method of blowing the real Flute. "
A QUERFLOTE, 8 FT., of the compass of the orchestral Flute, was
inserted in the Organ reconstructed by Engelbert Maas, in 1821,
for the Cathedral of Cologne. For further particulars see FLAUTO
TRAVERSO and ORCHESTRAL FLUTE.
QUINT. Fr. and Ger., QUINTE. Ital. and Lat., QUINTA.
The stop, correctly termed, which speaks a fifth above the unison
pitch of the division of the Organ in which it is placed: it is, accord-
ingly, of 5]^ ft. pitch on the manuals, and 10% ft. pitch in the Pedal
Organ. The pipes forming the manual QUINT are properly of open
metal, and of medium scale; while those of the Pedal Organ QUINT,
though preferably of open metal, may be made of wood. Covered
pipes have been used, but they are not to be recommended; they do
not satisfactorily fulfil the office for which they are intended.
TONE AND REGISTRATION. The QUINT, 5^ FT., is a mutation or
harmonic-corroborating stop, belonging to the i6ft. harmonic series,
and, as it is strictly a member of the foundation-work of the Organ,
its voice should be of pure organ-tone of medium strength, produced
by cylindrical metal pipes of the DIAPASON formation. Its presence
is not called for in any save the First or Great Organ, and only there
une des parties constituantes, il se trouve a la quarte au-dessus du nasard. Aussirappelle-t-on
r6ellem^nt quarte de nasavd, et ce n'est que par abrSviation qu'on dit simplement quarte."
Hamel.
218 ORGAN-STOPS
when a DOUBLE DIAPASON, 16 FT., is present. Under certain condi-
tions, however, it may be introduced in the division when no foun-
dation stop of labial character below 8 ft. pitch is present, for the
purpose of generating the differential 16 ft. tone, in combination
with the DIAPASON, 8 FT. In registration, the QUINT, 5^ FT., may
be effectively combined with any lingual stop of 16 ft. pitch, pro-
ducing beautiful compound tones, not possible of production with-
out it. Much, however, depends on the quality of the voice of the
stop and the pipes from which it is produced. To save expense,
English organ-builders have commonly used covered pipes, inartis-
tically voiced, producing undesirable thick and dull effects in com-
bination. Better results are obtained by Continental builders, who
use open metal and wood pipes, cylindrical, quadrangular, conical,
and inverted conical in form; producing various tones, all of which
introduce colorings, and affect artistic registration.
The QUINT, 10% FT., should be introduced, in one form or
another, in every important Pedal Organ in which there is no stop
of 32 ft. pitch. The stop may be either of metal or wood, and either
open or covered preferably open according to the stop appor-
tionment of the department. In the Organ built by Schulze, in
1850, for the Cathedral of Bremen, there is a stop labeled QUINTEN-
BASS, 10% FT., and another labeled GROSSQUINTENBASS, 21 J^ FT.
The former is legitimate, corroborating the second upper partial
tone of the 32 ft. prime, and in combination with the DIAPASON,
1 6 FT. , corroborating the 32 ft. tone ; while the latter could have been
introduced solely with the view of generating the differential tone
of the 64 ft. pitch, just as it has been employed in other important
Organs. See GRAVISSIMA.
QUINTADENA. Stops bearing this name have been intro-
duced in numerous Organs. The earliest instances of its introduc-
tion we have been able to find are those in the Organ in the New
Church, Amsterdam. It is an old instrument, enlarged in 1673 by
Duyshor van Goor, of Dordrecht. In the Great there is a QUINTA-
DENA, 16 FT., and in both the Choir and Echo a QUINTADENA, 8 FT.
The stop seems to have been a favorite with the old Dutch organ-
builders.
The QUINTADENA is in all essentials similar to the covered stop
correctly designated QUINTATEN (q. u), yielding a compound tone
in which the twelfth, or second upper partial tone, is present in a
pronounced degree along with the prime or fundamental tone. The
stop is formed of covered pipes of metal or wood: as a rule the metal
THEIR ARTISTIC REGISTRATION 219
stop is to be preferred, if not too large a scale, and artistically voiced.
We find some variations of the prevailing name; namely, QUIN-
TADEN, QUINTADENE, and QuiNTADEMA. Under the last name, the
stop exists in the Swell of the Organ in the Music Hall, Cincinnati.
It is formed of covered pipes of tin, of 8 ft. pitch.
QUINTATEN. The name derived from the Latin words quin-
tain tenentes (holding the fifth), and properly applied to covered
stops which yield compound tones, in which the second upper
partial tone is almost as pronounced as the prime or ground tone.
Helmholtz correctly remarks : ' ' Narrow stopped pipes let the twelfth
be very distinctly heard at the same time with the prime tone;
and have hence been called Quintaten (quintam tenentes)"* The
term QUITATEN has comparatively seldom appeared in organ speci-
fications or stop-knobs , several corruptions having been substituted
according to the caprice of different organ-builders in different
countries. We give these corruptions so that they may be known
and avoided in the stop nomenclature of to-day: QUINTATON,
QUINTATON, QUINTADEN, QlJINTADON, QUINTADINER, QUINTADENE,
and QUINTGETON.
The QUINTATEN is formed of covered pipes of metal and wood,
and of 1 6 ft., 8 ft., and 4 ft. pitch: all are very valuable in their re-
spective places; and collectively they form a remarkable family, the
importance of which has been altogether overlooked by organ de-
signers; in this direction it is on an equality with the LIEBLICH-
GEDECKT family.
TONE AND REGISTRATION. The compound tone of the QUIN-
TATEN, when produced from pipes correctly scaled, and artistically
voiced on wind of proper pressure, is extremely valuable in combina-
tion and artistic registration. As before stated, the tone yielded by
a QUINTATEN pipe is a compound of the prime and its second upper
partial tone, the first harmonic produced by a covered pipe,
and it is from this acoustical phenomenon that the stop derives its
special value, enabling the Octave-quint or Twelfth to be introduced
in registration in a degree of softness and refinement impossible to
be imparted by an independent fifth-sounding stop. The relative
strength of the Twelfth varies in different stops. In some it is sub-
dued, while in others it is almost as prominent as the prime tone.
In a really fine QUINTATEN, 8 FT., the prime tone should be firmly
*" On the Sensations of Tone," by Hermann L. F. Helmholtz, M.D. Translated by A. J.
Ellis, B.A., 1875.
220 ORGAN-STOPS
established as the pitch of the stop; but as regards the relative
strength of the upper partial, much will depend on the general stop
apportionment in which the QUINTATEN is inserted : it should, how-
ever, in all cases be distinctly pronounced, otherwise the stop will
lose much of its value.
An artistically voiced QUINTATEN forms an effective solo stop,
notwithstanding the argument advanced against its use on account
of the production of consecutive fifths. But this objection is of no
real value, for it must be recognized that the natural fifths, produced
as harmonic upper partial tones, have a widely different tonal effect
from the independent consecutive fifths condemned in musical
theory and composition. The fifth-sounding stops and ranks in
compound stops, introduced in the Organ as harmonic-corroborators
if properly apportioned and scientifically graduated in strength of
tone, produce no objectionable tonal effects: on the contrary, they
are demanded in .the tonal structure of the Organ, and are abso-
lutely necessary for the creation of many desirable qualities of com-
pound tone; and, accordingly, have very important offices in artistic
registration.
The QUINTATEN, 16 FT., although by no means so valuable in
manual divisions as the unison stop, is greatly to be preferred to a
BOURDON, 16 FT., of the ordinary class. Its subordinate upper
partial, sufficiently representing the sH ft. tone, gives great fulness
and richness to all combinations in which the stop is introduced.
The stop imparts great solidity and dignity to the tones of lingual
stops of 1 6 ft. pitch. Voiced with the harmonic-bridge, the QUIN-
TATEN, 1 6 FT., can be made an imitative DOUBLE BASS suitable for
a small Organ.
The QUINTATEN, 4 FT., is comparatively of little value unless in
association with the 16 ft. and 8 ft. stops completing the family:
but certain organ-builders have introduced it as the only octave stop
in a manual division, rendering a TWELFTH, 2% FT., unnecessary,
but the presence of a SUPER-OCTAVE, 2 FT., is imperative to cover the
prominent upper partial of the QUINTATEN.
QUINTENBASS, Ger. The name used by German organ-
builders to designate the mutation or harmonic-corroborating stop,
of 10^3 ft., pitch belonging to the 32 ft. harmonic series, and prop-
erly introduced in the Pedal Organ along with the DOUBLE DIAPA-
SON, 32 FT. ; or, in the absence of that important stop, with the view
of producing in combination with the DIAPASON, 16 FT., the differen-
tial tone of 32 ft. pitch, sometimes designated "acoustic bass." The
THEIR ARTISTIC REGISTRATION 221
stop has been made of wood and metal, and of both open and covered
pipes; but open pipes are to be desired in all cases. An example of
the stop exists in the Pedal of the Organ in the Cathedral of Bremen;
another, under the name QUINTBASS, exists in the Pedal of the Organ
in the Cathedral of Riga; and one labeled MAJOR-QUINTE, 10^ FT.,
is inserted in the Pedal of the Organ in the Marienkirche, Lubeck.
Under the name GROSSQUINTENBASS, 21^ FT., a stop exists in the
Pedal of the Organ in the Cathedral of Bremen. See QUINT.
QUINT FLUTE. Ger., QUINTFLOTE. An unimitative flute-
toned stop of 5^ ft. and 2% ft. pitch. Examples of the QUINT-
FLOTE, 5 J^ FT., exist on the Second Manuals of the Walcker Organs
in the Cathedral of Ulm and St. Paulskirche, Frankfurt a. M. A
QUINTFLOTE, 2% FT., is inserted in the Choir of the Organ in the
Music Hall, Cincinnati. This stop is of open metal pipes, yielding a
flute-tone which combines well, as a harmonic-corroborator and
timbre-creator, with the unison and octave flute-work of -the divison.
A compound QUINT FLUTE, 8 FT., has recently been introduced
by Mr. George W. Till, of Philadelphia, Pa., and inserted in the so-
called Etherial Organ of the large Concert instrument in the Wana-
maker Store, in Philadelphia. So far as our knowledge extends, this
is the most noteworthy dual stop of 8 ft. pitch ever constructed; and
one that could only be inserted, under favorable conditions, in
Concert-room Organs of the first magnitude. Its compound voice
is of considerable grandeur, its tonality surpassing in depth and
richness of color that of any other flute-toned stop known to have
been produced up to the present time (1920).
The principal rank is a CLEAR FLUTE, 8 FT., formed of open wood
pipes from CC to c 4 , to which are added twelve open metal pipes for
octave coupling. The scale is large, the CC pipe measuring, inter-
nally, 8 inches in width by ioj^ inches in depth. The subordinate
rank is a QUINT, 5^ FT., formed of covered wood pipes from CC to
f$ 3 43 notes, and open metal pipes from g 2 to c s 30 notes, all
yielding a normal tone. The stop speaks on wind of 25 inches
pressure.*
- QUINTVIOLE. The name given by Zollner to a string-toned
stop, of 8 ft. pitch, introduced by him in the Organ of the Stadt-
kirche, Wittenberg, f
* Full particulars of formation and scales of this QUINT FLUTE are given in our ^ork, " The
Organ of the Twentieth Century," pp. 102-4.
t" QUINTVIOLE 8' ist eine Stimme, welche im Hauptwerke der Jahre 1814 vom Orgelbau-
meister Zdllner aus Hubertusburg erbau+en Orgel der Stadtkirche zu Wittenberg steht. Die
222 ORGAN-STOPS
R
RANKET, RACKET. An old and now obsolete lingual stop,
of 1 6 ft. and 8 ft. pitch, the resonators of which were short, and
closed with exception of a few small perforations near their lower
ends, necessary for the egress of wind and sound, after the fashion of
certain REGALS. Under such treatment, the tone was necessarily
muffled and probably humming or buzzing in character. When the
stop was first introduced is not known ; but Prastorius, in his " Theat-
rum Instrumentorum seu Sciagraphia" (1618), mentions it along
with other lingual stops of a similar character. See REGAL.
RAUSCHFLOTE, RAUSCHPFEIFE, Ger. Literally, Rustling
Flute. A dual stop commonly formed, according to Wolfram (1815),
of two ranks of open metal pipes, of 2 ft. and I J^ ft. pitch, respec-
tively. A RAUSCHPFEIFE of three" ranks, probably 2% ft., 2 ft.,
and i }^ ft. , was inserted in the Brustwerk of the Organ built by
Hildebrand, in 1762, for St. Michaeliskirche, Hamburg. The name
seems to have disappeared from German stop nomenclature now in
use.
RAUSCHQUINTE, Ger. Literally, Rustling Quint. The stop
found in German Organs, usually formed of two ranks of open metal
pipes, of 2% ft. and 2 ft. pitch respectively, standing at the interval
of a fourth apart. An example exists in the Hauptwerk of the Organ
in the Church of St. Mary Magdalen, Breslau. The RAUSCHQUINTE
of high pitch consists of two ranks, of i J^ ft. and i ft. pitch respec-
tively, as in the stop in the Hauptwerk of the Organ in the Christ
Church, Hirschberg. As the interval between the pitches of the
ranks of the RAUSCHQUINTE is a fourth, the name QUARTE, or
QUARTA, has been given it by old German organ-builders.
While the value of such a dual stop in harmonic-corroboration is
unquestionable; yet, in its usual tonality, its ranks would be still
more useful as separate stops. Dual stops are only really desirable
when they are so formed as to produce new compound tones of ex-
ceptional beauty and value; which could not be created, under
ordinary conditions, by the combination of stops of the usual or
standard tonalities. Dual stop formation opens up a field for in-
vention and skill, only just touched by the plow of the artistic pipe-
maker and voicer.
Eigenschaften dieses Registers sind dem Verf . nicht bekannt; soil es vielleicbt eine QuintatSn
die eine, der Viola ahnliche Intonation hat, sein? " Seidel-Kothe.
THEIR ARTISTIC REGISTRATION 223
RECORDER. A rare term in old organ-stop nomenclature,
but one not difficult to understand and account for. The employ-
ment of the term by the early English organ-builders is shown in the
following extracts: In an Agreement entered into by John Loose-
more, in 1665, for the building of a Chamber Organ, we find, among
the wood stops, "One Recorde" specified. And in the "Articles of
Agreement" between the Dean and Chapter of York Cathedral and
Robert Dallam, organ-builder, of London, in 1632, for an Organ to
be erected in the cathedral, we find specified for the Great Organ:
" Itmone Recorder unison to the said Principall. vi li. "; and for the
Chaire Organ: "Itm one Recorder of tynn, unison with the voice,
viij li. Accordingly, these stops were respectively of 4 ft. and 8 ft.
pitch. The instrument called Recorder was in all essentials similar
to the Flute Douce, belonging to the Whistle Flute or Flute & Bee
family. It is therefore very probable that the old organ RECORDER
was simply a flute-toned stop, imitating the voice of the old instru-
ment.
REGAL, Ger. Ft., REGALE. Ital., REGALE. The generic
name for a large family of ancient lingual stops, which in their
original forms have long ceased to be used in the stop appointments
of Organs. The name, however, still lingers in some old German
Organs, as in those in certain Liibeck churches. A REGAL, 8 FT.,
occupies a place in the Choir of the great Haarlem Organ; and there
are in the Choir of the Organ in the Church of St. Dominick, Prague,
a REGAL, 8 FT., and a JUNGFERNREGAL, 16 FT.
The term "Regal" was originally used to designate a portable
reed organ, or " Portative," used in court ceremonies; from which
fact it is understood to have derived its name. Subsequently the
term was extended to certain lingual stops, introduced in large
Organs, or " Positives," which in their voices more or less closely
resembled those of stops in the earlier Regal. Widely different
names have been employed to designate the various REGAL stops;
some of which refer directly to the character of the tones they pro-
duce, while others refer to the peculiar forms of the resonators used.
The old organ-builders certainly exercised their inventive powers
and fancy in devising curious shapes for the resonators of their
lingual pipes termed REGALS. In the accompanying illustration,
Fig. 30, are given the forms of REGAL pipes which have been pre-
served on the pages of old treatises. No. i is the SORDUNREGAL,
having a capped resonator pierced with four holes for the emission
of sound. As the name implies, its tone was very subdued. No. 2
224
ORGAN-STOPS
is the KNOPFREGAL, deriving its name from its pear-shaped head,
cut after the fashion of a sleigh-bell. No. 3 is the APFELREGAL,
deriving its name from the form of its head, which is spherical and
pierced with numerous small holes for the emission of sound. No. 4
is the KRUMMHORNREGAL, the tone of which is said to have re-
sembled that of the old Krummhorn (see CROMORNE). No. 5 is
the SCHALMEI, a stop of the REGAL family, the tone of which imi-
tated that of the old instrument called Schalmei or Shawm. Other
forms have been given to the pipes of the SCHALMEI (see CHALU-
MEAU). No. 6 is the BARPFEIFE, also a stop of the REGAL family,
which yielded a low growling tone (See BARPFEIFE). No. 7 is the
MESSINGREGAL, the tone of which had a brazen clang, probably re-
sembling that of a Trumpet. No. 8 is the RANKET, the resonator
rn
FIG. 30
of which is cylindrical, furnished with a tuning stopper, and pierced,
in the neighborhood of the reed, with sound-holes (see RANKET).
Fig. 9 is a REGAL which received the name JEU SRARD, after its
inventor (see JEU RARD). Other REGALS have been introduced in
old Organs, of which the following are the names and probable
tonalities: CYMBELREGAL, the tone of which was singularly bright
and ringing. GEDAMPFTREGAL, the tone of which was subdued or
muffled. GEIGENREGAL, the tone of which somewhat resembled
that of the Geige or Violin. HARFENREGAL, the tones of which bore
the character of the sounds of roughly-plucked harp strings. JUNG-
FERNREGAL, the tones of which were of so refined a character as to
resemble, to some degree, the youthful female voice "La Voix ou
Regale virginale." KALBERREGAL, the voice of which was of a soft
and lowing character, like that of a calf (Kalb). KLEINREGAL, ai
octave or 4 ft. lingual stop of soft intonation. SCHARFREGAL, yield-
ing a keen and cutting tone. SINGENDREGAL, the voice of which
was of a singing tonality. SUBTILREGAL, the tone of which was sub-
dued in character. TRICHTERREGAL, which derived its name from
THEIR ARTISTIC REGISTRATION 225
Its funnelshaped resonators, probably yielded a bright horn-like
tone. The term REGAL has now entirely disappeared from organ-
stop nomencl attire.
REGULA MINIMA, Lat. The usual open metal SUPEROCTAVE
or FIFTEENTH, 2 FT., belonging to the foundation-work, and corrob-
orating the third upper partial tone of the foundation unison or
prime 8 ft. tone. Yielding pure organ-tone. See FIFTEENTH.
REGULA MINOR, Lat. The ordinary open metal OCTAVE,
4 FT., yielding pure organ-tone; belonging to the foundation-work,
and corroborating the first upper partial tone of the prime 8 ft. tone
yielded by the PRINCIPAL or DIAPASON, 8 FT. See OCTAVE.
REGULA MIXTA, Lat. A compound harmonic-corroborating
stop formed of several ranks of high-pitched pipes. See MIXTURE.
REGULA PRIMARIA, Lat. An open metal stop, of full scale,
and unison pitch, yielding pure organ-tone. It forms the founda-
tion of the tonal structure of the manual department of the Organ,
being identical with the PRINCIPAL or DIAPASON, 8 FT.
REIM, Ger. The name given to a lingual stop, of 16 ft. pitch,
the voice of which is of medium strength and good mixing quality.
An example is to be found in the Pedal of the Organ in the Cathedral
of Bremen, constructed by Schulze, the celebrated organ-builder, of
Paulinzelle.
Lingual stops of comparatively soft intonation have been greatly
neglected by organ-builders and organ-designers in the stop appor-
tionment of modern Pedal Organs ; yet the value of a unison lingual
stop, that could be used alone or in combination with such stops
as the LIEBLICHGEDECKT or DULCIANA, 16 FT., could not well be
overrated.
REINFORZA A LIGNE, ItaL A stop, of 16 ft. pitch, formed
of free-reeds without the addition of resonant tubes, resembling in
this respect the PHYSHARMONIKA in the Organ in the Cathedral of
Lucerne. Italian organ-builders have introduced the REINFORZA
A LIGNE only in cases where space was too limited for the accom-
modation of a proper lingual stop of 16 ft. pitch. An example ex-
ists in the largest Organ in the Basilica of St. Peter, Rome. See
PHYSHARMONIKA.
RIPIENFLOTE, PtlLLFLOTE, Ger. A stop, of 8 ft. pitch,
yielding an unimitative flute-tone, of considerable volume without
226 ORGAN-STOPS
being unduly assertive or penetrating, used for filling up or imparting
firmness and body to combinations of unison or foundation tone in a
manual division, chiefly in the Hauptwerk or Great Organ. The
term might be applied, with propriety, to such stops as the CLARA-
BELLA or OFFENEFLOTE; also to the DOPPELFLOTE, which possesses
remarkable filling-up properties in all combinations in which it is
introduced.
REGISTRATION. Stops of the RIPIENFLOTE tonality are of great
importance in artistic registration; forming backgrounds for the
production of numerous tonal colorings which otherwise might lack
firmness or richness; as is notably the case with lingual stops of thin
quality and medium tone.
RIPIENO, Ital. The name used by Italian organ-builders to
designate a MIXTURE. As the term Ripieno signifies filling-up, it is
appropriately applied to a compound harmonic-corroborating stop
of the Organ. MIXTURES of two, three, four, and five ranks are re-
spectively labeled RIPIENO DI DUE, RIPIENO DI TRE, RIPIENO DI
QUATTRO, and RIPIENO DI CINQUE.
ROHRBORDUN, Ger. The name that has been given to a full-
scale labial stop, of 16 ft. pitch, the lower octaves of which are
formed of covered pipes, and the higher octaves of half-covered
ROHRFLOTE pipes of large scale. The stop is properly made of wood
throughout, but the two higher octaves are occasionally made of
metal, capped and tubed, or fitted with perforated wooden stoppers.
ROHRFLOTE, ROHRSCHELLE, Ger. Dtch., ROERFLUIT.
Fr., FLUTE A CHEMINEE. The names given to a half-covered stop
of metal or wood, or partly of both, usually of 8 ft. and 4 ft. pitch.
When as a Pedal Organ stop of 16 ft. pitch, it is appropriately named
ROHRFLOTENBASS. Other names are employed by German organ-
builders to designate certain members of the ROHRFLOTE family.*
The stops of both 8 ft. and 4 ft. pitch have been held in high estima-
tion and justly so by all the great German organ-builders. It is
* ' * ROHRFLOTE ist ein sehr angenehmes und wohl anwendbares Flotenwerk von Zinn, Metall
und Holz, welches zwar gedeckt ist, aber in dem Hut oder Deckel eine Rohre hat, wodurch der
Klang heller wird als bei gewohnlichen Gedaeckten. Die Pfeifen warden der besseren Intonation
wegen mit Seitenund Querbarten versehen und haben weitere Mensur als die Quintatdn. Die
Rohrflote kommt zu 16, 8, 4, 2 und i Fusston (mit letzterer Grosse unter dem Namen Rohr-
schelle) sowohlim Manual als Pedal vor. Im Pedal heisst sie Rohrflotenbass. Als Quintregister
trifft man diese Stimme zu 10 %, 5% und i ^ Fusston an, wo sie Rohrflautquinte oder Rohr-
quinte genannt wird. Es giebt Rohrfloten, welche nach Art der Doppelflote mit doppelten
Labien versehen sind und daher noch einen helleren Klang als die gewohnlichen Rohrflfiten
haben, diese heissen Doppelrohrfloten. Die Namen : Gross-, Klein-, und Superflote sind Benen-
nungen, welche die Grosse des Registers naher bezeichnen." Seidel-Kothe.
THEIR ARTISTIC REGISTRATION
227
usual to find the ROHRFLOTE, in one form or another, inserted in
their Organs of any pretensions. Of 8 ft. pitch, it is inserted on the
Second Manual and, of 4 ft. pitch, on the First Manual of the Organ
in the Cathedral of Riga; and it appears, of 8 ft. pitch, on the First
Manual of the Organ in the Cathedral of Vienna. Of 8 ft. pitch, and
made of tin, it is inserted in the second division of the Second Man-
ual of the Organ in Schwerin Ca-
thedral. It exists, of 8 ft. pitch, in
the Great and, of 4 ft. pitch, in the
Swell of the Organ in the Centen-
nial Hall, Sydney, N. S. W. It is
remarkable that this beautiful and
valuable stop should have been,
and still is, so systematically neg-
lected by French, English, and
American organ-builders. It calls
for too much labor and skill in its
formation we suppose.
FORMATION. ROHRFLOTE pipes are
made of both wood and metal, the latter
being preferred. The wood pipes are
quadrangular in form, the only radical
difference between them and the pipes
of the wood GEDECKT or BOURDON lying
in the longitudinal perforation of their
stoppers. The distinctive feature of the
metal ROHRFLOTE pipe is its peculiar cap,
to which is attached an open tube com-
municating directly with the interior of
the pipe and prolonging its air-column,
in the manner shown at A, Section 2, in
the accompanying illustration, Fig. 31.
As the relative diameter and length of
the tube, or so-called chimney, in propor-
tion to the scale and length of the body
of the pipe, affect the tone in a marked
manner, its dimensions vary considerably
in examples made by different builders.*
The internal diameter of the tube varies in ordinary examples from one-sixth to
one-third of the internal diameter of the body, while its length varies from one-
FIG. 31
* " De tous les bourdons mStalliques, il est facile de faire une Rohrflcete en percant la calotte
et y dressant une chemine'e du calibre trace par le trou qu'on vient d'y faire. ' Elle doit Stre/dit
Dom Bed os, 'd'autant plus haute qu'on la fait grosse; et plus elle est menue, plus elle doit Sire
courte.' II est juste, en eff et, de proportionner la taille & la hauteur. ' Les plus grosses, ajoute-t-
il, out la moitie" du diametre du corps de tuyau. En ce cas, elles doivent 6tre presque aussihautes
228 ORGAN-STOPS
fourth to one-half the speaking length of the body of the pipe. In the fine illus-
tration given by Dom Bedos, the tube is exactly one-third the diameter and one-
half the length of the body of the pipe. In all cases, the changes in the proportions
of the tube not only affect the pitch but also the quality of the tone produced, due
to the creation of certain inharmonic upper partial tones. The acoustical problem
involved is somewhat obscure, and is complicated by the strange fact that the
tone of the pipe is in no way affected by the tube being turned downward into the
body of the pipe, in the manner indicated at B, in Section 3, Fig. 3 1. The central
drawing, i, shows the complete pipe and the most approved form of mouth. The
rough tuning is done by moving the cap, and the fine tuning is done by bending
the large flexible ears to or from the mouth. In a ROHRFLOTE, 8 FT., it is neither
usual nor necessary to carry the pipes with tubed caps below tenor C. The bass
octave may be formed of covered pipes of wood or metal, preferably the latter.
Large scales have been adopted by both German and French builders, but it is
questionable if they are desirable in modern Organs. A scale, in the ratio i : 2.66,
giving the tenor C pipe an internal diameter of 2.62 inches, would be generally
suitable.
TONE AND REGISTRATION. The exact tone of a fine ROHRFLOTE
is difficult to describe. It is an unimitative flute-tone of a liquid,
bright, and singing quality, created by the presence of a special
combination of harmonic upper partials. Professor Helmholtz
treats of the question very slightly. He only remarks: " Narrow
stopped pipes let the Twelfth be very distinctly heard at the same
time with the prime tone, and have hence been called Quintaten
(quintam tenentes). When these pipes are strongly blown, they also
give the fifth partial [fourth upper partial], or higher major Third,
very distinctly. Another variety of quality is produced by the
ROHRFLOTE, here a tube, open at both ends, is inserted in the cover
of a stopped pipe, and in the examples I examined its length was
that of an open pipe giving the fifth partial tone of the stopped pipe.
The fifth partial tone is thus proportionately stronger than the
rather weak third partial on these pipes, and the quality of tone
becomes peculiarly bright." This is doubtless correct so far as it
goes, but it does not go far enough to account for the refined and
beautiful tones of certain ROHRFLOTES, voiced on winds of low
pressures.
The peculiar liquid and singing quality of the properly propor-
tioned and artistically voiced ROHRFLOTE, 8 FT., renders it highly
suitable for insertion in the chief accompanimental division of the
que le corps de leurs tuyaux. ' Les plus petites ont le quart et mme le demi-quart du diametre,
Le timbi e done tient i la f ois du tuyau ouvert et dubourdon ; mais il tient d'autant plus du tuyau
ouvert, que la chemine'e est plus grande et grosse; et d'autant plus du bourdon, que la chemin6e
est plus mince et basse. La douceur des Rohrflceten est toujours m61ang6e de finesse, et c'estavec
raison que souvent les facteurs les pre"ferent aux bourdons dans une grande masse de fonds de
grosse taille, parce qu'elles en relevent la rondeur." Regnier.
THEIR ARTISTIC REGISTRATION 229
Church Organ. There it would be associated with stops of a simi-
larly refined character, and would lend itself to highly effective
combinations. As, tonally, it stands midway between covered and
open unimitative flute-tones, its medium voice, combining body and
brightness, is invaluable in artistic registration, especially with such
lingual stops as the OBOE, CLARINET, FAGOTTO, and Vox HUMANA.
ROHRFLOTENQTJINTE, Ger. The name given to the ROHR-
FLOTE of 5^ ft. pitch, an example of which exists on the First Man-
ual of the Organ in St. Nicolaikirche, Leipzig, built by Ladegast
in 1862.
ROHRQUINTE, Ger. The name commonly employed by
German organ-builders to designate the half -covered stop, of 2^
ft. pitch, belonging to the ROHRFLOTE family. The stop does not
appear a common one in German Organs, but it exists on both the
Second and Third Manuals of the Organ in St. Nicolaikirche, Leip-
zig. A ROHRQUINTE, 5 J/ FT. , is inserted in the second division of the
First Manual of the Organ in the Cathedral of Schwerin. The name
ROHRNASAT has been given to the stop of 2%j ft. pitch.
SACKBUT, SAKE UT. The original wind instrument of this
name was a species of Trombone in use during the Middle Ages : and
the name was applied by Hill, of London, to a lingual stop, of 32 ft,
pitch, inserted by him in the Pedal of the Organ he erected in York
Cathedral in 1833. This was the first lingual stop of that grave
pitch introduced in an English Organ; and it remained in use until
removed by Walker, when he reconstructed the Organ in 1903, and
inserted, in its stead, the combined Contra-Trombone and Trom-
bone in the north aisle Pedal Organ.
SADT. In the "Schedule/ 7 prepared by Bernard Smith for the
stop-appointment of the Organ he built for Temple Church, London,
dated 1688, we find in the Choir list the following: "A SADT of
m ettle 61 pipes 06 foote tone." This is the only instance of the
use of the term we have been able to find. While the exact nature
and tonality of the stop are not known, it is believed to have, re-
sembled the stop now named GEMSHORN. The meaning of the term
has not been determined.
SALAMINE. This is one of the fancy names that organ-
builders have introduced for reasons only known to themselves. It
230 ORGAN-STOPS
is always desirable that a stop name should have some meaning, and
should convey either some idea of the tone or the characteristic
form of the pipes forming the stop. The present name is meaning-
less from a constructional or a musical point of view. Meyer, organ-
builder, of Hanover, gave the name to certain stops he inserted in
Organs he constructed for churches in that city. The name was
also used by Poster & Andrews, of Hull, in an Organ they built for
the Church of All Souls, Halifax. The SALAMINE in this Organ (now
removed) is stated to have been of 8 ft. pitch, formed of small-scale,
open metal pipes, yielding a delicate tone slightly inclining to stringi-
ness. This would point to a stop that would have been more ex-
pressively labeled ECHO SALICIONAL. Meaningless stop-names
should be condemned by every organ-lover.
SALICETBASS, Ger. The name given by Ladegast to an open
wood stop, of 1 6 ft. pitch, inserted in the Piano-Pedal of his Organ
in the Cathedral of Schwerin. The stop is of small scale, constructed
of pine, and voiced to carry down the tone of the SALICIONAL, 8 FT.,
on the Third Manual. The value of a soft Pedal stop of this tonality
is unquestionable.
SALICIONAL, SALICET. The names given to open labial
stops formed of cylindrical metal pipes of medium or small scales,
belonging to the VIOL family. The stops are frequently introduced
in German, French, English, and American Organs; usually of 8 ft.
pitch, but other pitches are occasionally adopted, chiefly by Ger-
man builders. Seidel says the SALICIONAL is "on the Manuals of
four, eight, or sixteen feet pitch; and in the Pedal Organ, of eight or
sixteen feet pitch, called SALICETBASS." A SALICIONAL, 8 FT., is
inserted on the First Manual, and one, of 16 ft., on the Second Man-
ual, of the Walcker Organ in the Cathedral of Vienna. In the Organ
in the Cathedral of Riga, SALICIONALS, of 16 ft. and 8 ft. are placed
on the Third Manual; a SALICET, 4 FT., is inserted on the Second
Manual; and a SALICET, 2 FT., is placed on the Fourth Manual.
The last stop is very rarely introduced. In French Organs, only the
SALICIONAL, 8 FT., seems to have been introduced; and it is usually
inserted in the Positif, as in the Organ in the Cathedral of Notre-
Dame. It appears, however, in the Bombarde of the Organ in the
Church of St. Eustache; and in the Grand-Orgue of the instrument
in the Madeleine, Paris. In English and American Organs, the
SALICIONAL, 8 FT., occupies its usual place in the Swell; but it cannot
be said to be held in the estimation it deserves, for it is, by no means
of general introduction. The omission of so valuable a stop from
THEIR ARTISTIC REGISTRATION 231
the tonal appointment of several very important Organs shows little
thought on the part of their designers. There is no SALICIONAL,
8 FT., in the Organ in St. George's Hall, Liverpool, nor in one of the
most important Concert-room Organs recently constructed in this
country: but there is a SALICIONAL, 16 FT., in the Pedal of the St.
George's Hall Organ, the only one of that grave pitch known to us
in England.
The name of the stop has been rendered in different spellings,
but that followed in this article is clearly the most correct and desir-
able.* It is also desirable to adopt Walker's mode, as shown in the
Riga Organ, applying the term SALICIONAL to the stops of 8 ft. and
1 6 ft., and the term SALICET to the stops of 4 ft. and 2 ft.
SCALE AND FORMATION. The pipes forming the SALICIONAL, when of metal,
are invariably cylindrical, varying in scale according to the ideas of the- organ-
builder regarding the most desirable tone. A satisfactory medium scale for the
stop, speaking on a wind of from 3 to 4 inches, is that adopted by F. Haas, the dis-
tinguished organ-builder, of Lucerne. This scale, in the ratio 1 : 2.66, gives the CC
pipe a diameter of 3.21 inches; the tenor C pipe a diameter of 1.97 inches ; and the
middle c 1 pipe a diameter of 1.20 inches. A slightly larger scale was favored by
T. C. Lewis, of London : this scale, in the same ratio, gives the CC pipe a diameter
of 3.34 inches; the C pipe a diameter of 2.05 inches; and the c 1 pipe a diameter of
1.25 inches. Roosevelt, of New York, used a similar scale for his beautiful SALI-
CIONALS. Various widths of mouths have been adopted and their heights have
rarely exceeded one-third their widths. The finer stops have been voiced with
harmonic-bridges, or some form of beard attached to the lower lip. Some ex-
amples of the stop are slotted and others have plain pipes, their tones being affected
accordingly. The German and French stops have usually been made of tin, or
what is described by German organ-builders as " 14 lothig Zinn " that is, an alloy
composed of 14 parts pure tin and 2 parts pure lead. We are of opinion that for
this and similar stops not requiring thick walls, the hard-rolled Hoyt two-ply
pipe metal will be found in every way satisfactory and economical.
TONE AND REGISTRATION. The tone of the SALICIONAL varies
considerably in different examples of the stop : the difference being
greater between the tones of those made by the Continental organ-
builders of the latter half of the last century and those usually made
by the English and American builders of to-day, than the difference
between the tonalities of purely modern examples, which are usually
characterized by too pungent and cutting voices. A broad survey
of the subject inclines one to ask: What is the proper tone for the
* " Le Salcional, Solcional (on dit aussi Salicional et Solicioncf), ou Salicet, est un jeu de flute
ouverte dont lestuyaux sont fort 6troits et dont les sons ont quelque analogic avec ceux de
violoncello. Ce jeu, qui se trouvait dans "beaucoup d'orgues allemandes s'est introduit depuit
peu de temps dans celles de France. " Orgue de 1'eglise royale de Saint-Denis: Rapport par J.
Adrien de La Fage, 1844.
232 ORGAN-STOPS
true SALICIONAL? A question, perhaps, not easily answered, because
the tone of such a stop, if artistically conceived, should be dictated
by the position the stop occupies in the Organ, and the nature and
tonalities of the stops with which it is directly associated, and with
which it will have, chiefly, to be combined in artistic registration.
The late T. C. Lewis, of London, who was recognized as the greatest
authority on matters of organ-tone among English organ-builders,
places the SALICIONAL third in importance in his list of manual stops.
After describing the tones of the DIAPASONS as "full, mellow,
brilliant, and powerful ; and the string-tone of the GEIGENPRINCIPAL
as "next in power to the Great DIAPASONS, and of a bright and tell-
ing quality/' he says the SALICIONAL is "another description of
small reedy and quiet DIAPASON, but still retaining the clear, life-
like quality of the preceding stops. It can be made with various
widths of mouths, and therefore of various strengths of tone, accord-
ing to the place it occupies with regard to other stops." This de-
scription points to a rounder and richer voice than that commonly
given to the SALICIONAL made to-day. The most desirable tone for
the stop would seem to be a combination of that of the true English
DULCIANA and the delicate singing string-tone of the VIOLA D' AMORE
(q. v.). Anything approaching a keen, thin, and cutting string-tone
is to be avoided; for not only is this quality furnished by the pro-
nounced and imitative string-toned stops; but because the SALI-
CIONAL has a valuable office to fulfil in artistic registration in which
such a pronounced tone would prove undesirable. The SALICIONAL
when artistically voiced, with just the proper proportion of pure
organ-tone and string-tone, is a beautiful solo stop ; while, as a body-
giver and timbre-creator, it is extremely valuable in effective regis-
tration with both open and covered flute-toned stops, and also with
the more delicately-voiced lingual stops, producing many very
charming and refined compound tones, very seldom heard in these
days of hurriedly-voiced and over-blown work.
SANPTGEDECKT, SANFTFLOTE, Ger. The names that
have been given to covered stops of wood or metal, and commonly of
8 ft. pitch, yielding a quiet unimitative flute-tone. They properly
belong to the LIEBLICHGEDECKT family, but are softer in tone. Stops
of this quality would be admirably suited for insertion in the true
Chamber Organ or the Echo Organ of larger instruments.
SAXOPHONE. The name which has been given, in some few
instances, to stops made in both lingual and labial forms. Several
essays have been made by ingenious pipe-makers and voicers to
THEIR ARTISTIC REGISTRATION 233
construct an organ-stop which shall imitate in a satisfactory manner
the peculiar compound tones of the single-reed instruments of the
Saxophone family. All attempts hitherto made in the direction of
lingual stops have fallen short of being satisfactory, for it has been
found difficult, by the employment of either striking- or free-reeds
to produce the rich compound tones of the brass Saxophones. While
the Saxophone strictly belongs to the Clarinet family, and is fitted
with the single reed of the Clarinet, its tone is decidedly sui generis.
On carefully studying the tone, one finds it to be a remarkable com-
pound of orchestral reed- and string-tone, with a slight admixture of
flute-tone, a remarkable combination and one difficult, if not im-
possible, to imitate in a single lingual stop. Dr. W. H. Stone, speak-
ing of the instrument, says: "The Saxophone, though inferior in
compass, quality, and power of articulation to the Clarinet, and
Bassethorn, and especially to the Bassoon, has great value in mili-
tary combinations. It reproduces on a magnified scale something
of the Violoncello quality, and gives great sustaining power to the
full chorus of brass instruments, by introducing a mass of harmonic
overtones." It would seem highly probable that the SAXOPHONE
of the Organ will in its most satisfactory form be dual, constructed of
lingual and labial pipes of CORNO DI BASSETTO and VIOL tonalities.
The very few stops which have appeared in English Organs, under
the name SAXOPHONE, have been merely full-tone CLARINETS, and,
accordingly, by no means satisfactory.
Up to this point we have alluded to lingual stops only. We have
now to speak of the wonderfully imitative labial SAXOPHONE, in-
vented by W. E. Haskell, America's most distinguished artist in
labial pipe formation and voicing; and first introduced, in 1897,
in the Organ in the Church of the Holy Trinity, Philadelphia,
Pa. This remarkable stop is of 8 ft. pitch and extends throughout
the compass of the clavier. It is formed entirely of straight, quad-
rangular wood pipes, of small scale, the CC pipe measuring inter-
nally 3-5^- inches in width by 4^- inches in depth; the ratio of the
scale apparently being i: 2.66. The pipes have sunk blocks, in-
verted mouths, beveled caps, and are fitted with cylindrical har-
monic-bridges.*
Voiced on wind of 3^ inches, this fine stop yields a compound
tone so closely imitative of that of the true Saxophones as to be posi-
tively deceptive to the ear. The tones of the Saxophones are thus
* Further details of formation, accompanied by illustrations of the sound-producing portion
of a pipe, are given in "The Art of Organ-Building, " Vol. II., p. 485; and in "The Organ of the
Twentieth Century," pp. 450-1.
234 ORGAN-STOPS
described by Berlioz as possessing "most rare and precious qualities.
Soft and penetrating in the higher part, full and rich in the lower
part, their medium has something profoundly expressive. It is, in
short, a quality of tone sui generis, presenting vague analogies with
the sounds of the Violoncello, of the Clarinet, and Corno Inglese,
and invested with a brazen tinge which imparts a quite peculiar
accent. " We were naturally very doubtful regarding the possibility
of producing so complex a tonality from wood labial pipes: but all
doubts were put to rest, on our being afforded the means of judging
by direct comparison of the tones of the stop with those of the true
Saxophone, performed upon within the Organ immediately along-
side the stop. The imitation was practically perfect; while in cer-
tain parts of the compass the SAXOPHONE of the Organ was more
even and pleasing than the reed instrument. This is only one of the
inventor's notable achievements in wood pipe formation and voicing
as these pages show.
SCHALMEI, Ger. Ital, SCIALUMO, A lingual stop, of 8 ft.
pitch, the tone of which is supposed to imitate the voice of the old
Schalmei or Shawm, the precursor of the Clarinet, and an instru-
ment commonly used during the Middle Ages. See CHALUMEAU.
SCHARF, Ger. Dtch., SCHERP. A compound harmonic-
corroborating stop, composed of three or more ranks of metal pipes,
of high pitch and moderate scale, voiced to yield a bright and sharp
tone, hence its name. Alluding to the old stop, Wolfram says it was
usually of three ranks, starting with a I5th, igth, and 22nd. Seidel,
on the other hand, says the SCHARF differs from the ordinary MIX-
TURE by having one of its ranks third-sounding; and gives the start-
ing composition for the three-rank stop, I5th, lyth, and igth; for
the four-ranked stop, isth, 17th, igth, and 22nd; and for the five*
ranked stop, I2th, isth, i;th, and 22nd. The introduction of the
third-sounding rank adds greatly to the sharp intonation of the stop,
especially under the somewhat crude system of voicing of compound
stops followed by the German organ-builders, who made the SCHARF
of pipes of too large scales, and voiced it too loud and piercing in
tone. It seems strange that the old German builders either mis-
understood or systematically ignored the true and scientific office of
the compound stops. And it seems equally strange that the organ-
builders of to-day are omitting, or discouraging the introduction of,
the compound harmonic-corroborating stops in modern Organs.
Is it through ignorance of their invaluable office in the tonal struc-
ture of the Organ? Or is it the narrow trade desire to avoid having
THEIR ARTISTIC REGISTRATION 235
to make such stops, requiring both scientific knowledge and high
artistic skill in their proper formation? It unquestionably must be
one or the other, and neither is creditable to the organ-builders of
the twentieth century. See ACUTA.
SCHARFFLOTE, Ger. A metal labial stop, of 4 ft. pitch,
voiced to yield a bright and piercing flute-tone. It is valuable, in
the absence of a COMPENSATING MIXTURE, for imparting brightness
to the Pedal Organ. A SCHARFFLOTE, 4 FT. , exists in the Pedal of the
Organ in the Cathedral of Merseburg.
SCHARFREGAL, Ger. An old and obsolete lingual stop, of
4 ft. pitch, and very keen intonation, as its name implies. See
REGAL.
SCHLANGENROHR, Ger. An old and now disused name for
the lingual stop, of 16 ft. pitch, commonly inserted in the Pedal
Organ, which is better known as the SERPENT (q. v.).
SCHGNGEDECKT, Ger. The name that has sometimes been
used to designate a small-scaled covered stop, yielding a beautiful
tone. The name is practically synonymous with LIEBLICHGEDECKT.
The prefix Schon signifying beautiful has been applied to other
names of labial stops.
SCHREIER, SCHREIERPFEIFE, Ger. The names given
by old German organ-builders to compound labial stops yielding, as
the name implies, screaming tones; and also to a labial stop of a
shrill and penetrating flute-tone, an example of which is said to exist
in the Organ of the Barfusskirche, at Erfurt. The stop was usually
of three ranks, and was simply a large-scaled and loudly-voiced
ACUTA (q. z>.).* Stops of this noisy class are very objectionable, and
have too often been introduced by inartistic organ-builders.
SCHUFFLET, Ger. An old name given to the mutation har-
monic-corroborating stop, of iJ/ ft. pitch, representing the fifth
upper partial tone of the fundamental manual unison, 8 ft. pitch.
A stop of this name existed in the old Organ in the Church of St.
Lambert, Munster.
SCHWEIZERFLOTE, SCHWEIZERPFEIFE, Ger. Liter-
* ' * SCHREIER, SCHREIERPFEIFE, Schryari, ist eine veraltete, gewohnlich 3 f ache Mixtur, welche
aus Oktavchoren zu i', J# und % r besteht. Sie wurde, also disponiert, mit der Cymbel ein und
dasselbe Register sein, wenn nicht vielleicht eine Verschiedenheit in der Mensur und namentlich
in der Intonation obwaltet. Auch 2 f ach soil diese Stimme gefunden werden, wo sie aus 2' und i'
besteht. Der Ton dieses Registers muss, dem Namen nach, sehr grell sein. " Seidel-JCothe,
236 ORGAN-STOPS
ally Swiss Flute. An open labial stop of 8 ft., 4 ft., 2 ft., and rarel>
i ft. pitch; the pipes of which are cylindrical, of small scale, and
have low mouths, yielding a yoice between a flute- and string-tone,
of a refined and pleasing character. In good work, the pipes are
made of tin. Seidel says: " Dieses Register findet man sowohl im
Manual als im Pedal, wo es alsdann SCHWEIZERPFEIFBASS oder
SCHWEIZERBASS heisst. Im Manual trifft man diese Stimme zu-
weilen nur in den oberen Octaven an, wo sie unter dem Namen
SCHWEIZERPFEIFDISKANT vorkommt." Locher says there is no
foundation for the name SCHWEIZERFLOTE, any more than for the
name WIENERFLOTE; and "in spite of its name it belongs to the
string family, as for example in the Great Organ of the instrument
in Magdeburg Cathedral, where it assumes the form of a full-toned
GAMBA."
SCHWIEGEL, SCHWAGEL, Ger. The term that has been
employed by old German organ-builders to designate metal labial
stops of different formation and intonation. According to Schlim-
bach, the stop belongs to the Flute-work, and is of the scale of the
QUERPFEIFE, with the intonation of the BAUERFLOTE. Seidel, on
the other hand, describes the stop as formed of pipes having cylin-
drical bodies surmounted by truncated cones; resembling those of
the FLACHFLOTE or SPILLFLOTE. Respecting the tone, he remarks:
<k Der Klang dieser Stimme ist angenehm und dem Klang der Quer-
flote ahnlich, aber er soil noch etwas sanfter als der Ton der Spill-
flote sein." The stop was made of 8 ft., 4 ft., 2 ft., and I ft. pitch;
that of 4 ft. pitch being termed SCHWIEGELDISKANT or DISKANT-
SCHWIEGEL. An example, of 8 ft. pitch, was introduced in the Echo
of the Organ built by Jagermann, of Dresden, for the Kreuzkirche
in that city.
SEPTADECIMA, Lat. Eng. SEVENTEENTH. The mutation
stop, of i % ft. pitch, corroborating the fifth upper partial tone of
the manual unison prime, 8 ft. See SEVENTEENTH.
SEPTIEME, Fr. The mutation harmonic-corroborating stop,
of 4 4 /? ft. pitch in the Pedal Organ, and 2 2 /7 ft. and i l fy ft. pitch in
the manual divisions. The pipes forming the stops are of open
metal, cylindrical in form, of medium scale, and voiced to yield soft
pure organ-tone. The SEPTIEME represents the sixth upper partial
tone of the prime tone, produced by seven times the number of
vibrations that belong to the prime tone. Thus, if the CC prime
tone is of 64 vibrations per second, the sixth upper partial tone,
THEIR ARTISTIC REGISTRATION 237
lying between aft 1 and bb 1 (of the physical scale), is yielded by a
pipe of say i 1 /? feet speaking length, having 448 vibrations per
second. Although very much higher partial tones are corroborated
by stops in the Organ, the sixth upper partial tone has been very
seldom corroborated by the introduction of the SEPTIEME. The
only example, in complete form, in an English Organ, known to us,
is to be found, under the name " SHARP TWENTIETH," in the instru-
ment in the Collegiate Institution of Liverpool, built by Jackson, of
that city, in 1850. We know of no example in an American or Ger-
man Organ. The only Organ in which the SEPTIEME has been
systematically and scientifically introduced is that built by Cavaille-
Coll, in 1868, for the Cathedral of Notre-Dame, Paris. In the Pedal
Organ the stop is inserted, of 4 4 /7 ft. pitch, belonging to the 32 ft.
harmonic series. In the Clavier des Bombardes it appears, of 2 2 /7
ft. pitch, belonging to the 16 ft. harmonic series: and in the Grand
Choeur it is introduced, of i 1 /? ft. pitch, belonging to the 8 ft. har-
monic series.
This systematic introduction of the SEPTIEME in all the har-
monic series was due to Aristide Cavaille-Coll's scientific knowledge
and researches in tone production ; and it is to be regretted that we
do not see evidences of a similar knowledge and investigation in
even the more important Organs built to-day. It is, however,
hardly to be expected that such a stop as the SEPTIEME should ap-
pear in Organs in which very little attention is paid to the provision
of even a reasonably adequate harmonic structure. In the foun-
dation-work of an important Organ the SEPTIEME should certainly
be introduced, preferably as a rank in a through compound stop in
which it can be correctly adjusted tonally. The following four-
rank CORNET would be a favorable stop for its introduction:
CORNET IV. RANKS.
I. SEVENTEENTH Metal. 1 3/s Feet
II. NINETEENTH Metal. 1 1/3 "
III. SEPTIEME Metal. iV? "
IV, TWENTY-SECOND .... Metal, i "
The pipes forming the stop to be of medium scale, and voiced
to yield pure organ-tone. Ranks II. and IV. to be the most pro-
nounced in strength of voice; rank I. to be softer; and rank III. to
be regulated so as not to be unduly assertive. It is in the voicing
and regulating of such an important stop that the artist can show
himself.
SERAPHONFLOTE, Ger. The name given to a flute-toned
238 ORGAN-STOPS
stop, of 8 ft. pitch, and very powerful voice; invented by "W. P.
Weigle, of Stuttgart. An example exists in the Organ of St. Sebal-
duskirche, Nurnberg. Under the name SERAPHONPFEIFE, a similar
stop was, for the first time, introduced in the Organ constructed by
Steinmeyer for the City Church of Wertheim, Baden.
The pipes forming these stops are of metal, cylindrical in form
and of large scale. They have two mouths, formed in the usual
manner, and placed as close to each other as practicable.* The
large lineal measurement of the combined mouths about four-
tenths of the circumference of the pipe affords the opportunity for
the production of a remarkable volume of tone. The stop so formed
is practically a metal DOPPELFLOTE.
TONE AND REGISTRATION. A powerful stop of this class, pro-
ducing a pronounced unimitative flute-tone, would be of consider-
able value in two manual expressive divisions of the Concert-room
Organ; namely, the Solo Organ and the expressive portion of the
First or Great Organ, where its voice would be under control. In
the former it would impart great body, solidity, and color to the
more powerful lingual stops, while it would give great dignity to the
Flute-work. A powerful normal flute-tone is absolutely necessary
for the production of broad effects and tone-coloring in a properly
stop-apportioned Solo Organ. In the expressive subdivision of the
First Organ, a fine SERAPHONFLOTE would be extremely valuable,
as it would, under control as to its strength of voice, combine per-
fectly with the pure organ-tones of the PRINCIPALS or DIAPASONS;
increasing their volume, and infusing into their somewhat monot-
onous voices an agreeable coloring by the introduction of certain
of the lower harmonic upper partial tones. The stop would
also prove highly effective in registration with the lingual stops
which are enclosed along with it; and, generally, in all full-toned
combinations.
SERPENT. ItaL, SERPENTINO. A lingual stop, of 16 ft. pitch,
the tone of which is between those of the BASSOON and the TROM-
BONE stops of the Organ. It is supposed to imitate in its voice the
old instrument invented by Edme Guillaume, a Priest of Auxerre,
France, in the year 1590. The instrument was an improvement on
the older Bass Zinken. It was a conical tube of wood, covered with
leather, and bent into the form of a serpent, hence its name. It was
played with a mouthpiece, not with a reed. Examples of the stop,
* Further particulars respecting this stop, accompanied by illustrations, as patented by G.
F. Weigle, are given in "The Art of Organ-Building," Vol. II., pp. 534-6.
THEIR ARTISTIC REGISTRATION 239
of 1 6 ft. pitch, exist in the Pedal of the Organ in the Cathedral of
Uim; and in the Schwell-Pedal of the Organ in the Cathedral of
Riga. In the latter it would seem to be a species of DOUBLE CORNO
DI BASSETTO. Under the name SERPENTINO, a similar stop exists
in the Organ in the Church of Sanctissimo Crucifisso, at Como.
An artistically-voiced CORNO DI BASSETTO, 16 FT., would be a
valuable addition to the tonal forces of the Pedal Organ of the
Twentieth Century. The name SERPENT might be retained for such
a stop.
SESQUIALTERA, SEXQUIALTERA. The compound har-
monic-corroborating stop, formed of two or more ranks of open
metal pipes of medium scales, yielding, in the foundation- work, pure
organ-tone. The true SESQUIALTERA consists of two ranks of pipes
only, carried throughout the compass of the clavier without a
break; the ranks standing at the interval of a major sixth apart.
This interval is secured by placing a fifth-sounding rank below a
third-sounding one; as, in the eight feet harmonic series, a TWELFTH
2% FT., and a SEVENTEENTH, i% FT., sounding G e 1 on the CC
key. On the First Manual of the Organ in the Cathedral of Riga
there is a SEXQUIALTERA, belonging to the sixteen feet harmonic
series, formed of a QUINT, 5^ FT., and a TIERCE, 3% FT. ; while on
the same manual there are, in addition, independent stops of the
same pitches, but of different tonality a unique apportionment, so
far as our knowledge extends. As independent stops, the QUINT, 5^
FT., and TIERCE, 3)^ FT., are inserted on the First Manual of the
Organ in the Cathedral of Ulm; and they appear in other Walcker
Organs. A QUINT, 2^ FT., and TIERCE, i% FT., as independent
stops, are inserted on the Second Manual of the Riga Organ. When
a SESQUIALTERA, belonging to either the eight feet or sixteen feet
harmonic series, is placed in any manual division, its ranks should
be made to draw separately.
The SESQUIALTERA belonging to the thirty-two feet harmonic
series is necessarily rare; but one, formed of a QUINT, 10% FT., and a
TIERCE, 6 2 /s FT., exists in the Pedal of the Riga Organ; and in addi-
tion there is a QUINTBASS, 10% FT., and a TERZBASS, 6 2 /s FT. This
grand Pedal Organ has a remarkable harmonic structure; and in this
direction affords a valuable lesson to organ-designers of to-day.
The so-called SESQUIALTERAS, of the old English organ-builders,
were invariably harmonic-corroborating stops, formed of several
rknks of open metal pipes, requiring two or more breaks in their
compass. In these stops, the sexts did not invariably obtain in
240 ORGAN-STOPS
every break ; those omitted being formed by the addition of an in-
dependent TWELFTH, 2% FT. An example of this incomplete form
of SESQUIALTERA is furnished by the stop inserted by John Snetzler
in the Organ he constructed, in the latter part of the eighteenth
century, for St. Mary's Church, Nottingham. The composition of
the stop is here given :
SESQUIALTERA IV. RANKS.
CC to G 15 17 19 22.
G# to g 1 12* 15 17* 19-
g# J to top 8 12* 15 17*.
The following example of a five-rank SESQUIALTERA, in which
sexts of different pitches obtain in the two breaks, gives the com-
position of the stop inserted in the Great of the Organ constructed by
Harris and Byfield, of London, in 1740, for the old Parish Church of
Doncaster.
SESQUIALTERA V. RANKS.
CC tO C 1 . . . 19* 22 24* 26 20.
c# r to top ... 8 12* 15 17* 19.
The sexts are indicated by the asterisks. This SESQUIALTERA is
here given in modern compass: the original stop in the Harris-By-
field Organ had the old compass of GGG to d 3 . It is interesting to
note that in this Organ the SESQUIALTERA was accompanied by a
TWELFTH, 2% FT., and a TIERCE, i% FT., both complete and in-
dependent stops. A good four-rank SESQUIALTERA can be formed
by omitting the acute fifth-sounding rank in both breaks,
TONE AND REGISTRATION. The tone of the SESQUIALTERA may
properly vary with the stop-apportionment of the division of the
Organ in which it is inserted. When associated with the foundation-
work, in the First or Great Organ, it should, as essentially a har-
monic-corroborating stop, yield pure organ-tone: but when inserted
in any other manual division, its tone may properly be dictated by
the stop-apportionment with which it is associated. When con-
sidered desirable, the SESQUIALTERA may be formed of through
ranks of pipes of special and different tonalities, becoming a timbre-
creating stop. In all cases it is desirable that, in its voicing and
regulating, the third-sounding rank should be made subordinate in
tone to the fifth-sounding rank, so as to avoid undue assertiveness.
The value of harmonic-corroborating stops of the SESQUIALTERA
class in the production of distinct tonal coloring deserves to be better
THEIR ARTISTIC REGISTRATION 241
known. The neglect of such stops in the tonal appointment of
modern Organs is a mistake; and seriously cripples the means of
producing compound tones of great variety and beauty. The regis-
tration of unison, octave, and super-octave stops alone, whatever
their tonal character may be, must fail to satisfy the cultivated ear
of the musician seeking to paint his tone-pictures in rich musical
chiaroscuro. It is the tradesman, not the artist, who is to-day
omitting such stops from the Organ.
SEVENTEENTH. Fr., TIERCE. Ger., TERZ, TERTIE ItaL,
DECIMA SETTIMA. A third-sounding mutation stop, formed of open
metal pipes of medium scale, of i y$ ft. pitch in the manual divisions,
standing at the interval of a seventeenth above the unison, 8 ft., and
a major third above the SUPER-OCTAVE, 2 FT.* The SEVENTEENTH
represents and corroborates the fourth upper partial tone of the
prime or unison tone In the Pedal Organ, the SEVENTEENTH is of
33/5 ft. pitch, belonging to the 16 ft, harmonic series. When a stop of
3/^ ft. pitch is introduced in a manual division, it is termed by
French organ-builders GROSSE TIERCE, | as in the Clavier des Bom-
bardes of the Cavaill-Coll Organ in the Church of Saint-Sulpice,
Paris. A GROSSE TIERCE, 6% FT., exists in the Pedal of the Organ
in the Cathedral of Notre-Dame. A TERZ, 3)^ FT., is placed on the
First Manual, and a TERZBASS, 6% FT., in the Haupt Pedal of the
Walcker Organ in the Cathedral of Riga. A SEVENTEENTH, 6% FT.,
under the name TERTIA, exists in the Pedal of the Church at Perle-
berg: this stop is formed of wood pipes.
The SEVENTEENTH, 6% FT., made either of open metal or wood
pipes, is of considerable value in the Pedal Organ, brightening and
enlivening its normal foundation tone: here it may be more pro-
nounced in tone than is desirable for third-sounding stops in the
manual divisions. This SEVENTEENTH belongs to the 32 ft. har-
monic series.
*"La TIERCE, jadis nomme'e par certains auteursSwgwj-Ocfatfl et mieux Sesqui-Quarta, est
unjeuouvertdedix-neufpouces,oucinquante-uncentimetres,degrossetaille r detain oud'etoffe
selon qu'on veut lui donner ou lui 6ter du tranchant. II parle a la Tierce de la Doublette, par con-
sequent a la dixieme du Preslant; c'est pourquoi les Italiens 1'appellent quelquefois Decima. Au
positif , on la designe par le nom de Petite-Tierce, en opposition avec celle du grand orgue quiest
de taille plus forte, mais de mme degre. Les Allemands 1'appellent Terz et Tertia, et encore Des t
abr6g6 de Decima." Regnier.
f " La GROSSE-TIERCE parle a r octave inf erieure de la Tierce ordinaire, c'est-a-dire a la Tierct
du Preslant, et a la dixieme du huit-pieds. Elle est grosse taille, tout ouverte et d'etoffe, I'Stak
serait trop mordant. Pour que ce registre fasse bon effet, il lui faut associer une grande mass*
de f onds, surtout en hurt pieds; un bourdon de seize ne nuira point. Le plus grand tuyau de Is
Grosse-Tierce a trois-pieds deux pouces, lorsqu'elle se trouve aux claviers de la main. En AUe
magne, on la trouve en p6dale de six-pieds, sous le nom de Decem-Bass" Regnier.
242 ORGAN-STOPS
SHARP MIXTURE. The term that has been used to designate
a compound harmonic-corroborating stop formed of four or more
ranks of open metal pipes, of high pitch, and penetrating tone. The
SHARP MIXTURES of the old English organ-builders appear to have
been invariably formed of octave- and fifth-sounding ranks; third-
sounding ranks being confined to their SESQUIALTERAS.
Properly scaled, voiced, and regulated, a SHARP MIXTURE would
be a useful stop, in either the wood-wind or brass-wind division of
the artistically appointed Concert-room Organ, for the production
of ringing qualities of compound tone. But as MIXTURES are made
to-day by the generality of organ-builders, without regard to the
dictates of science or art, the SHARP MIXTURE had much better be
omitted altogether from the Organ.
SIFFLOTE, SIEFFLOT, Ger. Fr., SIFFLET. A small open
metal stop, of medium scale, and 2 ft. and i ft. pitch, yielding a clear
unimitative flute-tone. In some rare instances it has been intro-
duced of i J^ ft. pitch. The stop, of 2 ft. pitch, exists on the Second
Manual of the Organ in the Protestant Church, Mulhausen; and,
of i ft. pitch, in the Oberwerk of the Organ in the Cathedral of
Merseburg.
SINGENDREGAL, Ger. An old lingual stop, of 8 ft. pitch,
which received its appellation on account of the singing character of
its voice. Both the stop and name are obsolete. See REGAL.
SOAVE, Ital. The name given to an open labial stop, of 8 ft.
pitch, yielding an unimitative flute-tone of great softness and
beauty.* It is to be regretted that unison stops of this refined
tonality are so greatly neglected in the appointment of modern
Organs.
SORDUN, Ger. Fr., SOURDINE. Ital., SORDINI. The name
given by old Continental organ-builders to a covered labial stop, of
1 6 ft. and 8 ft. pitch, the pipes of which were commonly made of
wood, and so treated as to produce a subdued or muffled tone;
considered desirable before the swell was invented, f
*" SOAVE (ou suabile), htdt-pieds, de grosse taille, lent parler, mais tres-doux et gracieux,
commel'mdique son nom italien. On le de"signe aussi sous le nom d'englische Flcete, flute-ange~li-
que, qu'il ne faut pas confondre avec le jeu d' anche, qui porte a peu pres le nom de angelica vox."
Regnier.
t" SORDINI (ital.), Sordun (all.), Sourdine. C'est un ancien bourdon de huit, ou de seize,
selon qu'il se trouvait k la main ou & la pe"dale. Les tuyaux 6taient caches chacun dans une boite,-
pDur rendre un son plus e"tient. " Regnier.
THEIR ARTISTIC REGISTRATION 243
SORDUNREGAL, Ger. An old lingual stop of the REGAL
family, of 8 ft. pitch, softly voiced, and probably muted by being
inclosed in some form of box. Both the stop and its name are now
obsolete. See REGAL.
SPILLFLOTE, SPINDELFLOTE, Ger. Eng., SPINDLE FLUTE.
A half-covered labial stop, of 8 ft., 4 ft., and 2 ft. pitch. The
name is expressive of the peculiar form of its pipes, the bodies of
which are cylindrical surmounted by long conical portions, trun-
cated, leaving small openings at top. The form of the
pipe, including its tapering foot, somewhat resembles
that of a covered spindle, and has suggested the name.
The form is shown in the accompanying illustration, Fig.
32. As it is not possible to tune the pipe by any manip-
ulation of its small top orifice, without altering its tone,
the fine tuning is done by means of large projecting ears,
bent toward or from the mouth. As its name implies, the
SPILLFLOTE belongs to the Flute-work; but its tone is
bright and without any distinctive flute quality, lying
between the tones of the ROHRFLOTE and the SPITZFLOTE.
Seidel compares it with the SCHWIEGEL (q.v.). Under the
hands of a skillful and artistic voicer, and probably with
the addition of the harmonic-bridge, there is little doubt
but the SPILLFLOTE pipe could be made to yield both a
characteristic and a beautiful compound tone. The stop
seems to have been introduced during the second quar-
ter of the sixteenth century; but has become practically
a curiosity in organ-building, doubtless due to the time
and trouble involved in its construction.
SPITZFLOTE, Ger. Fr., FLUTE A POINT, FLUTE A
FUSEAU. Eng., SPIRE FLUTE. A metal labial stop, of 8ft.,
4 ft., and 2 ft. pitch; which derives its names from
the form of its pipes, which is that of a slender truncat-
ed cone, open at top; closely resembling the form of the
GEMSHORN pipe (See Fig. 19, Plate III). The stop is
justly held in high estimation by German organ-builders,
being introduced, usually of 8 ft. and 4 ft. pitch, in all FIG. 32
their more important Organs. In Walcker's Organ in
the Cathedral of Vienna, it is inserted, of 8 ft. and 4 ft. pitch, on
the Second Manual. In the Organ in the Cathedral of Riga, it is
inserted, of 4 ft. pitch, on the Third Manual.
244 ORGAN-STOPS
SCALE AND FORMATION. The scale of the SPITZFLOTE varies in different ex-
amples and according to the character and strength of the tone desired ; and the
proportion of the diameters at the mouth line and at the top of the pipe also varies
slightly. The dimensions given by Topfer, and apparently adopted by Friedrich
Haas, the rebuilder of the Lucerne Organ, are practically as follows: For the CC
8 ft. pipe, 4.60 inches diameter at the mouth line, and 1.50 inches diameter at the
open top. The diameter at top is, accordingly, a trifle under one-third of the
diameter at the mouth line, approaching closely the proportion recommended
for the GEMSHORN pipe. But Seidel says the SPITZFLOTE pipes are more pointed
than those of the GEMSHORN; and this would seem desirable for the production of
the characteristic tone of the stop. The diameter at top should, however, never be
less than one-fourth of that at the mouth. The width of the mouth may be one-
fourth or two-ninths of the larger circumference; and its height may range from
one-fourth to one-third of its width, according to the wind-pressure and the tone
desired. As in certain half-covered stops, the upper lip of the mouth may be
slightly arched, and the lower lip and languid nicked moderately fine. As the pipes
must not be slotted at top, or coned in tuning, fine tuning must be done at the
mouth by means of flexible ears. While the 4 ft. and 2 ft. stops have invariably
been made of metal throughout, the bass octave of the 8 ft. stop has been made of
pyramidal pipes of wood. This practice is not to be recommended.
TONE AND REGISTRATION. The voice of the true SPITZFLOTE
is compound, which, owing to its delicate harmonics, partakes of
both flute- and string-tone; inclining to either one or the other
according to the manner in which the pipes are voiced, and, to some
extent, according to their scales. This variable compound tone is a
valuable property of the stop, as the artist voicer can adapt it to
suit the tonal apportionment of which it is to form a part. We know
that such refinement in tonal adjustment is paid very little atten-
tion to in this age of commercial organ-building; but it is, never-
theless, just such refinement, carried consistently throughout the
stop-work of an Organ, that marks the artist-craftsman and makes
the Organ a work of art.
It must be obvious to everyone conversant with compound tone
production, that such a stop as the SPITZFLOTE, standing midway in
voice between the flute-toned and string-toned stops, must lend it-
self to the production of refined colorings in registration with the
softer-voiced stops, both labial and lingual. With the latter es-
pecially, to which it will impart a desirable body and richness with-
out destroying their characteristic tonalities. The true SPITZFLOTE,
8 FT., combines perfectly with the CLARINET, CORNO DI BASSETTO,
FAGOTTO, and COR ANGLAIS, producing beautiful compound tones.
It also adds richness and body to the imitative string-toned stops;
and, in both 8 ft. and 4 ft. pitch, a certain crisp fullness to the imita-
tive flute-toned stops. The SPITZFLOTE should find a place in the
Choir of the Church Organ; and in the Second or accompanimental
THEIR ARTISTIC REGISTRATION 245
Organ of the properly stop-apportioned Concert-room Organ.
SPITZQUINTE, QUINTSPITZ, Ger. A SPITZFLOTE, of
ft. and i^i ft. pitch, the pipes of which are formed in all respects
similar to those of the unison SPITZFLOTE (q. v.). As a harmonic-
corroborating and timbre-creating stop, the SPITZQUINTE may be
found valuable in special stop-apportionments: and, of 2% ft. pitch,
it may, with advantage, take the place of the ordinary TWELFTH,
2% FT., in Organs of small size; lending itself more effectively to
artistic registration, and proving more generally useful in combina-
tion.
STENTORFLOTE, Ger. The name given by Herr Weigle to a
loudly-voiced, high-pressure, labial stop, of 8 ft. pitch, the pipes of
which are constructed in accordance with his patented system.* An
example exists on the First Manual of the Walcker Organ in the
Synagogue, Strasbourg.
STENTORGAMBA. The name used by Walcker to designate
a loudly-voiced, string-toned, labial stop, of 8 ft. pitch, the pipes of
which are constructed in accordance with the high-pressure, Weigle
system. An example exists on the Second Manual of the Organ in
the Synagogue, Strasbourg. Under the name SOLO-GAMBE, a simi-
lar stop exists on the Second Manual of the Weigle Organ in the
Grand Hall of the Liederhalle, Stuttgart.
STENTORPHONE. The name, derived from the Greek
SiivTwp Stentor, and <PUVTQ voice or sound, and employed to
designate a large-scaled metal labial stop, of 8 ft. pitch, the pipes of
which are cylindrical in form and made of thick metal, so as to with-
stand the powerful pulsations of the columns of air within them,
generated by the high pressures employed in the voicing. Properly
this wind-pressure should range between seven and ten inches; but
during the present prevailing craze for high wind-pressures and loud
voicing, such reasonable pressures will be considered insufficient.
Properly voiced, the STENTORPHONE yields a tone of great breadth,
richness, and dignity. True to its name, it should be the most
stentorian labial stop introduced in the Organ. Its large scale and
its tonality places it, strictly considered, in the Flute-work of the
Organ. This position is shown by the fact that the best substitute
for the true STENTORPHONE, at present known, is a large-scaled
* Royal Letters Patent, Great Britain, No. 17718. United States Patents, Nos. 457686 and
520344-
246 ORGAN-STOPS
open wood flute-toned stop, to which the name TIBIA PLENA has
been given.
The powerful voice of the STENTORPHONE naturally points to the
only two manual divisions in which it could be properly introduced;
namely, the First or Great Organ and the Solo Organ. In its full
tonality it is of most value in the latter, imparting a remarkable
breadth and grandeur to the tones of the powerful lingual stops
properly inserted there; and that without unduly increasing their
assertiveness. In the First Organ, it should be inserted in the ex-
pressive subdivision, where its dominating voice will be under con-
trol. There it will lend itself to the production of impressive
crescendoes, and impart great dignity to the TRUMPETS.
Examples of the metal STENTORPHONE, 8 FT., appear in several
American Organs, notably in those by Roosevelt. He inserted it in
the Solos of the Organs in the Auditorium, Chicago; and the Cathe-
dral of the Incarnation, Garden City, L. I. A fine STENTORPHONE
is inserted in the Solo of the Organ in the Cincinnati Music Hall.
We inserted one in the Fourth Organ of our scheme for the Grand
Concert Organ installed in the Festival Hall of the Louisiana Pur-
chase Exposition (1904) ; now in the Wanamaker Store, Philadelphia,
Pa. The STENTORPHONE has been introduced in certain German
Organs. An example exists in the Organ in the great Concert Hall
in Mannheim; and Walcker has inserted one, under the name STEN-
TORFLOTE, 8 FT., on the First Manual of the Organ in the Synagogue,
Strasbourg. The STENTORPHONE is one of the large-mouthed, high-
pressure, objectionable stops introduced and patented by Weigle,
an example of which exists in the Organ he constructed, in 1895, for
the Liederhalle, Stuttgart. The STENTORPHONE has not commended
itself to either French or English organ-builders. We known of no
example in a French Organ; and of only one in an English Organ,
and that, strange to relate, in a Chamber Organ.
STARKGEDECKT, Ger. A large-scaled covered stop, of 16
ft. pitch, the pipes of which are of wood, copiously blown with wind
of moderate pressure, and voiced to yield a full and round tone of
good mixing quality. Of this stop, Regnier says: "Grand bourdon
de seize fortement embouche et donnant aux flfites ouvertes de huit
un velout et une profondeur remarquables. "
STILLCcEDECKT, Ger. This name, which signifies a quiet-
toned covered stop, is employed by German organ-builders to desig-
nate stops which are softer in their intonation than the LIEBLICH-
GEDECKTS. The STILLGEDECKT is usually made of wood, and of 8 ft.
THEIR ARTISTIC REGISTRATION 247
and 4 ft. pitch. The pipes are of a small scale, and speak on low-
pressure wind. The stop is highly suitable for insertion in the softer
toned manual divisions of the Concert-room Organ, and in the chief
accompanimental division of the Church Organ; it is also an ideal
stop for the true Chamber Organ. An example, of 8 ft. pitch, exists
in the Choir of the Organ in the Town Church of Fulda; and one, of
4 ft. pitch, is inserted in the Echo of the Organ in the Church of
Waltershausen.
STOPPED DIAPASON, Eng. Ger., GEDECKT. Fr., BOURDON.
Span., TAPADO, TAPADILLO. A covered labial stop, of 8 ft. pitch in
the manual divisions, and of 16 ft. pitch in the Pedal Organ. The
English name, though time-honored, is neither correct nor desirable,
for the stop has no resemblance to a true DIAPASON in form or ton-
ality. The stop yields an unimitative flute-tone, and belongs to the
Flute-work of the Organ. The proper English equivalent is STOPPED
FLUTE. The English practice is to make the pipes of the stop en-
tirely of wood, but in some late examples metal pipes have been
used in the higher treble octaves. The so-called STOPPED DIAPASON,
1 6 FT., is invariably constructed of wood throughout : the stop is only
a medium-scaled BOURDON, voiced to be as free from the second
upper partial as possible. The old builders in England, and notably
Bernard Smith, made their STOPPED DIAPASONS of oak, and no
better material could be used for the purpose. In the "Schedule"
of the Organ constructed by Smith for the Temple Church, London,
are mentioned two "GEDACKTS of wainescott" a superior quality
of straight-grained oak grown abroad. The German organ-builders
have made their GEDECKTS of both wood and metal; while the
French builders have preferred metal for their manual unison BOUR-
DONS. See BOURDON and GEDECKT.
SUABE FLUTE. The name given to a quadrangular wood
stop, of 4 ft. pitch, invented by William Hill, organ-builder, of Lon-
don. The stop is of medium scale; the pipes of which are open, have
sunk blocks and inverted mouths, and yield a soft and clear unimita-
tive flute-tone which lies between the tones of the CLARABELLA and
the WALDFLOTE, but is softer than either. It is difficult to under-
stand the signification of the name SUABE, for it seems to have no
appropriate derivation ; it certainly cannot be derived from the law
term suable. The term SUABILE certainly obtains in Seidel-Kothe's
list of stops, applied to a flute-toned stop of 8 ft. pitch.*
* " Suabileist em Flotenwerk zu 8', von angenenmer Intonation, welches sick zum sanften und
langsamen Vortrage besonders eignet. Diese Stimme wird auch englische Flote genannt."
248
ORGAN-STOPS
SUAVE FLUTE. A name, suggested by the Latin suavis
sweet, pleasant; and which we have considered appropriate for a
flute-toned stop of peculiar formation, and of singularly smooth and
agreeable intonation. The pipes of the stop are of wood, open, and
of medium scale; their peculiarity obtaining in the formation of the
mouth, the upper lip of which has a cylindrical piece of polished hard
wood attached to it, in the manner shown in Fig. 33. This form of
FIG. 33
mouth clearly demonstrates the value of the thick and rounded
upper lip in the production of smooth and full flute-tone. Organ-
builders of to-day have resorted to the easy, objectionable, and per-
ishable expedient of covering the upper lip with leather in both
metal and wood pipes. It is a cheap expedient to save labor and
care in voicing; otherwise, it would never have been so readily
adopted by organ-builders.
SUAVIAL. According to Locher, a name given to a soft-toned
stop of the GEIGENPRINCIPAL class, found in old Organs; usually of
8 ft. pitch, and of short compass, beginning at C. An example of
which exists in the Organ in the French Church, at Berne.
SUB-BASS. Ger., SUBBASS. Fr., SOUBASSE, SOUS-BASSE.
According to Wolfram (1815), the old German organ-builders used
the name to designate Pedal Organ GEDECKTS, of 32 ft. and 16 ft.
pitch, constructed of wood; and the practice has been followed by
THEIR ARTISTIC REGISTRATION 249
the French builders, as is shown by the Pedals of the Organs built by
Cavaille-Coll for the Cathedral of Orleans and the Church of Saint-
Ouen, Rouen, in which we find a SOUS-BASSE, 32 FT., and a SOUS-
BASSE, 16 FT., inserted. It also appears as a manual stop, of 16 ft.
pitch, in the Bombarde of the Organ in the Cathedral of Notre-
Dame, Paris; and in the Positif of the Organ in the Cathedral of
Albi. The SUBBASS, 32 FT., exists in the Pedal of the Organ in the
Church of St. Michael, Hamburg; and, of 16 ft. pitch, in the Pedals
of the Organs in the Cathedrals of Riga and Ulm. In English instru-
ments, we find the term SUB-BASS applied to both open and covered
stops of 32 ft. pitch. In the Pedal of the Organ in the Temple
Church, London, the SUB-BASS, 32 FT., is a covered stop of wood;
while the SUB-BASS, 32 FT., in the same department of the Organ in
the Town Hall of Leeds is an open stop of metal. It would seem very
desirable, in the stop nomenclature of to-day, to confine the term
SUB-BASS to the Pedal Organ covered wood stop of 32 ft. pitch. See
CoNTRA-BouRDON and SUB-BOURDON.
SUB-BOURDON. A covered wood stop, of large scale, and of
32 ft. pitch, commonly introduced in the Pedal Organ when there is
no accommodation for an open stop of 32 ft. A SUB-BOURDON, 32
FT., was introduced in the Pedal of the Organ constructed by Willis"
for the Alexandra Palace, Muswell Hill, near London. A SUB-
BOURDON, 32 FT., is inserted in the Great of the Organ in the Parish
Church of Doncaster; and another, of wood and metal, of tenor C
compass, is inserted in the Great of the Organ in the Parish Church
of Leeds, Yorkshire. For other names used to designate stops of
this class, see CONTRA-BOURDON.
SUBPRINZIPAL, GROSSPRINZIPALBASS, Ger. Fr., PRIN-
CIPAL BASSE. The names employed to designate the open labial
stop, of wood or metal, and 32 ft. pitch, belonging to the Pedal
Organ. Its usual name in English Organs is DOUBLE OPEN DIAPA-
SON, 32 FT. Under that name, metal and wood stops exist in the
Pedal of the Willis Organ in St. George's Hall, Liverpool. A PRINCI-
PAL BASSE, 32 FT. exists in the Pdale of the Concert Organ in the
Salle des Fetes, Palais du Trocadero, Paris. A SUB-PRINCIPAL, 32
FT. , is inserted in the Pedal of the Organ in the Cathedral of Haarlem.
The pipes of this fine stop are of pure Cornish tin, burnished, and
displayed in the case. The CCCC pipe, which stands in the left
tower, is nearly forty feet long and is fifteen inches in diameter.
Under the name PRINCIPALBASS, 32 FT., the stop exists in the Pedals
of the Organs in the Cathedrals of Riga, Vienna, and Ulm. Under
250 ORGAN-STOPS
the simple term FLUTE, 32 FT., this important stop exists in the
Organ in the Basilique du Sacre-Cceur, Montmartre, Paris. For
further particulars see DOUBLE DIAPASON.
SUBTILREGAL, Ger. The name given to an old lingual stop,
of 8 ft. pitch, the tone of which was of a soft and agreeable quality,
probably differing slightly from that of the GEDAMPFTREGAL (q. u.).
Both stop and name are obsolete. See REGAL.
SUPER-OCTAVE. The open, metal, labial stop, sounding two
octaves above the foundation unison tones of the manual and pedal
departments of the Organ, being of 2 ft. pitch in the former and of
4 ft. pitch in the latter. The term, unqualified, signifies that the stop
is harmonic-corroborating and belongs to the Foundation-work, and
yields pure organ-tone. The term SUPEROCTAV is commonly used
by German organ-builders. In both the Riga and Ulm Organs,
built by Walcker, the term is incorrectly applied to stops of i ft.
pitch. The French builders commonly use the name DOUBLETTE,
2 FT. See FIFTEENTH.
TAPADILLO, TAPADO, Span. The names given by Spanish
organ-builders to covered labial stops, in all essentials similar to the
English so-called STOPPED DIAPASON and the German GEDECKT.
Under the name TAPADILLO, the stop exists in the Organs in the
Cathedrals of Burgos and Valladolid; and under the name TAPADO,
in the Organ in the Cathedral of Orense. A stop named FLAUTADO
TAPADO exists in the Organ on the south side of the Coro in Burgos
Cathedral, built in 1706.
TENOROON. The name given by certain English organ-
builders during the middle years of the last century, when incom-
plete stops were more in favor than they happily are now, to a
covered stop, of 16 ft. pitch, which was carried downward only to
tenor C on the manual clavier. The term is now obsolete, and it is
to be hoped the incomplete stop it designated is equally so.
The term Tenoroon was originally employed to designate the
Tenor Bassoon or Alto Fagotto in F; and it is, accordingly, evident
that the name given to the organ-stop had no reference to the or-
chestral Tenoroon. Should it, however, be considered desirable to
introduce in the Organ a FAGOTTO, of 4 ft. pitch, the name TENOR-
OON could be given, appropriately, to it. Such a stop would
THEIR ARTISTIC REGISTRATION 251
complete the FAGOTTO family, and be of great value in artistic
registration. At present the Organ is sadly deficient in soft octave
reed-tone.
TENTH. ltd., DECIMA. Fr., GROSSE TIERCE. Ger., TERZ.
A third-sounding mutation stop, introducing and corroborating the
fourth upper partial tone in the 16 ft. harmonic series, introduced on
the manuals, of 3)^ ft. pitch ; and corroborating the fourth upper
partial tone in the 32 ft. harmonic series, introduced in the Pedal
Organ, of 6% ft. pitch. The unscientific English organ-builders
have seriously failed to realize the value of this stop, for it appears
in only one of their Organs known to us. On the other hand, both
the French and German builders have thoroughly realized its value
in both the manual and pedal departments. The GROSSE TIERCE,
3> FT., exists in the Solo of the Organ in the Church of Saint-Sul-
pice, and in the Bombarde of the Organ in the Cathedral of Notre-
Dame, Paris. The GROSSE TIERCE, 6% FT., is introduced in the
Pedal of the Notre-Dame Organ. The TERZ, 3^ FT., exists on the
First Manuals of the Organs in the Cathedrals of Riga, Vienna, and
Ulm. The TERZ, 6% FT. , exists in the Pedal of the Riga Organ ; and,
under the term TERZBASS, 6% FT., it exists in the Pedals of the Riga
and Vienna Organs. We agree with Topfer that BASSTERZ would
be the better name. The extended term TERZENBASS, 6% FT,, has
also been used by German organ-builders. Under the names GREAT
TIERCE, 6% FT., and TIERCE, 3% FT., both stops exist in the Pedal
of the Organ in the Parish Church of Doncaster, Yorkshire, built
by Edmund Schulze, of Paulinzelle, in 1 862. The term TIERCE, 3>
FT., is, in our opinion, more expressive and desirable than the term
TENTH.
The manual stop is formed of open metal pipes; and, as it belongs
to the 1 6 ft. series, its scale should be derived from that of the
DOUBLE DIAPASON, 16 FT., which is, properly, smaller than that of
the principal DIAPASON, 8 FT. All third-sounding stops, wherever
they are introduced, should be softer in tone than the unison, octave,
and fifth-sounding stops with which they are associated. For ar-
tistic registration, this regulation of tone will be found of great im-
portance. The Pedal Organ TENTH, 6% FT., should be formed of
open pipes of wood or metal, preferably yielding a bright tone, im-
parting a desirable life to this grave department of the Organ. See
SEVENTEENTH.
TERPODION. The term derived from the Greek words
to delight, and O>&TQ a song. The musical instrument
252 ORGAN-STOPS
which gave this name to an organ-stop was invented by David
Buschmann, of Berlin, in 1816. It was a clavier instrument, vari-
ously described as frictional and percussive, in form resembling a
Pianoforte: its sounds were produced from sonorous wood in a
manner not clearly described by writers on old instruments.
The stop to which the name was originally given was invented by
J. Priedrich Schulze, of Paulinzelle, and inserted, for the first time,
in the Organ he constructed, in 1838, for the Cathedral of Halber-
stadt. Other examples of the TERPODION, 8 FT., exist in the Ober-
werk of the Schulze Organ in the Cathedral of Bremen; on the Third
Manual of the Organ in the Marienkirche, Lubeck; and in the Swell
of the Organ in the Parish Church of Doncaster, Yorkshire, built
by Edmund Schulze, of Paulinzelle. As made by this master,
the stop is formed of open cylindrical metal pipes, which have wide
and low mouths, yielding a tone of a pronounced raedy quality.
The stop, even at its best, seems to be of little tonal value. Accord-
ing to Schlimbach-Becker (1843), the TERPODION in the Halberstadt
Organ, when combined with the LIEBLICHGEDECKTS, of 16 ft. and
8 ft. pitch, and the HARMONIKA, 8 FT., produced atone resembling
that of the instrument invented by Buschmann. What must the
tone of this wooden instrument have been?
TERPOMELE. The term, derived from the Greek words
Tepicsiv to delight, and ^sXo<; melody or song; and first used to
designate a free-reed stop, of 8 ft. pitch, which was inserted, about
the year 1828, in the Organ in the Cathedral of Beauvais. The
reeds were furnished with slender resonators, introducing the proper
construction of free-reed stops. Availing himself of the fact that
free-reeds can produce varied strengths of tone under different pres-
sures of wind without alteration of pitch, the organ-builder arranged
to impart powers of expression to the TERPOMELE by means of a con-
trivance placed under the control of the organist.
TERTIAN, TERZIAN, Ger. A compound harmonic-corrob-
orating stop, formed of two ranks of open metal pipes, properly of
medium scales, which stand at the interval of a major third apart.
The position is, accordingly, an inversion of that of the SESQUIAL-
TERA, the third-sounding being larger, and lower in pitch, than the
fifth-sounding one. The TERTIAN is composed of ranks of i % ft.
and i^i ft. pitch when it belongs to the 8 ft. harmonic series; of
ranks of 3> ft. and 2%j ft. when it belongs to the 16 ft. harmonic
series; and of ranks of 6% ft. and 5}^ ft. when it belongs to the 32
feet harmonic series. Seidel tells us that in certain old Organs it
THEIR ARTISTIC REGISTRATION 253
comprised three ranks; namely, of 4. ft, 3^ ft., and 2% ft. pitch.
When the two relative ranks are introduced as separate stops in
any division of the Organ, it is not necessary to contemplate the
addition of a TERTIAN, unless on special grounds directly relating to
the very important matter of artistic and varied registration. In
this case, the stop should be of small scale and carefully voiced,
giving slight prominence in tone to the third-sounding rank. The
stop may also, and very properly, be made timbre-creating, its
ranks having contrasting tonalities. Only those who have made a
special study of compound-tone production can realize the value of
such a TERTIAN in imparting color in refined and artistic registra-
tion.
THIRTY-FIRST. Ital., TRIGESIMA PRIMA. A third-sounding
MIXTURE rank, of open metal pipes, corroborating the fourth upper
partial tone of the super-octave (2 ft. pitch) stops; its extremely
acute pitch preventing its being considered, in anything save a
purely philosophical sense, as belonging to a lower harmonic series.
Owing to the smallness of the pipes forming this rank, it can only be
introduced in the two lower octaves of a manual MIXTURE. While
it appears to have been used, in some way, by the Italian organ-
builders, as in one of the Organs in the Cathedral of Milan, we have
been unable to find an instance of the insertion of this extremely
acute third-sounding rank in a MIXTURE made by any known organ-
builder. It may, however, exist in the ninety-five ranks of MIXTURE
which are inserted in the Organ in the Monastic Church, Weingarten.
THIRTY-SIXTH. Ital. , TRIGESIMA SESTA. An octave-sound^
ing MIXTURE rank, of open metal pipes, corroborating the
seventh upper partial tone of the super-octave (2 ft. pitch) stops;
its extremely acute pitch preventing its being considered, in any-
thing save a purely philosophical sense, as belonging to a lower
harmonic series. Owing to the very small size of the pipes forming
this rank, it can only be introduced in the bass octave of a manual
MIXTURE. It appears, as in the case of the THIRTY-FIRST (j. v.)>
in one of the Organs in the Cathedral of Milan; but we have not been
able to find the THIRTY-SIXTH introduced in any MIXTURE of Eng-
lish, French, or German origin. The only instance of its introduc-
tion known to us, is afforded by the seven-rank CYMBALE in the
Great of the Organ in the Music Hall, Cincinnati; in this compound
stop, the THIRTY-SIXTH is confined to the bass octave.
THIRTY-THIRD. Ital., TRIGESIMA TERZA. A fifth-sound-
254 ORGAN-STOPS
ing MIXTURE rank, of open metal pipes, corroborating the fifth
upper partial tone of the super-octave (2 ft. pitch) stops; its ex-
tremely acute pitch preventing its being considered, in anything
save a philosophical sense, as belonging to a lower harmonic series.
The THIRTY-THIRD was introduced by the old English builders in
their compound harmonic-corroborating stops. Harris inserted it,
in conjunction with the TWENTY-NINTH, in the bass octave of the
two-rank MIXTURE in the Organ erected by him, during the closing
years of the seventeenth century, in the Church of St. Peter Man-
croft, Norwich. It appears, in conjunction with the THIRTY-SIXTH,
in the North Organ in the Cathedral of Milan. The THIRTY-THIRD
is inserted in the bass and tenor octaves of the CYMBALE in the Organ
in the Music Hall, Cincinnati.
TIBIA ANGUSTA. (Lat. Tibia a, pipe; A ngustus narrow).
The term that has been employed by German organ-builders to
designate a flute-toned stop, of small scale, and 8 ft. pitch. It
resembles in formation and tone the DOLZFLOTE (q. v.).
TIBIA BIFARA. (Lat. Tibia- -a pipe; Bifarius, double).
The stop, commonly named BIFARA (without the prefix), is formed
in two ways ; both with the view of imparting a tremulous or wavy
effect to its voice. For full particulars of this stop, see BIFARA.
TIBIA CLAUSA. (Lat. Tibia & pipe; Clausus closed).
The name given to a large-scaled covered stop, of 8 ft. pitch, the
pipes of which are of wood. The mouths of the pipes are cut up
about half their widths; their upper lips being thick, and very
smoothly rounded and polished with black-lead. To save trouble
and labor, the rounding and polishing are omitted, and the lips are
covered with leather a perishable expedient. When properly
voiced, on a copious supply of wind of moderate pressure, the stop
yields a beautiful and singularly pure quality of tone. A full scale
for this stop, in the ratio I : V8, gives the CC pipe a width of 5.56
and a depth of 7.52 inches; the C pipe a width of 3.30, and a depth
of 4.46 inches; and the middle c 1 pipe a width of 1.96, and a depth
of 2.66 inches. The name and modern treatment of this large-scaled
GEDECKT were introduced by Hope- Jones.
TONE AND REGISTRATION. The tone of the TIBIA CLAUSA,
when artistically produced, is freer from harmonic over-tones than
the voices of the generality of covered stops, and especially those of
the BOURDONS; hence its peculiar value. In this respect it ap-
proaches the voice of the true LIEBLICHGEDECKT; but, on account
THEIR ARTISTIC REGISTRATION 255
of its large scale, the TIBIA CLAUSA yields a much greater volume of
sound, and, accordingly, lends itself more effectively to the building-
up of a valuable series of highly-colored compound tones. Its proper
place is in the expressive divisions of the Organ, in which a firm
ground-work of unison tone is required to balance and carry the
assertive string-, reed-, and brass-tones of the orchestral stops : such
as the so-called Swell in the Church Organ; and the wood-wind,
brass-wind, and string-toned divisions of the properly stop-appor-
tioned Concert-room Organ. In registration with the octave, muta-
tion, and softly- voiced compound harmonic-corroborating stops, the
TIBIA CLAUSA, 8 FT., produces a family of very beautiful tones.
TIBIA DURA. (Lat. Tibia & pipe; Item hard). The
name given to an open wood stop, of 4 ft. pitch, introduced by Hope-
Jones; the tone of which is cold and penetrating ; hence the name.
The approved form of the pipes is quadrangular and inverted pyra-
midal, and wider than deep, so as to allow of a large mouth, which
is of the inverted form. The tone of the stop has no special charac-
ter to render it valuable in artistic registration. Examples exist in
several Organs constructed under the superintendence of Hope-
Jones.
TIBIA MAJOR, (Lat., Tibia -a pipe; Major greater).
The term employed^-by German organ-builders to designate a man-
ual, covered labial stop, of the BOURDON class, and of 16 ft. pitch.
It is commonly of large scale and of a full tone. Examples exist on
the First Manuals of the Walcker Organs in the Cathedrals of Ulm
and Vienna. Another exists in the Echo of the Schulze Organ in the
Parish Church of Doncaster; in this instance the stop is of a com-
paratively soft tone. All the Echo Organs made by this builder
being characterized by extreme delicacy and beauty of tone, as all
Echo Organs should be.
TIBIA ]$INOR. (Lat., Tibia -a pipe; Minor smaller).
German organ-builders have, in some instances, given the name to a
large-soiled covered stop, of 8 ft. pitch, which is practically the
Octave^of the TIBIA MAJOR, 16 FT. (q. v.). An example of the stop,
under the simple name TIBIA, 8 FT., exists on the First Manual of
the Organ in St. Paulskirche, Frankfurt. We are not certain, how-
ever, if this stop is covered. The name TIBIA MINOR has been given
by John H. Compton, of Nottingham, England, to covered stops,
introduced by him in certain of his Organs, the pipes of which are of
large scale, and of wood and metal. The wood stop differs only
256 ORGAN-STOPS
slightly in treatment and tone from the TIBIA CLAUSA (q. *>.) The
metal TIBIA MINOR, 8 FT., as made by Compton, is a fine and valu-
able contribution to the flute-tone of the Organ of the Twentieth
Century; and deserves the attention of every progressive organ-
builder. The chief characteristics of its pipes are their very large
scale and their very narrow mouths. The CC pipe has a diameter
equal to that of a medium-scaled CC DIAPASON pipe. The mouths
are cut high, arched, and leathered.
TONE AND REGISTRATION. The tone of the Compton metal
TIBIA MINOR, 8 FT., is sui generis; differing widely from that of the
ordinary large-scaled GEDECKT; and therein lies much of its value.
The Organ of to-day calls loudly for new voices of a refined, building-
up, and timbre-creating character; not old voices coarsened by the
use of inordinate pressures of pipe-wind, which seems to be the chief
aim of the generality of voicers at the present time. The tone of the
f stop under consideration is thus clearly described by J. I. Wedgwood,
who has had ample opportunity of forming an accurate estimate of
its character: "The tone of the TIBIA MINOR is extraordinarily
effective. In the bass it is round and velvety with a suspicion of
smooth French Horn quality. In the treble the tone becomes very
clear and full. The top notes of the stop, indeed, bear in them some
resemblance to the full liquid notes of the Ocarina, though free, of
course, from the undesirable features of that instrument. Whilst
entirely devoid of the objectionable hooting quality sometimes dis-
played by powerful FLUTES, it forms a solo stop of remarkably fine
effect, and in combination serves to add much clearness and fullness
of tone to the treble, and, in general, exercises to the fullest extent
the beneficial characteristics of the TIBIA class of stop/' In regis-
tration, this stop lends itself to the creation of many beautiful com-
pound tones and artistic colorings. See TIBIA CLAUSA.
TIBIA MOLLIS. (Lat, Tibia & pipe; Mollis soft). The
name given by Hope-Jones to an open wood stop, of a soft flute-
tone; the pipes of which have their mouths cut parallel to their sides,
and the caps so placed as to direct the wind-stream transversely, in
a manner similar to that from the mouth while playing the orches-
tral Flute. An example exists in the Organ in St. George's Church,
Blackheath, London. The name is equally appropriate for any very
soft flute-toned stop.
TIBIA PLENA. (Lat, Tibia -a pipe; Plenus full). The
name given to an open wood stop of 8 ft. pitch; the pipes of which
are of greater depth than width, of very large scale, and have mouths
THEIR ARTISTIC REGISTRATION 257
cut moderately high, their upper lips being thick and carefully
rounded and polished when properly finished. Leathering the lips
has been adopted to produce sufficient smoothness. Lips carefully
rounded and burnished with black-lead are to be preferred, as
leather is certain to change and decay in a few years. The TIBIA
PLENA is the most powerful of the wood stops of unimitative flute-
tone; and should find a prominent place in every important Organ,
preferably on the First or Great Manual and in its expressive sub-
division. The stop has been made of several different scales, all of
which are necessarily large. For instance, that stated for the stop
in the Organ in the Cathedral of Worcester, gives the CC (8 ft.) pipe
a measurement of 7yf inches by 9 inches, and middle c 1 (2 ft.) pipe
2^-| inches by 3^ inches. This is an extreme and unnecessarily
large scale. Examples of the stop exist in several Organs in England
and the United States.
TONE AND REGISTRATION. The tone of the TIBIA PLENA when
artistically voiced, is singularly full and dignified; the presence of
certain low harmonics imparting to it considerable richness, and
rendering it of the greatest value in* registration with the DIAPASONS,
to the voices of which it imparts a remarkable color and grandeur.
It is also of the greatest value in combination with such lingual stops
as the DOUBLE TRUMPET and unison TRUMPET of the First or Great
Organ, to which division it properly belongs. To be of maximum
value it should be rendered expressive and flexible, so as to be largely
available throughout the other more important divisions of the
Organ. In a very large Concert-room instrument, a TIBIA PLENA
may with the greatest advantage be inserted, as a tone-builder, in
either the Solo Organ or that devoted to the brass-wind division; or,
indeed, in both.
TIBIA PROFUNDA. (Lat., Tibia -a pipe; Proj undis deep) .
The name given to the Pedal Organ stop, of 16 ft. pitch, the pipes
of which are similar in formation and quality of tone to those of the
TIBIA PLENA (q. v.). An example exists in the Pedal of the Organ
in the Church of St. Mary, Warwick, England.
TIBIA RURESTRIS, FISTULA RURESTRIS. (Lat., Fis-
tula, Tibia a pipe; Ruralis rural). Names that have been used
by the early organ-builders to designate a flute-toned stop in all
essentials similar to the BAUERFLOTE (q. v.).
TIBIA SILVESTRIS. (Lat., Tibia a, pipe; Silvestrisdi a
wood). The term that has been used to designate an open wood.
258 ORGAN-STOPS
flute-toned stop, identical with that commonly named WALDFLOTE
(q. v.). Seidel and Wolfram render the name TIBIA SYLVESTRIS.
TIBIA TRANSVERSA, TIBIA TRAVERSA. (Lat., Tibia
a pipe; Transversus, Traversus across). The names given by old
builders to the open labial stop, yielding imitative flute-tone, which
is now commonly designated FLAUTO TRA VERSO (q. v.) or ORCHES-
TRAL FLUTE.
TIBIA VULGARIS. (Lat., Tibia a pipe; Vularis common).
The name given in early times to the unimitative flute-toned stop,
now commonly termed BLOCKFLOTE (q. v.).
TIERCE, Fr. Ger., TERZ. Eng., SEVENTEENTH. A muta-
tion harmonic-corroborating stop, of 1^5 ft. pitch in the manual
divisions, where it belongs to the 8 ft. harmonic series; and 3^ ft.
pitch in the Pedal Organ, where it belongs to the 16 ft. harmonic
series. See SEVENTEENTH.
TIERCINA. The name given to a metal labial stop, of 8 ft.
pitch, the pipes of which are of small scale, covered, and voiced to
yield a tone in which the fourth upper partial tone (the Seventeenth)
is strongly in evidence; hence the name. This compound tonality is
enriched by the presence, in a subordinate degree, of the second
upper partial, the Twelfth. To enrich the harmonic structure, we
inserted a TIERCINA, 8 FT., in the Second Subdivision (String) of the
Third Organ of the instrument installed in the Festival Hall of the
Louisiana Purchase Exposition (1904).
TIERCE FLUTE. The name given by its introducer, George
W. Till, of Philadelphia, Pa. to a labial dual stop, formed of a metal
HARMONIC FLUTE, 8 FT., having the compass of CC to c s , and a
metal TIERCE, 3)^ FT., of the same compass, yielding pure organ-
tone, and belonging to the 16 ft. harmonic series. In this instance,
however, the TIERCE is a timbre-creating rather than a harmonic-
corroborating stop or rank. In combination with the HARMONIC
FLUTE it produces a compound solo voice of an absolutely new and
remarkable tone-color, impossible to be produced from a single-
ranked FLUTE, however made and voiced.
This dual stop is inserted in the Choir of the Organ in the Wana-
maker Store, in Philadelphia, where it speaks on wind of 15 inches.
The stop is unique; and it is safe to say that such a dual stop has
never been even contemplated in the preparation of the tonal scheme
for any other Organ hitherto constructed. We know of no Organ
THEIR ARTISTIC REGISTRATION 259
in which the component parts of the stop exist in any single manual
division. The tone of the TIERCE FLUTE must be heard ; it cannot be
described. It is especially a solo stop.*
TRAVERSPLGTE, Ger. Ital., FLAUTO TRAVERSO. The lab-
ial stop, properly formed of cylindrical or quadrangular wood pipes
voiced to yield a tone as closely imitating that of the orchestral Flute
as possible. Fine examples, of 8 ft. and 4 ft. pitch, voiced by Ed-
mund Schulze, exist in the Choir and Echo of the Organ in the Parish
Church of Doncaster, England. See FLAUTO TRAVERSO and OR-
CHESTRAL FLUTE.
TRAVERSENBASS, Ger. The name given to a labial metal
stop, of 1 6 ft. pitch, the higher notes of which produce tones similar
to those of the TRAVERSFLOTE, 8 FT. According to Wolfram (1815),
an example exists in the Pedal of the Organ in the Church of St.
Bonifacii, Langensalza. A clear flute-toned stop, of this class, and
of medium strength of tone, would be extremely valuable in the
Pedal of a Concert-room Organ.
TRICHTERREGAL, Ger. Literally, Funnel Regal An old
lingual stop of the REGAL class furnished with funnel-shaped re-
sonators; hence its name. A TRICHTERREGAL, 8 FT., was inserted
by Schnittker in the Organ he built for the Church of St. Jacobi,
Hamburg. See REGAL.
TRINONA. The name given by the builder of the Organ in the
Church of St. Vincent, Breslau, to stops, of 8 ft. and 4 ft. pitch, in-
serted on the Second or Upper Manual. These stops are said to be
of wood and of a soft viol quality of tone. The name would seem to
imply the presence, in some prominence, of the fourth upper partial
tone the Seventeenth in the voices of the stops, quite compatible
with good viol-tone.
TRIPLETTE. The term employed to designate a compound
harmonic-corroborating stop formed of three unbroken ranks of
open metal pipes, similar in treatment to the DOUBLETTE of the
German organ-builders. The TRIPLETTE furnishes a favorable op-
portunity for the formation of a valuable timbre-creating stop. Its
ranks may be of metal, or of wood and metal, open or covered, pipes,
voiced to yield any desirable tones. Its ranks may be of 2% ft., 2 ft.,
and i% ft.; 2 ft, i% ft, and i^ ft; or i% ft, ij^ ft, and i ft
* Full particulars of the formation and scales of this stop are given in "The Organ of the
Twentieth Century." Page 104.
26o ORGAN-STOPS
pitch; all the ranks representing and corroborating upper partial
tones of the 8 ft. harmonic series. Unless designed as a strictly
timbre-creating stop, the TRIPLETTE has nothing to specially re-
commend its adoption : its ranks would be more useful as separate
stops.
A finely-voiced timbre-creating TRIPLETTE would unquestion-
ably be of great value in artistic registration; producing compound
tones of great variety and beauty, unknown in the Organs hitherto
constructed. It should be so placed as to be available on any manual
clavier.
TROMBA, Ital. Italian organ-builders have employed this
term to designate lingual stops of the TRUMPET class, of both 8 ft.
and 16 ft. pitch, as in the Organ in the Church of St. Allessandro,
Milan. We find the term applied to a stop, of 32 ft. pitch, in the
Great of the Organ in the Church of Sta. Maria di Carignano,
Genoa; reconstructed by C. G. Bianchi, in 1863. Spanish and Por-
tuguese organ-builders have applied the name, commonly with the
addition of certain qualifying terms, to their TRUMPETS of powerful
and special intonation. In the Organ on the north side of the Coro
in Burgos Cathedral there are two lingual stops, labeled, respec-
tively, TROMBA REAL (Royal Trumpet) and TROMBA BATALHA
(Battle Trumpet). We find stops in the Organ in the Church of the
Martyros, Lisbon, labeled TROMBA REAL, TROMBA BATALHA, and
TROMBA MAGNA (Great Trumpet).
The TROMBAS, of 8 ft. and 16 ft. pitch, as made by German and
English organ-builders, differ widely in tonality from those of the
old Spanish builders : they are TRUMPETS yielding a full and smooth,
unimitative brass-tone, more powerful than that of the ORCHESTRAL
TRUMPET, but less assertive than the tones of the TUBAS. A
TROMBA, 8 FT., made by Walcker, exists on the First Manual of the
Organ in the Accademia di Sta. Cecilia, Rome. An example, by
Norman & Beard, exists in the Organ in the Church of All Saints,
Netting Hill, London. The late Edmund Schulze of Paulinzelle
invented a labial TROMBA, 16 FT., the pipes of which were of wood,
square and inverted pyramidal in form, and of medium scale, and
voiced with the harmonic-bridge.
TONE AND REGISTRATION. Given that the tone of the TROMBA,
8 FT., is distinct in strength and quality from the tones of the HAR-
MONIC and ORCHESTRAL TRUMPETS, on the one hand; and from the
tones of the TROMBONE, OPHICLEIDE, and TUBA, on the other hand,
there can be no question as to the great value of the stop in tne ap-
THEIR ARTISTIC REGISTRATION 261
pointment of a large Church Organ or a Concert-room instrument.
Like all lingual stops, its place is in an expressive manual division.
In a properly stop-apportioned Concert-room Organ, its proper
place is in the division devoted to the representatives of the ' ' brass-
wind" forces of the orchestra.* Its presence there will be of the
greatest service in artistic registration, furnishing a rich and full
lingual tone, of a neutral and good mixing quality, combining per-
fectly with any pure organ-tones, flute-tones, and viol-tones.
TROMBA CLARION. The appropriate name for a TROMBA,
of 4 ft. pitch, voiced considerably softer than the unison TROMBA
(q. v.). Softly voiced octave lingual stops have been too much neg-
lected in the stop-appointments of modern Organs; yet their value
in artistic registration cannot well be overrated. They effectively
enrich, brighten, and impart life to volumes of unison and double
tones; which, without them, would be dull, heavy, and unmusical.
TROMBONE, Ital., Eng., Pr A lingual stop of 8 ft. pitch in
the manual divisions and of 16 ft. pitch in the Pedal Organ. The
stop belongs to the TRUMPET family; and its pipes are similar in
form to, but of larger scale than, those of the ordinary TRUMPET,
8 FT., only having tongues and reeds yielding a distinctive and more
powerful intonation. The stop is properly voiced to closely imitate
the tones of the orchestral Trombones when played forte, but not
fortissimo. The more powerful intonation being desirable in the
POSAUNE (q. v.).
FORMATION. The manual TROMBONE, 8 FT., is constructed of metal through-
out; its resonators being of the inverted conical form, of large scale, and either
of good pipe-metal or zinc; the latter being commonly used at the present time.
The Hoyt Two-ply Pipe-metal is strongly to be recommend for high-class work.
The Pedal Organ TROMBONE, 16 FT., is made either entirely of metal or of wood
and metal. When of metal, it differs in no essential, save in dimensions, from the
TROMBONE, 8 FT. When chiefly of wood, the pipes are necessarily different in
form and construction. The resonators are quadrangular and of inverted pyra-
midal shape. The boots, when of wood, are square and of ample dimensions to
receive the feet of the resonators, the reed-blocks, and all fittings. The reeds or
echalotes are of the open class; and the tongues or languettes are of hard-rolled
brass, of medium thickness, and finely curved so as to produce a bright quality
of tone.
TONE AND REGISTRATION. The proper tone of the manual
TROMBONE, 8 FT., should resemble as closely as possible the tones of
the orchestral Trombones when played without the production of
* See stop-apportionment of the Fourth Organ, in our scheme for the Concert-room Organ.
i,' ven on page 31 1 of "The Organ of the Twentieth Century."
262 ORGAN-STOPS
their extreme brazen clang. Fine examples of this stop, voiced by
the masterly hand of the late George Willis, exist in the Great and
Solo of the Organ in St. George's Hall, Liverpool. The value of this
unison stop seems to be overlooked by the organ-builders and organ-
designers of to-day, but its introduction in important instruments is
strongly to be recommended. So firmly convinced are we of the
value of its distinctive tone, that in our suggestive tonal scheme for
the Concert-room Organ of the Twentieth Century, we have in-
serted in the First Organ (Great) a TROMBONE, 8 FT., and in the
Fourth Organ (Brass-wind) the complete TROMBONE family; namely,
a CONTRA-TROMBONE, 16 FT., TROMBONE, 8 FT., TROMBONE QUINT,
5J" FT., and TROMBONE OCTAVE, 4 FT.* It would be difficult to over-
rate the importance of such a combination of lingual stops at the
disposal of the organist; yet it has never appeared in any executed
Organ. The neglect of the manual TROMBONE, 8 FT., is as remark-
able as it is unwise, yet it is the only one which can properly repre-
sent the orchestral Trombone. In a large collection of 'stop-lists of
Organs, designed in this country, which lie before us, we fail to find a
single TROMBONE, 8 FT., mentioned. Indeed, in the list of an im-
mense divided Organ, embracing 283 speaking stops, there is only
one TROMBONE given, and that, of 16 ft. pitch, in one of the Pedal
Organs. In the Fourth or Solo Organ of the large instrument we
designed, and which was installed in the Festival Hall of the Louis-
iana Purchase Exposition, St. Louis, 1904, there was a TROMBONE,
8 FT., and a BASS TROMBONE, 16 FT. ; both of metal.
TROMPETTE EN CHAMADE, Fr. The term employed by
French organ-builders to designate a TRUMPET, the pipes of which
are projected horizontally and fan wise from the front of the case, as
in the Organ in the Church of Saint-Ouen, Rouen (See Frontispiece).
This treatment is common in important and noisy Spanish Organs.
On both artistic and common-sense grounds the treatment is to be
condemned. No lingual stop should be without expression.
TROMPETTE HARMONIQUE, Fr. This important lingual
stop was invented by Cavaill6-Coll, and introduced, for the first
time, in the Organ of the Royal Church of Saint-Denis, constructed
in 1 841 . In this remarkable instrument there were inserted no fewer
than five examples of the stop of 8 ft. pitch; one in the Positif, two
in the Grand-Orgue; and two in the R6cit-ficho Expressif. In his
" Rapport" on the Organ, J. Adrien de la Page remarks:
* See " The Organ of the Twentieth Century, " page 31 1.
THEIR ARTISTIC REGISTRATION 263
"La trompette harmonique du clavier de rScit est, par sa puissance et par Tex-
cellence des sons qu'elle produit, incomparablement suprieure a tout ce qtie Ton
connait en ce genre. Le caractere tout-a-fait particulier de ses basses, lorsqu'on
1'emploie comme partie chantante accompagnee des jeux de fond, est d'un effet
admirable. On ne peut dire autant des series de jeux de flUtes harmoniques qui
donnent a I'ensemble de 1'orgue tant de rondeur et de puissance."
For further particulars respecting the TROMPETTE HARMONIQUE,
see HARMONIC TRUMPET.
TRUMPET. Fr., TROMPETTE. Ger., TROMPETE. Ital., TROM-
BA. Dtch., TROMPET. Span., TROMPETA. The manual lingual
stop, of 8 ft, pitch, to which these names are given by the organ-
builders of different countries, may reasonably, and under usual
conditions, be considered the most generally useful, if not the most
important lingual stop in the Organ. So much so is this the case,
that no Organ of any pretension is constructed without one, com-
monly occupying a place in the Great or foundation division of the
instrument. Unless when it is strictly of the solo or orchestral tonal-
ity, the TRUMPET is a chorus-reed of the first importance. Realizing
this fact, the German organ-builders place the stop on different
manuals. In the Organ in the Cathedral of Ulm the TROMPETE,
8 FT., is inserted on the Second and Third Manuals only, and also in
the Pedal Organ. It commonly occupies a place in the Pedal Organ
of Walcker's important instruments. In the Cavaill-Coll Organ
in the Church of Saint-Sulpice, Paris, there are two TROMPETTES
in the Grand-Choeur, one in the Positif , one in the R6cit, one in the
Solo, and one in the Pdale six TROMPETTES, of 8 ft. pitch (none
harmonic), in an Organ of 100 speaking stops. The builder of this
Organ fully realized the value of the chorus TRUMPET.* In English
* " La TROMPETTE est le registre d'anches le plus pur, le plus rond, le plus fin d'harmonie et
de sonorit6, quand il est bien fait; le plus desagreable, quand il est manque*. Chose singuliereil
reussit moins souvent en Allemagne qu'en France; j'ai dj& cherch6 la raison de cette difference
dans la difference d'allures des deux nations: le bruit guerrier, I'eclat, tout ce qui tend a accen-
tuerla musique, et mieux marquer son rhythme, semble etre de notre domaine. De leurcdte,
les f acteurs allemands se vengent de la maigreur de leurs Trompettes par la variete, la finesse
melancolique de leurs jeux de fonds.
' ' Une bonne Trompette doit avoir le tuyau un peu plus long que de rigueur, de grosse tailleet
de solide epaisseur, en etain fin et bien battu, pos6 sur un-pied in6branlable et bien proportionne"
en hauteur et en embouchure. La languette, de laiton f ortement 6croui, mediocrementrecour-
bee al'entr6e de 1'anche, s'y posera bien 6galement sans devier, et sera comme cette anchede
large'ur et d'epaisseur proportionnees a la vigueur du courant d'air centre lequel elle combat.
Le tuyau trop long ferait octavier; trop court, il pourrait donner un son plus tranchant , mais
criard. Trop d'epaisseur serait inutile, et la minceur ferait grincer le tuyau. La languettetrop
lachefait raler le son, trop ferine ou d6viee de son aplomb, courbde trop haut ou trop bas, elle
devient capricieuse, raide jusqu'au mutisme. Les justes proportions du tuyau de Trompette
combineesaveccelles du vent doivent donner un son egal, brillant sans trop d'eclat, mais male et
doux ct la f ois. C'est cette derniere qualite, si rare que les bons accordeurs distinguent au bour-
don quiressort toujours d'une Trompette arriv6e IL sa juste harmonic. Toutes ces perfections
264 ORGAN-STOPS
Organs of the usual Church type a single TRUMPET, 8 FT., is com-
monly introduced, inserted almost invariably in the Great. Even
in the important Organ, of 70 speaking stops, in the Cathedral of
York, there are only two TRUMPETS, one in the Great and the other
in the Swell. In the Concert Organ in St. George's Hall, Liver-
pool, there is a TRUMPET, 8 FT., in the following manual divisions,
Great, Choir, Swell, and Solo; and also one in the Pedal Organ.
Judging from a survey of the tonal-appointments of American
Organs, it would seem that organ-builders here have failed, as a
rule, to realize the great value of the TRUMPET. An example of
this neglect may be given: in an important College Organ, of 46
speaking stops (exclusive of 13 borrowed stops), built by a dis-
tinguished firm, there is not a single TRUMPET, or any lingual stop
that can take its place in registration. Surely this is a step in the
wrong direction. In another Organ, of 52 speaking stops, there is
not a single TRUMPET or any stop of its tonal character. Further
comment is unnecessary.
FORMATION. The pipes of the TRUMPET are invariably made of metal; their
resonators being of the inverted conical shape and of medium scale, formed of a
good alloy of tin and lead (spotted-metal) or zinc. The Hoyt Two-ply Pipe-metal
is to be recommended. The resonators are of about the same lengths as open
labial metal pipes of the corresponding pitch, and are entirely open at top : they
should be accurately cut to length and not slotted. The form of the complete
TRUMPET pipe is shown in the accompanying illustration, Fig. 34. The scale of
the resonators varies according to character and strength of the tone desired. The
reeds or echalotes are properly of the open class; but in the case of the ECHO TRUM-
PET, or a stop suitable for a true Chamber Organ, the reeds may properly be of
the closed variety. The tongues or languettes must in all cases be such as to im-
part a true Trumpet tonality: this should be pronounced in the case of the ORCHES-
TRAL TRUMPET.
TONE AND REGISTRATION. The tone of the ordinary chorus
TRUMPET, while of a true Trumpet character, should not too closely
resemble that of the orchestral instrument played forte. A certain
amount of brazen clang is desirable to give individuality to the stop;
but it should not be such as to prevent its free use in general regis-
tration with stops of organ-tone. When of this tonality, the stop is
doivent se trouver dans chaque tuyau s6par6ment, et dans tous compares 1'un & 1'autre. II faut
y joindre enfin celle d'une grande promptitude et nettetS de langage.
"La taille des Trompettes doit 6tre en rapport avec leur place; ainsi, la pSdale de Trompelte
sera de plus forte taille que celle qui chante k la main; celle du recit, plus delicate ue celle du
grand orgue. Quoiqu'on joue avantageusement la Trompette seule, 1'alliance de ces octaves
extremes avec celles du Cl&iron lui donne une grande vigueur, surtout & la p^dale; 1'habitude de
certains facteurs de loger sur un seul clavier deux Trompettes de moyenne taille ne vaut Hen,
parce qu'elles sont rarement d'accord, et qu'il vaut mieux un seul instrument de forte taille,
dont les vibrations aient 1'avantage de la clarU et de I'unitS." Regnier.
THEIR ARTISTIC REGISTRATION
265
most valuable in the First or Great Organ, where it can enter into
combination with the DIAPASONS and the entire foundation-work,
producing compound tones of great richness and dignity; this is due
to the great number of upper partial tones its voice comprises. When
enclosed and rendered flexible and expressive, as it
should be in every artistically appointed Great Organ,
the value of the TRUMPET is increased ten-fold. The
solo stop, properly designated ORCHESTRAL TRUMPET,
8 FT., should be voiced to imitate as closely as pos-
sible the characteristic tones of the orchestral instru-
ment. Its voice will, accordingly be singularly bright,
silvery, and jubilant; differing, in these desirable qual-
ities, from every other lingual stop of the Organ. It will
be used in registration for the production of rich orch-
estral effects, in which its brilliant voice will prove invalu-
able. It properly belongs to the brass-wind division of
the Concert-room Organ; where it will be found to be of
the utmost service both in solo and combinational effects.*
TUBA, TUBA MIRABILIS. Known by both names,
this stop, of the TRUMPET class, is the most assertive
lingual stop in the modern Organ, having a voice of
great sonority and grandeur. The stop is of 8 ft. pitch,
and has been placed in different divisions of important
Organs; but its most desirable place is in the manual
division containing the principal solo stops. In this posi-
tion it exists in the Concert Organs in the Centennial
Hall, Sydney, N. S. W., the Auditorium, Chicago, and
the Music Hall, Cincinnati. It also exists in the Solos
of the Organs in Westminster Abbey, the Cathedral of
York, and the Cathedral of the Incarnation, Garden
City, L. I. It exists on the First Manual of the Organ
in the Cathedral of Riga, and in the Dome Organ in St.
Paul's Cathedral, London.
FIG. 34
FORMATION, The pipes of the TUBA are, like those of the
TRUMPET (Fig. 34) of inverted conical form, of large scale, and of
thick and firm metal. Its reeds or &chalotes are properly of the open
class, and its tongues or languettes are wide, thick, and boldly
curved. The TUBAS of different organ-builders have been
voiced on, winds ranging from eight to twenty-five inches: the higher octaves of
the stop in the Dome Organ in St. Paul's Cathedral speak on the latter pressure.
The scale of the resonators varies in different examples ; but the maximum may be
* See "The Organ of the Twentieth Century, " page 311.
266 ORGAN-STOPS
accepted as seven inches diameter at top for the resonator of the CC (8 ft.) pipe.
A remarkable labial stop, designated TUBA MIRABILIS, has recently been in-
vented by the distinguished artist in stop formation, W. E. Haskell, of Brattle-
boro, Vt. This is an open wood stop of very peculiar form and general treatment,
in which the harmonic-bridge holds a prominent position. The form and construc-
tion of the pipes of this unique stop are fully shown on Plate IV. All the details
given deserve the careful study of those interested in pipe formation. The stop
speaks on a very copious supply of wind of fifteen inches pressure ; yielding a tone
of great volume and dignity, and of a most valuable quality. It may be said to
open an entirely new chapter in the history of wood pipe development, and points
the way to further achievements.
TONE AND REGISTRATION. The Tuba of the modern orchestra,
favored by Wagner on account of its majestic and impressive bass
tones, belongs to the Saxhorn family; and its voice should, properly,
be imitated by that of the TUBA of the Organ ; but such a desirable
imitation seems to have been completely neglected in the desire to
produce the greatest volume and dominance of tone possible from
lingual pipes and high-pressure voicing. This aim has certainly
been reached in the HARMONIC TUBA (q. v.). A truly imitative
TUBA, 8 FT., similar in power to the pure and unstrained voice of the
orchestral instrument in Bb, would be an invaluable addition to the
brass-tone forces of the Concert-room Organ. The extremely power-
ful tone of the TUBA, as it usually obtains in English and American
Organs, is of very limited use in registration ; but in full organ effects
and fortissimo climaxes it is most impressive, dominating all other
tonalities. Its presence is of most value in the Fifth or Solo Organ
of the properly stop-apportioned Concert-room Organ.
TUBA CLARION. A TUBA of octave or 4 ft. pitch, the pipes
of which are formed in all respects similar to those of the unison
TUBA, 8 FT. As this stop will accompany the unison stop, in what-
ever division of the Organ it is inserted, it should be subordinate in
scale and strength of voice.
TUBA MAGNA. The name given either to the TUBA MIRA-
BILIS or the HARMONIC TUBA when voiced to yield a tone of great
volume and assertiveness.
TUBA MAJOR. The name employed in some instances to
designate the CONTRA-TUBA, 16 FT., but more commonly applied to
such a stop as the TUBA MAGNA (q. v. ).
TUBA MINOR. The name appropriately given to a TUBA,
8 FT., of small scale, yielding a softer voice than that of the other
TUBAS. This stop may be introduced in the Swell or in the expres-
PLATE IV
THE HASKELL LABIAL TUBA MIRABILIS
THEIR ARTISTIC REGISTRATION 267
sive subdivision of the Great Organ; its subdued voice rendering it
of considerable value in full combinations, especially with the DIA-
PASONS, of 8 ft. and 16 ft. pitch. Good examples of this useful stop
exist in several English Organs.
TUBA PROFUNDA. The term employed to designate a stop
of the TUBA class and of 16 ft. pitch, properly belonging to the Pedal
Organ. When introduced in a manual Organ, the stop is commonly
and appropriately named CONTRA-TUBA (q. v.). An example exists
in the Solo of the Organ in the Centennial Hall, Sydney, N. S. W.
TUBA QUINT. A mutation stop of the TUBA class, of 5^ ft.
pitch, voiced to yield a tone considerably softer than that of the
unison TUBA (8 ft.) in the same division. As it strictly belongs to
the 1 6 ft. harmonic series, it can only be properly introduced in asso-
ciation with the CONTRA-TUBA, 1 6 FT. The complete family will
accordingly comprise the CONTRA TUBA, 16 FT.; TUBA, 8 FT.; TUBA
QUINT, 5^ FT.; and TUBA CLARION, 4 FT.; a series of powerful
lingual stops capable of producing compound tones of surpassing
grandeur. The four stops alone admit, in registration, of eight
perfectly satisfactory combinations, each of which would have a
distinctive coloring. We know of no Organ in which the complete
family is introduced.*
TUBA SONORA. The name given by Robert Hope- Jones to a
TUBA, 8 FT., voiced to yield a peculiarly full and pure tone, largely
free from the clang due, in other stops of the TRUMPET class, to the
prominence of high upper partial tones. A stop of this tonality is
much more generally useful in artistic registration than the ordinary
dominating TUBA. A fine example of the TUBA SONORA, 8 FT.,
voiced on wind of twenty inches pressure, exists in the Solo of the
Organ in the Cathedral of Worcester, England.
TUBASSON, Fr. The name tljat has been used by French and
Belgian organ-builders to designate a Pedal Organ lingual stop of
the TROMBONE class, of 16 ft. pitch, and softer in tone than the or-
dinary TROMBONE, 16 FT., as made by English and American organ-
builders.
TWELFTH. A fifth-sounding mutation stop, belonging to the
8 ft. harmonic series in the manual divisions, where it is of 2% ft.
pitch; and to the 16 ft. harmonic series in the Pedal Organ, where it
* We have inserted the four stops in the tonal-apportionment of the F^th Organ in our
scheme for the Concert-room Organ. See "The Organ of the Twentieth Century," page 315-
268 ORGAN-STOPS
is of 5 K ft* pitch. In both cases it corroborates the second upper
partial tone of the prime tone, as yielded by the DIAPASON, 8 FT.,
of the manual divisions, and the DIAPASON, 16 FT., of the Pedal Or-
gan. The TWELFTH is properly formed of open cylindrical metal
pipes, of a smaller scale than those of the OCTAVE, 4 FT., and voiced
to yield pure organ-tone, softer than that of the OCTAVE. In old
Organs, the TWELFTH, 2% FT., was almost invariably made of too
large a scale and voiced too loudly. The practice of making the
TWELFTH of the same scale as the DIAPASON is strongly to be con-
demned, for with so large a scale it is not possible to obtain a scien-
tific adjustment of power or the refined singing tone so desirable in
this harmonic-corroborating stop.
The TWELFTH, 2% FT., frequently enters into the composition
of compound harmonic-corroborating stops ; and as it requires to be
covered by the SUPER-OCTAVE, 2 FT., it is sometimes associated with
that stop as a two-rank MIXTURE, as in the RAUSCHQUINTE (q. v.).
For further particulars, see NAZARD.
TWENTY-FOURTH. ItaL, VIGESIMA QUARTA. A third-
sounding MIXTURE rank, corroborating the ninth upper partial tone
in the 8 ft. harmonic series, and the fourth upper partial tone in the
4 ft. harmonic series. The TWENTY-FOURTH does not often appear
in modern MIXTURES, but it should always be introduced in those in
which a third-sounding rank is used. The old English builders in-
troduced it along with the SEVENTEENTH, in many of their MIX-
TURES. When properly scaled and proportioned in tone, the
TWENTY-FOURTH has a very good effect in the bass and tenor oc-
taves. It appears, associated with the SEVENTEENTH (DECIMA
SETTIMA), in the North Organ in Milan Cathedral.
TWENTY-NINTH. Ital, VIGESIMA NONA. An octave-
sounding MIXTURE rank, corroborating the seventh upper partial
tone in the 4 ft. harmonic series, and, philosophically considered,
the fifteenth upper partial tone in the 8 ft. harmonic series. The
TWENTY-NINTH was commonly introduced in the acute MIXTURES
of the old builders. It appears, associated with the TWENTY-SIXTH
(VIGESIMA SESTA), in both the Organs in the Cathedral of Milan.
TWENTY-SECOND. ItaL, VIGESIMA SECONDA. An octave-
sounding harmonic-corroborating rank, of I ft. pitch, commonly
found in MIXTURE work, but sometimes as an independent stop, as
in the Grand Choeur of the Organ in the Cathedral of Notre-Dame,
and the Positif of the Organ in the Church of Saint-Sulpice, Paris.
THEIR ARTISTIC REGISTRATION 269
In both these noted instruments it appears as a PICCOLO, I FT. The
extremely small size of the pipes forming the TWENTY-SECOND
renders it almost impossible to carry the stop throughout the man-
ual compass without a break. In MIXTURE work it is very seldom
carried above the tenor octave. Associated with the NINETEENTH
(DECIMA NONA) it appears in the North Organ, and separately in
the South Organ, in the Cathedral of Milan.
TWENTY-SIXTH. Ital, VIGESIMA SESTA. A fifth-sounding
MIXTURE rank, corroborating the eleventh upper partial tone in the
8 ft. harmonic series, and the fifth upper partial tone in the 4 ft.
harmonic series. The TWENTY-SIXTH appears in the majority of
high-pitched MIXTURES, but never above middle eft 1 , and rarely
going so high as that note. Associated with the TWENTY-NINTH
(VIGESIMA NONA) it appears in the North Organ in Milan Cathedral.
u
UNDA MARIS, Lat. Literally Wave of the Sea. A name used
to designate a single or a dual stop, of 8 ft. pitch, the pipes of which
are usually open and made of metal, though occasionally of wood.
In old Organs, the stop was formed of either open or covered flute-
toned pipes, having soft intonation, and tuned slightly flat, so as to
produce, when drawn with a unison stop of correct pitch, an undu-
lating or wave-like effect. When in its best dual form, the stop
produces slow undulations that may be counted; and not the objec-
tionable fluttering tones which characterize it in Organs constructed
by inartistic builders. This slow undulation, coupled with a soft
flute-tone, properly distinguishes the UNDA MARIS from the Voix
CELESTE.* There seems, however, to be no recognized standard
of tonality for the UNDA MARIS at the present time. In Prance it
has been formed successfully of two ranks of pipes of a QUINTATEN
tonality, tuned flat and sharp. In England and Germany the stop
varies but slightly, if any, from the recognized character of the Voix
CELESTE (g. .) The UNDA MARIS is to be fornd in numerous
* ' ' UNDA- MARIS. Flute de huit, en bois ou en e"toffe. Elle s'accorde une idee plus haut que
la montre, aveclaquelle ce disaccord doitfaire constamment unel6gere oscillation, unbaltement
comme disent les accordeurs. Les ondulations que produit ce battment sont traduites parson
nom latin d'unda-maris. Son effet.assez bizarre, est plus sensible avec des jeux moins accentue"s
que la montre ; et quand ce dernier rgistre est f ortement embouche\ il parle trop haut pourlaisser
entendre son auxiliaire. Quelquefois, Vunda-maris est composee de maniere se passer de ses
voisins; alors, comme la flute double, appel^e bifara, elle a deux bouches & chaque tuyau, maissa
puissance estd'autant moindre que les deux bouches sont aliment6es par un seul pred." Regnier.
270 ORGAN-STOPS
Organs constructed by French and German builders, as on the
Fourth Manual of the Walcker Organ in Riga Cathedral, and in the
Positifs of the Cavaille-Coll Organs in the Cathedral of Notre-Dame,
the Church of Saint-Sulpice, and the Palais du Trocadero, Paris.
Good examples by English builders, formed of two ranks of pipes,
exist in the Organs in Norwich Cathedral and the Centennial Hall,
Sydney, N. S. W. An example exists in the Echo of the Roosevelt
Organ in the Auditorium, Chicago. A dual stop of unusual forma-
tion, named UNDA MARIS by its inventor, George W. Till, of Phila-
delphia, Pa., has recently been inserted in the Choir of the Organ
in the Wanamaker Store in that city. It consists of a GEMSHORN,
8 FT., of wood and metal, and a GAMBA, 8 FT., of tin. The former is
tuned flat so as to produce about three undulations at tenor C, in-
creasing to about double the number in the top octave.* How far it
is desirable to introduce stops of the UNDA MARIS class in an Organ
is open to question; but it is quite certain it is not required in dig-
nified organ music.
UNTERSATZ, Ger. Eng., CONTRA-BOURDON, SUB-BOURDON.
Fr., SOUSBASSE. A large-scaled covered wood stop, of 32 ft. pitch.
It is inserted in both the manual and pedal departments of large
Organs. When in a manual division it is usually labeled MANUAL-
UNTERSATZ, 32 FT., as on the First Manuals of the Walcker Organs
in the Cathedrals of Ulm and Vienna, and the Church of St. Paul,
Frankfurt a. M. It appears, labeled UNTERSATZ, 32 FT., in the Pedal
Organs of the Ladegast instruments in the Nicolaikirche, Leipzig,
and the Cathedrals of Merseburg and Schwerin. See CONTRA-
BOURDON.
v
VIOL. Fr. and Ger., VIOLE. A name that may be accepted
as generic of all labial stops of unimitative string-tone, commonly
known by the inappropriate name GAMBA; which, signifying leg,
has no possible relation to tone. The term VIOL is, accordingly, to
be strongly recommended for adoption in all appropriate cases;
confining the term GAMBA to the only one stop to which it can be
properly applied. The term Viol or Viole was originally applied to a
family of string instruments commonly used in the sixteenth and
seventeenth centuries. The Chest of Viols, mentioned by old
* Pull particulars of the formation of this unique stop are given in " The Organ of the Twen
tieth Century, " pages 106-107.
THEIR ARTISTIC REGISTRATION 271
writers, consisted, in its complete form, of two Bass Viols, two Tenor
Viols, and two Treble Viols : but a smaller Chest, which was much
more common, comprised one Bass, one Tenor, and two Treble
Viols. These instruments were severally strung with seven, six, and
five strings; and had not the powerful tones of the perfected instru-
ments of the Violin family. The term VIOL may be variously
qualified, indicating tonal character ;, such terms as GRAND VIOL,
MAJOR VIOL, MINOR VIOL suggesting themselves.
VIOLA, Ital. The name properly given to an open labial stop,
of 8 ft. and 4 ft. pitch, formed of metal or wood pipes, of small scale,
voiced to yield an imitative string-tone of a fuller and richer quality
than that of the stop known as the VIOLE D'ORCHESTRE or ORCHES-
TRAL VIOLIN, 8 FT. The following names are to be found in impor-
tant Organs. On the Second Manual of the Organ in the Cathedral
of Riga, there is a VIOLA DI ALTA, 8 FT. ; on the Fourth Manual, a
VIOLA TREMOLO, 8 FT.; and in the Schwell-Pedal, a VIOLA, 4 FT.
On the First Manual of the Organ in the Cathedral of Vienna, there
is a VIOLA MAJOR, 16 FT. In the Organ in the Monastic Church,
Weingarten, there is a VIOLA DOUCE, 8 FT.
FORMATION. The pipes of the VIOLA are usually made of metal, preferably
tin, although fine examples have been formed of spotted-metal. The scale should
be somewhat larger than that proper for the VIOLIN, 8 FT. The mouths are of
medium width, low, and furnished with the harmonic-bridge. The pipes are
slotted, and tuned by slides, in preference to coiled or bent tongues. VIOLAS of
beautiful quality of tone have been made of wood. A fine example, made by
Edmund Schulze, exists in the Choir of the Organ in the Church of St. Peter,
Hindley, England. Its scale gives the CC (8 ft.) pipe an internal width of 2$^
inches and a depth of 3^' inches: the mouth is 13 /16 inch in height, and furnished
with a harmonic-bridge of unusual form, attached immediately above a sunk cap,
without the usual supporting ears. The stop speaks on wind of about 2 inches
pressure, and its tone is imitative and remarkably effective.
TONE AND REGISTRATION. It must be obvious to everyone in-
terested in correct and expressive stop nomenclature, that the name
VIOLA should be given only to the stop which yields tones imitative
of those of the orchestral Viola. But the fact is, that in the large
majority of cases the stops to which the name VIOLA has hitherto
been given yield no imitative tones, many being less effective in this
direction than ordinary SALICIONALS. It is, accordingly, most
desirable that in the stop nomenclature of the Organ of the Twen-
tieth Century the name VIOLA should be confined to the stop repre-
senting, in the tonal appointment of the Concert-room Organ, the
Viola of the orchestra; and that its tone should be as closely imita-
tive as the pipe-maker's and voicer's skill can reach.
272 ORGAN-STOPS
In artistic registration, the tones of the VIOLA will be found much
more valuable and of wider range in combination than the thinner
and more cutting tones of the VIOLIN, especially as that stop has
lately been made and voiced. The full and rich tones of the VIOLA,
8 FT., will impart strength and peculiar color to every combination
into which they enter. With pure organ-tones and flute-tones, un-
imitative and imitative, they will produce very beautiful compound
tonalities; while they will combine perfectly with the voices of all the
softer lingual stops, creating singular and valuable tonal colorings.
The VIOLA properly belongs to the String Division of the Concert-
room Organ; but it may, with considerable advantage, find a place
in other manual divisions, and in any expressive division of the
Church Organ.
VIOLA DA GAME A, Ital. FT., VIOLE DE GAMBE. The only
stop in the Organ in the name of which the word Gamba can be
properly introduced. The name of the stop is derived from the old
Viola da Gamba, a large string instrument, and the precursor of the
Violoncello; which, like the latter, was supported by the legs of the
player, hence the peculiar name. The stop is of 8 ft. pitch, and is
formed of open metal pipes of small scale, usually cylindrical in
form, but have frequently been made conical, surmounted by a long
and slender bell. Owing to the trouble and expense of making pipes
of this compound form, they now rarely, if ever, appear in the Organ.
See BELL GAMBA. The VIOLA DA GAMBA is to be found in nearly all
important German Organs; but it is very seldom of an imitative or
refined quality of tone. It exists, of 8 ft. pitch, in most of Walcker's
larger instruments, and placed invariably on the First Manual
(Great). On that of the Organ in the Cathedral of Riga, the stop is
inserted of 16 ft. and 8 ft. pitch. The VIOLE DE GAMBE, 8 FT., ap-
pears, often duplicated, in almost all of Cavaille-Coll's important
instruments. In the Organ in the Cathedral of Notre-Dame, it is
inserted in both the Grande-Orgue and the Recit. The stop is to
be found in numerous English Organs, though rarely in those of
recent construction. It has usually been inserted of tenor C com-
pass, as in the Organ in the Foundling Hospital, London, in which it
is placed in both the Choir and Swell. The VIOLA DA GAMBA, 8 FT.,
is inserted in the Choir, Swell, and Solo of the Organ in St. George's
Hall, Liverpool; furnishing the only unison string-tone in those divi-
sions of the instrument. Name incorrectly rendered VIOLA DI
GAMBA.
TONE AND REGISTRATION. Properly formed and artistically
THEIR ARTISTIC REGISTRATION 273
voiced, the stop should yield a tone imitating, as closely as practi-
cable, that of the old Viola da Gamba; the tone of which was of a
lighter and less expressive quality than that of the Violoncello, to
the construction of which it led the way, and before the superiority
of which it ultimately fell into disuse. The Viola da Gamba, or Bass
Viol, had seven strings tuned in fourths, with a major third between
the third and fourth strings. From these facts it will be realized that
the voice of the VIOLA DA GAMBA of the Organ should be distinctly
subordinate to that of the VIOLONCELLO, 8 FT. ; and on account of
this tonal subordination it will be of considerable value in artistic
registration. Imitative string stops of a powerful and penetrating
tonality are of very limited use in registration, save in that of a pro-
nounced orchestral character, or for the production of assertive solo
effects. With the DOPPELFLOTE, ROHRFLOTE, QUINTATEX, 8 FT.,
CLARABELA, and other flute-toned stops, the VIOLA DA GAMBA pro-
duces compound tones of great beauty and fine color, suitable for
both accompanimental and melodic passages.
VIOLA D'AMORE, ItaL Fr., VIOLE D'AMOUR. The name
given to an open labial stop, of 8 ft. and 4 ft. pitch; formed of very
small-scaled pipes of metal (preferably tin), or partly of wood and
metal when of 8 ft. pitch. The stop derives its name from the beau-
tiful old instrument called the Viola d'Amore; which is a Tenor Viol,
having seven strings played with the bow, and underneath which,
passing through the bridge and under the finger-board, are stretched
and tuned slender metal strings (from ten to fourteen in number)
which sound sympathetically with the bowed strings.
The VIOLA D'AMORE of the Organ is the softest and sweetest of
the imitative string-toned stops, and deserves a prominent place in
every true Chamber Organ, in which refinement and delicacy of
tone is preferred to musical noise. It also should find a place in the
Choir or Echo of the properly appointed Church Organ. When
formed of metal, the pipes are cylindrical, and are voiced with the
harmonic-bridge, or the appliance called the frein harmonique.
Wood pipes may be used, in the unison stop, from CC to tenor F,
made either straight or pyramidal, having thin sides of choice spruce
and fronts of a close-grained hard wood. The scale of the CC pipe
should not exceed 2 inches square. All to be voiced with the har-
monic-bridge. The VIOLA D'AMORE is by no means a common stop,
preference being given to string-toned stops of more assertive tonal-
ity. German examples exist, incorrectly labeled VIOLA D'AMOUR,
in the Organs in the Cathedrals of Riga and Liibeck, and St. Petri-
18
274 ORGAN-STOPS
kirche, Hamburg. A VIOLE D' AMOUR, 4 FT., is inserted in the Recit
of the Organ in the Church of Saint-Ouen, Rouen, where it is asso-
ciated with two unison soft-toned stops, the Voix EOLIENNE and
Voix CELESTE. A VIOLE D 'AMOUR, 8 FT., and another of 4 ft. pitch
exist in the Echo of the Organ in the Centennial Hall, Sydney, N.
S. W.
TONE AND REGISTRATION. When carefully made and voiced by
a master-hand, the tone of the stop should closely resemble that of
the old Viola d'Amore, thus described by Berlioz: "The quality of
the Viole d' Amour is faint and sweet; there is something seraphic
in it partaking at once of the Viola and the harmonics of the Violin.
It is peculiarly suitable to the legato style, to dreamy melodies, and
to the expression of ecstatic or religious feelings. " Though it does
scant justice to the appealing voice of this unique and beautiful
instrument, it is, perhaps, sufficient to convey a fair idea of the
peculiar singing tone of the Viola d'Amore to those who have not
enjoyed the privilege of hearing the instrument played upon, and
studying its complex tonality the privilege that has been enjoyed,
under varied and favorable conditions, by the writer.
It cannot be difficult for the organist, who has given some special
attention to the subject of tone creation, to realize the important
r6le the VIOLA D'AMORE, with its soft, singing, and sympathetic
voice, can play in the production of numerous delicate and beauti-
ful tone-colors, based on foundations of pure organ-tone, produced
by such stops as the ECHO DIAPASON and DULCIANA; of flute-tone,
produced by the MELODIA, FLAUTO DOLCE and FLAUTO D'AMORE;
and of reed-tone, produced by the CHALUMEAU and OBOE D'
AMORE. The VIOLA D'AMORE, 8 FT., should be introduced in the
String-Organ Ancillary or otherwise of every properly appointed
Concert-room instrument; and in the Choir of every large Church
Organ, where its sympathetic voice will be of the greatest value in
accompanimental music.
VIOLA SORDA, ItaL A stop, of 8 ft. pitch similar in form to,
but smaller in scale than, the VIOLA. Its voice is intended to imitate
the tone of the muted Viola of the orchestra. If a very subdued in-
tonation, combined with a slight modification of timbre, is desired,
the pipes forming the stop may be conical in form : the scale giving
the CC (8 ft.) pipe the diameter of 2 inches at the mouth line, and of
7 /s inch at top. Artistically voiced, the tone of this MUTED VIOLA
will be found extremely valuable in refined registration. Its proper
place is in the String Ancillary of the Concert-room Organ.
THEIR ARTISTIC REGISTRATION 275
VIOLE A PAVILLON, Fr. The name that has been employed
by French organ-builders to designate a string-toned labial stop, of
8 ft. pitch, the pipes of which have conical bodies surmounted by
bells (pavilions), similar to those of the BELL GAMBA (q, v.) ; and, like
them, tuned by means of larger flexible ears. The pipes were usually
made of tin; and voiced to yield a soft and agreeable string-tone.
The stop has fallen into disuse.
VIOLE CELESTE, Fr. The name given to a stop of the VIOL
class, of 8 ft. pitch, the pipes of which are tuned sharp so as to pro-
duce a bright undulatory effect in combination with another unison
stop (preferably of string tone) correctly tuned. As a dual stop, it is
properly formed of two softly-voiced VIOLS, of 8 ft. pitch, one of
which is tuned a few beats sharp, sufficient to create an agreeable
tremolo, but not sufficient to produce an objectionable out-of-tune
effect. See Voix CELESTE.
VIOLE D'ORCHESTRE, Fr. Eng., ORCHESTRAL VIOLIN.
The appropriate name given by William Thynne, organ-builder of
London, to a stop constructed by him; the tone of which is, in our
opinion, the most satisfactory imitation of that of the orchestral
Violin which has been produced by organ-pipes up to the present
time (1920).* In the construction and voicing, Thynne followed the
teaching of Edmund Schulze, but surpassed him in the production
of Violin tone from metal pipes. Fine examples of the VIOLE D'OR-
CHESTRE exist in Organs built by William Thynne. For further
particulars, see VIOLIN.
VIOLE SOURDINE, Fr. The name given by William Thynne,
organ-builder, of London, to a delicately voiced string-toned, open
metal stop, the tone of which is imitative of that of the muted
Violin. It was first introduced, by its inventor, in the Organ now in
Tewkesbury Abbey. The stop is formed of slender cylindrical pipes,
smaller in scale than those of the Thynne VIOLE D'ORCHESTRE, but
voiced in a similar manner with the harmonic-bridge. The tone of
the stop is extremely refined and beautiful; and, on account of its
subdued and perfect mixing quality, it may be recognized as the
most generally useful of the imitative string-toned stops in artistic
registration with stops of contrasting tonalities.
* We have had pipes of the Thynne VIOLE D'ORCHESTRE tested, by being sounded with corre-
sponding notes of a fine Violin, under conditions which prevented the hearers from positively
knowing which was speaking. It was found impossible to decide which instrument was produc-
ing the notes; or, when both were sounded together in unison, which one became silent. No
severer test could well be instituted. How many of the stops bearing the name, made by other
organ-builders, would stand so exacting a comparison?
276 ORGAN-STOPS
Modifications of the Thynne stop, subsequently introduced and
appropriately named MUTED VIOLS, are formed of very small-scaled
pipes, conical in shape, seldom exceeding, in the CC pipe (8 ft.), I J^
inches in diameter at the mouth line and tapering to about half that
diameter at top; having mouths rarely exceeding one-sixth of the
circumference, furnished with the harmonic-bridge. Like the orig-
inal VIOLE SOURDINE, these MUTED VIOLS yield tones of great
beauty and value.
VIOLETTA, ItaL Literally Small Viol. The name appro-
priately given by Italian organ-builders to a string-toned stop, of
4 ft. pitch, the pipes of which are open, cylindrical, of small scale,
and made of tin. The stop is properly voiced to yield an imitative
tone of medium strength, which is very valuable in combination
with the unison stops of imitative string-tone; brilliantly corrob-
orating their first and most important upper partial tones, while
enriching them with its own series of less pronounced harmonics.
As an Octave, the VIOLETTA, 4 FT., is extremely valuable in artistic
registration, imparting a bright and special coloring to all unison
tones with which it may be combined. The Organs of to-day are
very deficient in stops of 4 ft. pitch; far too much reliance being
placed on the crude expedient ot octave coupling, which, in its best
form, interferes with the desirable independence of the claviers. This
is a matter deserving careful consideration.
VIOLIN. ItaL, VIOLINO. Fr., VIOLON. Ger., VIOLINE. An
open, metal labial stop, of 8 ft. pitch, the tone of which imitates, as
closely as practicable in organ-pipes, that of the orchestral Violin.
This name appears at an early date in stop nomenclature; for in
Bernard Smith's "Schedule" of the Organ he erected in the Temple
Church, London, in 1688, we find mentioned "A VIOLL and VIOLIN
of mettle, 61 pipes, 12 foote tone/* The compass being GGG to g 3 .
Notwithstanding the early use of the name, it seems very improb-
able that the builders of the seventeenth century knew anything of
a highly imitative stop, such as we would now consider worthy of
representing the Violin in the stop-appointment of the Organ. The
modern use of the name appears to be extremely free and meaning-
less. On one hand, we find in the Pedal Organ of the Schulze instru-
ment in the Church of St. Bartholomew, Annley, a VIOLIN, 16 FT.,
and, on the other hand, turning to American organ-building, we find
in the Choir of the Organ in the Music Hall, Cincinnati a VIOLIN,
4 FT. In fact, neither of these stops can be correctly called a VIOLIN
for neither is of proper pitch or imitative in tonality. To prevent
THEIR ARTISTIC REGISTRATION
277
confusion, the practice should be adopted of labeling the strictly
imitative stop either ORCHESTRAL VIOLIN or VIOLON D'ORCHESTRE,
8 FT. The single terms are not in common use; but we find a Vio-
LINE, 8 FT., on the First Manual of the Organ in the Cathedral of.
Ulm, and a VIOLIN, 8 FT., in the Pedal of the Organ in Paulskirche,
Frankfurt a. M. Neither of these stops is likely to be strongly
imitative.
SCALE AND FORMATION. The variation of the scales which have been adopted
in the formation of stops of imitative string-tone is greater than that shown in any
other class of labial stops. This statement is supported by the fact that the scales
used by William Thynne and other eminent labial pipe voicers for the CC (8 ft.)
pipes of their VIOLINS, range from a diameter of 3.13 inches to the small diameter
of 1. 08 inches; the former scale being developed on the ratio of i : 2.519, halving on
the nineteenth pipe; and the latter mainly on the ratio 1:2, halving on the twenty-
fifth pipe. Thynne did not highly favor the adoption of very small scales, and our
FIG. 35
experience leads us to agree with his practice; for we have never heard VIOLINS
equal in fullness, richness, and imitativeness to those voiced by his master-hand.
The thin, scratchy, and penetrating tones of the over-blown, small-scaled VIOLINS
of certain makers are offensive to the cultivated ear, and of little value in artistic
registration.
The pipes of the VIOLIN are invariably cylindrical; and, chiefly on account of
their extremely small scale, should be made of tin or very high-grade alloy of tin
and lead. They are usually slotted, and tuned by a metal slide in good work. It
is very questionable if slotting is favorable to the tone, for it has a tendency to
impart a horn-like timbre. The mouths of the pipes vary slightly in their widths.
In the largest scale adopted by Thynne, the mouth of the CC pipe is two-ninths its
circumference; this width being graduated to one-third the circumference at treble
c a , that proportion being carried to the top note. The heights of the mouths vary
between one-fourth and one-third their widths, according to the wind-pressure
used, and the character of the tone desired. The VIOLIN has been artistically
voiced on pressures varying from 2% inches to 15 inches. The VIOLIN is invari-
278 ORGAN-STOPS
ably, and of necessity, voiced with the harmonic-bridge. In the accompanying
illustration, Fig. 35, is shown the mouth of a Thynne C pipe, accurately drawn
from one presented to us by its distinguished voicer and long a valued friend.
The harmonic-bridge is semi-cylindrical in form, and of hard alloy, is shown in the
Section, as carried between the ears, and in its position with regard to the mouth.
This position is a matter of extreme nicety. Cylindrical bridges, formed of alu-
minium tubing or some hard wood, are commonly used by voicers.
TONE AND REGISTRATION. Accepting the tone of the VIOLIN,
8 FT., to be as close as practicable to that of the Violin of the orches-
tra, as we have found it to be in certain examples (see VIOLE D'OR-
CHESTRE), its value in registration will be found in some directions
to be considerable, while in others it will be questionable. The dis-
tinctive tonality of the stop so assertive and penetrating renders
it unsuitable for general combinational purposes. With certain
labial stops, notably the FLUTES, it will not fully combine; always
dominating and asserting its individuality. With other and unimi-
tative string-toned stops it unites agreeably, imparting richness,
brightness, and force; creating volumes of string-tone very necessary
in orchestral effects : accordingly, its proper place is in the special or
Ancillary String Organ, where it joins in producing the volume of
orchestral string-tone required in the adequate and artistic rendition
of orchestral compositions. On this subject we have some right to
speak, having been the first, in the history of organ-building, to in-
troduce a complete and independent String Organ into the tonal
appointment of the Organ. The VIOLIN combines well with most of
the full-toned lingual stops ; largely losing its identity in their some-
what kindred voices, so far as their harmonic structure obtains,
which it improves in effectiveness and expressive force. A really
fine imitative VIOLIN, 8 FT., is par excellence a solo stop, especially
valuable in its compass from fiddle G to c 5 ; above that its imitative-
ness falls short of what is desirable. The bass and tenor notes from
CC to F$ will never be used in a correctly rendered Violin solo, for
these belong to the Viola and Violoncello, but their necessity and
utility cannot be denied. Besides, stops of short compass, or formed
of pipes of different tonalities in different portions of their compass,
are to be condemned from every artistic point of view, whatever
argument may be advanced in their favor.
VIOLINA. The name given by certain organ-builders to a
string-toned stop of 4 ft. pitch, similar to that called VIOLETTA, 4 FT.
(q. v.). Examples exist in the Swell of the Organ in the Music Hall
Cincinnati, and in the Swell of the Organ in the Cathedral of the
Incarnation, Garden City, Long Island.
THEIR ARTISTIC REGISTRATION 279
VIOLIN DIAPASON. The term that has been used by Eng-
lish organ-builders to designate an open labial stop, of 8 ft. pitch,
similar in all essentials to that more commonly known as the GEI-
GENPRINCIPAL (q. V.).
VIOLINO SORDO, ItaL Literally Muted Violin. A stop, of
8 ft. pitch, similar in form to, but voiced softer than, the VIOLIN.
Its voice is intended to imitate the tone of the muted Violin of
the orchestra. Under the name VIOLE SOURDINE (q. a.)., the stop
was first introduced by W. Thynne, of London. As the imitative
tone of the VIOLINO SORDO is much less assertive and penetrating
than that of the VIOLIN, 8 FT., it is accordingly much more gener-
ally useful in registration. It will be found a valuable stop in the
soft accompanimental division of the Church Organ, and should
find a place in the special or String Ancillary of the Concert-room
Organ, where it will be valuable in combinational and delicate solo
effects.
VIOLINO VIBRATO, Ital. A stop of 8 ft. pitch similar in the
form and scale of its pipes to the VIOLIN (q. v.) t but slightly softer in
tone; occupying, in this respect, an intermediate place between the
VIOLIN and the VIOLINO SORDO (q. .). The distinctive peculiarity
of this stop lies in its being tuned a few beats sharp, so as to produce
a wavering effect when sounded in combination with a correctly
tuned unison string-toned stop. As the VIOLINO VIBRATO occupies
a similar place in the tonal scheme of the Organ to the VIOLONCELLO
VIBRATO, its offices in the stop-apportionment of the Ancillary String
Organ may be considered identical. Such being the case, reference
may properly be made to what is said under VIOLONCELLO VIBRATO.
VIOLONCELLO, Ital. Fr., VIOLONCELLE. An open labial
stop, of 8 ft. pitch, the pipes of which are cylindrical when of metal
and square when of wood, voiced to imitate, as closely as practicable
in organ-pipes, the tone of the orchestral Violoncello. When made
of metal, the pipes are cylindrical and of small scale, but somewhat
larger than the scales commonly adopted for the VIOLIN; and are
properly voiced to yield a fuller and richer tone than that produced
by either the VIOLA or VIOLIN. It is obviously desirable, as it is
essential, that these three important unison imitative string-toned
stops should have their voices clearly distinct both in volume and
quality ; otherwise, it would be undesirable to introduce them all in
any one Organ. The VIOLONCELLO is the only one of the three which
is imitative down to the CC note; and this fact gives it special value
280 ORGAN-STOPS
both in solo and combinational effects. A very fine VIOLONCELLO,
8 FT., voiced by William Thynne, exists in the Organ built under his
directions, now in Tewkesbury Abbey.* This is one of the two
finest VIOLONCELLOES we have ever heard in Organs. The most
beautiful VIOLONCELLO formed of wood pipes, known to us, is that
in the Solo of the Concert-room Organ in the Battersea Polytechnic,
London, the work of John W. Whiteley, one of England's most cele-
brated labial pipe voicers. The pipes of this interesting and unique
stop present several novel features deserving the attention of all
interested in the formation of imitative string-toned stops.f A
VIOLONCELLO, 8 FT., of the ordinary type exists in the Great of the
Organ in St. George's Hall, Liverpool.
TONE AND REGISTRATION. Ttxe tone of the VIOLONCELLO being
imitative of that of the orchestral Violoncello, it is obvious that its
utility in the tonal appointment of the Organ, and especially of the
Concert-room Organ, does not admit of question. It is valuable
both in solo effects and in registration. As a solo stop, it is to be
desired in the Pedal and Solo of the Concert-room Organ ; while in
the Ancillary String Organ it is indispensable. It has, in some form
or other, been usually inserted in the Pedal Organ, and invariably,
hitherto, without being given powers of expression, seriously limit-
ing its usefulness. It is remarkable how seldom this beautiful and
valuable stop has been introduced on the manual claviers of even
large Organs. The tone of a properly formed and voiced VIOLON-
CELLO is much more valuable in artistic registration than the thinner
and more penetrating tone of the VIOLIN ; and this fact should not be
overlooked in scheming the tonal appointment of an Organ.
VIOLONCELLO SORDO, ItaL Literally Muted Violoncello.
A stop, of 8 ft. pitch, similar in form to, but voiced softer than, the
VIOLONCELLO. Its tone is intended to imitate that of the muted
Violoncello of the orchestra. As the imitative tone of the VIOLON-
CELLO SORDO is less assertive than that of the VIOLONCELLO, it will
be f ound more generally useful in registration with the softer-toned
stops, with which it will form numerous combinations of fine and
varied colorings. Drawn with the VIOLONCELLO VIBRATO (q. v.),
* During the Recital at the Dedication of this Organ, at which we were present, the most
beautiful number was a Sacred Song, accompanied by that stop only. The effect produced by
the pure soprano voice and the sympathetic and expressive SOLO VIOLONCELLO was one we have
never forgotten.
f Pull particulars and dimensioned drawings of the formation of the pipes of this stop are
given in "The Art of Organ-Building," Vol. II. t pp. 475-6; and in "The Organ of the
Twentieth Century, " pages 447~8, Plate XXIX.
THEIR ARTISTIC REGISTRATION 281
It will produce a beautiful VIOLONCELLE CELESTE, 8 FT. The most
important place for the VIOLONCELLO SORDO is in the Ancillary
String Organ ; but, when such a tonal division does not exist, it may
be properly inserted in the chief accompanimental division of the
Church Organ, or the softest-toned division of the Concert-room
Instrument, where it will be valuable in combinational and delicate
solo effects.
VIOLONCELLO VIBRATO, ItaL A stop, of 8 ft. pitch, simi-
lar in the form and scale of its pipes to the VIOLONCELLO (g. v.), but
slightly softer in tone; properly occupying, in this respect, an in-
termediate place between the VIOLONCELLO and the VIOLONCELLO
SORDO (q. v.). The distinctive peculiarity of the stop under con-
sideration lies in its being so tutted slightly sharp as to produce a
wavering effect when sounded in combination' with a correctly tuned
unison string-toned stop, preferably the VIOLONCELLO SORDO, imi-
tating as closely as possible, under the conditions, the effect of the
vibrato on the orchestral Violoncello. Drawn with such stops as the
VIOLA SORDA, VIOLE SOURDINE, or VIOLA D'AMORE, it forms an
effective VIOLONCELLE CELESTE. In full combinations of imitative
unison and harmonic-corroborating octave, mutation, and com-
pound stops, such as are provided in the proper stop-apportioned
Ancillary String Organ, the VIOLONCELLO VIBRATO imparts to the
volume of string-tone the nervous force which characterizes the full
string effects of the grand orchestra a forcefulness much to be
desired in the adequate and truly artistic rendition of important
orchestral compositions, or improvisations of an orchestral charac-
ter and coloring.
VIOLONE, ItaL Fr., VIOLON-BASSE. An open labial stop, of
1 6 ft. pitch, the pipes of which are of small scale and formed of either
metal or wood. It has almost invariably been confined to the Pedal
Organ, and when artistically voiced is accepted as the organ equiva-
lent to the Contrabasso of the orchestra; and its excellence is judged
in proportion to the closeness of its imitation of the tone of that
instrument. Artistic voicers, accordingly, endeavor to impart to
their VIOLONE s as much of the harmonic richness of the Contrabass
as their expedients in formation and skill in voicing can accomplish;
aiming also to secure that peculiar rasping effect which imitates the
attack of the bow on the string, and which is only heard in the finer
specimens of the stop. Although hitherto the VIOLONE, in its proper
form and tonality, has been practically confined to the Pedal Organ,
it is most desirable that it should find its proper place in the manual
282 ORGAN-STOPS
Organs, especially in those of the Concert-room Instrument. As its
pipes admit of mitering, it can be accommodated in swell-boxes of
the size proper in Organs of large dimensions. When necessary, its
lowest octave may be of covered pipes; or, indeed, the stop may be
formed of covered pipes throughout.
The ordinary metal VIOLONES, such as exist in important Eng-
lish Organs, are not strongly imitative, and are slightly slow in
speech, their notes sounding with more or less of a preparatory rasp.
Representative examples exist in the Willis Organ in the Royal
Albert Hall, London, and the Hill Organ in Centennial Hall, Sydney,
N. S. W. As a rule the VIOLONES of the German builders, com-
monly named VIOLONBASS, 16 FT., are somewhat tardy in speaking
their full tones, and on that account have frequently to be drawn
with another and prompt-speaking stop or "helper/ 1 such as a
softly-toned FLUTE, 8 FT. In the Pedal of the Organ in the Stifts-
kirche, Stuttgart, the stop, labeled VIOLON, 16 FT., is of two ranks of
pipes the stop proper of 16 ft. and its helper of 8 ft. Under the
name VIOLONBASS, 16 FT., the stop exists in the Pedals of the Organs
in the Cathedrals of Riga, Ulm, Lubeck, and Frankfurt, and in
almost all the important Church Organs built by Walcker. The
VIOLON-BASSE, 16 FT., is not a common stop in French Organs, but
examples exist in the Pedals of the Organs in the Palais du Troca-
dero and the Basilique du Sacre-Coeur, Paris.
The finest and most thoroughly imitative VIOLONE, 16 FT., that
we have ever heard, is that in the Pedal of the Organ in the Church
of St. Peter, Hindley, Lancashire. The stop is labeled VIOLONBASS;
and its pipes are of wood and of small scale, the CCC (16 ft.) pipe
measuring, internally, only 5% inches square.* Another fine
example, labeled VIOLON, 16 FT., exists in the Pedal of the large
Organ (94 speaking stops) in the Parish Church of Doncaster, York-
shire, England. Both these Organs were constructed by Edmund
Schulze, of Paulinzelle, and the stops alluded to were voiced by
him.f In the Organ in the Marienkirche, Lubeck, is another repre-
sentative example by the same artist.
* A full description of the formation of this stop, accompanied by accurate drawings, is given
in "The Art of Organ-Building, " Vol. II., pp. 470-1 ; and in ** The Organ of the Twentieth Cen-
tury, " pages 443-5-
f In a description of the Doncaster Organ, from the pen of an authority, the following per-
tinent remarks obtain: "The individuality of the stops is remarkable, and must be ascribed to
the artistic feeling possessed by all genuine organ-builders who perform that particular part of
the work the voicing themselves, and do not delegate so important a branch of the art to
workmen who, however skillful, will not on occasions devote the care necessary.
"The wood pipes of this instrument are very solid, clear, and firm in tone ; and the muddiness
often heard in many Organs from wooden stops, is not perceptible in this. Professor Tcepfer.
THEIR ARTISTIC REGISTRATION 283
VIRGINREGAL. Ger., JUNGFERNREGAL. An old lingual
stop, of 8 ft. and 4 ft. pitch, the tone of which was, in the best ex-
amples, soft and pleasing. The origin of the name has not been de-
cided; but it is supposed that the stop was originally in a Portative,
and was commonly played by young maidens. An example, of 4 ft.
pitch, was inserted in the Pedal of the Organ in the Church of SS.
Peter and Paul, Goerlitz, built by Eugenius Casparini & Son, and
finished in 1703. Under the name JUNGFERNREGAL, the stop ap-
pears in the Choir of the Organ, of 71 speaking stops, in the Church
of St. Dominico, Prague. According to Hopkins, this stop is of 16
ft. pitch. See REGAL.
VOCE FLEBILE, ItaL Literally Mournful Voice. The name
given to stops, of 16 ft. pitch, inserted in the Swell of the Organ, of
49 speaking stops, in the Church of St. Alessandro, Milan. The
name is expressive of the subdued and colorless character of their
tones. We have not found the name in any other Italian Organ.
VOGELFLOTE, Ger. Literally Bird-Flute. The name given
to a flute-toned stop, of 4 ft. pitch, voiced to yield a clear sound,
somewhat resembling that of a bird's song. The stop was inserted
on the First Manual of the Organ in the Church of Quittelsdorf,
constructed by Andreas Schulze, of Paulinzelle, in the year 1791
VOGELGESANG, Ger. Lat., AVICINIUM. Port., PASSARIN-
HOS. A stop which has been introduced in some old Organs, and
which may be classed amongst the curiosities of organ-building. It
was devised to imitate the warbling of birds. It was formed of three
or more small metal pipes, of different tones, bent, and partly im-
mersed in water.* The stop labeled PASSARINHOS in the Organ in
the Church of the Martyros, Lisbon, built in 1785, is formed of six
pipes so disposed. Stops of the class were introduced in several old
German Organs.
VOIX CELESTE, Fr. Lat., Vox CCELESTIS. Span., Voz
CELESTE. The name employed to designate an open labial stop, of
8 ft. pitch, formed of one, two, or three ranks of small-scaled pipes,
one or two of the ranks being tuned slightly sharp or flat to the
the author of elaborate works on organ-building, says, *that wooden pipes in theirloweroctaves
as voiced by Herr Schulze. are as good as, or even superior to, metal pipes.' "
*" VOGELGESANG. Unter alien lappischen Spielereien die grosste, indem das Gezwitscher
der Vogel nachgeahmt werden sollte ! Welch ein Einfall ! ! Diess wird auf f olgende Artbewerk-
stelligt : In einem blechernen Kastchen, das mit Wasser gef ullt wird, ragen 3 6 kleinePf eif chen
mitihrem obern Endein das Wasser, und geben dann eiuen gurgelnden, zwitscherndenTon."
W. Schneider (Merseburg, 1835)-
284 ORGAN-STOPS
correct unison pitch of the Organ in which the stop is inserted. In
the large majority of examples, the stop is of one rank only, of either
string-toned or pure organ-toned metal pipes, tuned a few beats
sharp; which, when combined with another unison stop of correct
pitch, produces a peculiar tremulous effect; and which, by a wonder-
ful stretch of the imagination, has been likened to a celestial voice,
whatever that may be. In some instances, the single-rank Voix
CELESTE has been tuned flat, but its effect is not so satisfactory as
in the case of the sharp stop. In our opinion, the flat tuning should
be confined to the UNDA MARIS (q. v.). The most important single-
rank stops, tuned sharp, are those designed to impart to the volume
of sound produced by the massing of imitative string-toned stops
the nervous force which characterizes the full string effects of the
grand orchestra. See VIOLINO VIBRATO and VIOLONCELLO VIBRATO.
The most satisfactory Voix CELESTE is that formed of two ranks
of pipes of special tonality, carefully voiced to produce a pleasing
and refined effect, one of which is voiced just sufficiently sharp to
create an artistic tremolo, not too pronounced. To avoid having to
tune one rank too discordant with the correct pitch of the Organ, the
discordancy has been divided by tuning one rank very slightly flat
and the other rank equally sharp. This variety of the stop is to be
preferred. In the Organ in St. Paul's Cathedral, London, recon-
structed by Willis, in 1901, there is a Voix CELESTE of three ranks,
tuned flat, correct, and sharp. A stop of this formation could be
made a valuable timbre-creator.
VOIX EOLIENNE, Pr. The name given by Cavaille-Coll to a
labial stop, of 8 ft. pitch, inserted in the Recit Expressif of the Organ
in the beautiful Church of St. Ouen, Rouen. The stop is tuned sharp
and according to Philbert was designed to produce undulations of
tone when drawn along with the FLUTE HARMONIQUE, 8 FT.* This
seems worthy of notice from the fact that the FLUTE alluded to is
placed in the Grand-Orgue. Such a disposition of the two stops
* Alluding to this stop, M. Philbert remarks: "A Saint-Ouen, on rencontre encore un troi-
siemejeuondulantja Voix EOLIENNE, destine a produire 1'ondulation avec la FLUTE HARMO-
NIQUE. Ilconsiste en unerangSe de tuyaux bouches accord6s a battements, et 1'effet m'ena paru
mSdiocrementsatisfaisant, parce qu'il est un peu lourd. Ce que j'ai vu de mietix comme jeu
ondulant destine a agir sur une FLUTE on un BOURDON, c'est le suavial de la Suisse allemande,
forni6 de tuyaux ouverts, de taille a peu pres identique a celle du SALICIOKAL, mais dont le pied
n'admetabsolument qu'un filet d'air, de facon que le son propre en est extremement faibls et se
perd pour ainsi dire dans celui du jeu auquel on 1'associe, La difference d'accord est en mme
temps treslegere, au point que 1'ondulation est k peine perceptible comme battement et nefait
qu'imprimer au timbre de la PLUTE et surtout du BOURDON comme une teinte crystalline un peu
vaguereellement suave et empreinte de recueillement. Vogt aimait assez ce jeu et s'en servait
habilement. " Causerie sur le Grand Orgue a Saint-Onen de Rouen, p. 30.
THEIR ARTISTIC REGISTRATION 285
deserves consideration, for it opens a nice question in artistic regis-
tration. Doubtless Cavaille-Coll, the most scientific organ-builder
of his time, realized the possibility of beautiful effects being produced
by the combination of the uniform flute-tone and the changing and
expressive tone of the Voix EOLIENNE.
VOX ANGELICA, Lat. Fr., Voix ANGELIQUE. ltd., FLAUTO
ANGELICO. Ger. , ENGELSTIMME. These names appear to have been
employed to designate stops of any description which yielded some
specially beautiful and refined quality of tone. They have, accord-
ingly, been applied to both labial and lingual stops. When applied
to the former, the term Vox ANGELICA properly indicates an open
metal labial stop, of 8 ft. pitch, the pipes of which are cylindrical
and of very small scale, voiced to yield the softest unison tone in
the manual department of the Organ. Fine examples of this form,
voiced, on very light wind, by Edmund Schulze, exist in the Echos
of the Organs in the Parish Church of Doncaster and the Church of
St. Bartholomew, Armley, Yorkshire, England. We have been able
to find one example in Walcker's Organs; the Vox ANGELICA, 4 FT.,
on the Fourth Manual of the Organ in Riga Cathedral. Soft-toned
stops have not been favored by German organ-builders to any note-
worthy extent. The labial Vox ANGELICA, 8 FT. may be either
organ-toned or string-toned, according as it is voiced to approach
the DULCIANA or the SALICIONAL or VIOL.
Both Wolfram and Seidel describe the Vox ANGELICA as a lingual
stop, of 8 ft. pitch, but do not give any particulars respecting its
formation. The former writer, alluding to its tone, says its name is
the best part of it, which, when one thinks of what the lingual stops
were in his day and country, may be readily accepted as a just state-
ment. Clarke, in his " Outline of the Structure of the Pipe-Organ,"
describes the stop as formed of " Free-reed pipes of the most delicate
voicing," giving no further information. In this form it was in-
serted in the Solo of the Roosevelt Organ erected in the Church of
St. Thomas, New York City. In the same form it exists in the Choir
of the Organ in the Church of St. Paul, Antwerp.*
VOX GRAVISSIMA, Lat. The appropriate name to designate
the acoustical tone or effect which is produced by the simultaneous
sounding of stops of 32 ft. and 21^ ft. pitch, standing at the inter-
val of a perfect fifth apart. It is the differential tone generated by
*"L' ANGELICA (Vox) est la premiere maniere de Voix humaine; elle avait stir la nouvelle
1'avantage d'avoir 6te construite d'apres une pens4e sinon ang&lique, du moms religieuse et non
purement humaine. " Regnier.
286 ORGAN-STOPS
the combination that is the Vox GRAVISSIMA, 64 FT. For full par-
ticulars see GRAVISSIMA.
VOX HUMANA, Lat Fr. Voix HUMAINE. ItaL, VOCE
UMANA. Ger., MENSCHENSTIMME. A lingual stop, of 8 ft. pitch, the
pipes of which, in the most satisfactory examples, have resonators
of cylindrical form, covered at top, and slotted for the emission of
wind and sound, and much shorter than the standard lengths of
open pipes. Other forms of resonators have been employed with
varying and, generally, very unsatisfactory results. The most
common form is cylindrical, very short, and entirely open at top;
the tones of pipes of this form are generally objectionable. The stop
is voiced with the view of producing tones, rich in harmonics, in
imitation of the human singing voice ; hence the name Vox HUMANA :
but even the best results that have been obtained up to the present
time fall far short of what is to be desired. The Vox HUMANA re-
quires the aid of the TREMOLANT to impart the characteristic in-
tonation to its voice ; and that is greatly improved by the addition
of a soft unison, unimitative flute-toned stop, such as the MELODIA,
8 FT., which would impart desirable body to the usually thin and
nasal voice of the stop. In high-class work, the solo Vox HUMANA
should be a dual stop, permanently formed of the lingual rank,
associated with a properly voiced, body-giving, labial rank of unison
pitch. If the Vox HUMANA is properly made and artistically voiced
it forms, when drawn without the TREMOLANT, a valuable stop in
registration, imparting a distinctive coloring to soft combinations
of contrasting tonalities. The tonal and imitative effects of the
Vox HUMANA depend to a considerable extent on the position it
occupies, and the manner in which it is treated in the Organ; and its
imitative quality is also greatly affected by the acoustical properties
ot the building in which the Organ is placed. Of all the stops in
the Organ, the Vox HUMANA is the one to which distance lends the
greatest charm. Speaking of this stop, Max Allihn remarks : " The
Vox HUMANA is intended to imitate the human voice, which will
only be possible when the stop occupies a distant and covered place
within a large instrument in a large room/' Alluding to the forma-
tion of the stop, he adds: "The blocks and tongues correspond to
those proper for an 8 ft. stop of large scale and delicate intonation,
but the bodies of the Vox HUMANA pipes are of quite different pro-
portions and form. Definite measurements or forms do not exist.
Every organ-builder follows his own predilection and experience.
Some use short cylindrical bodies, closed at top with the exception
THEIR ARTISTIC REGISTRATION 287
of a small opening; others make larger bodies with larger openings;
while others prefer conical bodies closed with the exception of a
small lateral opening. All treatments are alike in that the bodies
are more or less covered."
TONE AND REGISTRATION. There can be no question that the
only desirable tone for the Vox HUMANA is that which, when com-
bined with the tremolo, is as satisfactory as possible in its imitation
of a human voice; and, when used without the tremolo, is perfectly
suited for combination and artistic registration. That such a stop
can be produced we have had ample proof. We inserted in our own
Chamber Organ a VOCE UMANA, 8 FT., made by a most skillful and
artistic voicer E. Franklin Lloyd, of Liverpool which was emi-
nently satisfactory under both conditions. Mr. Clarence Eddy, the
distinguished organist, in a written criticism of our Organ, says:
"The reeds are exceptionally fine; and it would be difficult to find so
satisfactory a Vox HUMANA; while its accessory, the Tremolo, is
absolutely perfect." Mr. Eddy judged the stop while seated at the
claviers, not ten feet distant from its pipes and the TREMOLANT.
Apart from the TREMOLANT, the stop proved in combination as
usef til as any of the other four lingual stops Mr. Eddy has alluded to ;
while as a timbre-creator and color-giver it occupied the first place
in registration. Played, alone, in full chords its tones were rich and
beautiful. Such, in our opinion, the artistically voiced Vox HU-
MANA should be in every Organ.
The Vox HUMANA, of the tonality alluded to, enters into effec-
tive combination with all the softer-voiced labial and lingual stops
of unison pitch; giving a special coloring to the tones of stops more
assertive than itself, and intensity and fullness to the tones of stops
of its own value. It combines perfectly with such lingual stops as
the OBOE, CLARINET, COR ANGLAIS, and FAGOTTO, warming and
enriching their voices; with string-toned stops it produces refined
and effective -compound tones rich in harmonics; and with open,
half -covered, and covered flute-toned stops it produces a family of
compound tones of beautiful colorings. Considering its value in
artistic registration, the Vox HUMANA should find a place in all
Organs of any pretensions. Its position on the claviers may vary;
but it is imperative that it be placed in an expressive division. In
the properly stop-apportioned Concert-room Organ, it should cer-
tainly find a place in the wood-wind division, and also in one of the
soft-toned accompanimental divisions.*
* See "The Organ of the Twentieth Century, " pages 307, 329, 334, 483, 505, and 506.
288 ORGAN-STOPS
VOX MYSTICA. The name given to a lingual stop, of 8 ft.
pitch, the pipes of which have cylindrical resonators surmounted by
a bell which is slotted. The tone of the stop is a modification of
that of a Vox HUMANA. The stop exists in the Echo of the Concert-
room Organ in the Colston Hall, Bristol, constructed by Norman &
Beard in 1905.
VOX RETUSA, Lat. Literally Dull Voice. The name given
to a labial covered stop of subdued tone which exists in the Organ in
the Cathedral of Lund, Sweden.*
VOX VINOLATA, Lat. A very remarkable name given to a
labial stop, of 8 ft. pitch, the pipes of which are of metal and conical
in form, resembling the SPITZFLOTE. The apparently unique ex-
ample of the stop exists in the Organ in the Cathedral of Lund,
Sweden, f
w
WALDFLOTE, PELDPLOTE, Get. Lat., TIBIA SYLVESTRIS.
Ft., FLUTE DES Bois. Dtch., WOUDFLXJIT. Literally Forest Flute.
An open labial stop, of 8 ft., 4 ft., and, rarely, of 2 ft., and i ft. pitch,
the pipes of which are of large scale, and of either wood or metal,
preferably of the former material. The WALDFLOTE (under its
mixed name WALD FLUTE) appears in numerous English Organs, of
8 ft. or 4 ft. pitch, but very rarely of both pitches in the same Organ.
An instance, however, obtains of the insertion of both stops, in the
Great of the Organ in the Church of St. Matthew, Northampton.
As a rule, English organ-builders prefer the stop of 4 ft. pitch, com-
monly inserting it in the Choir Organ. In the instruments made by
Walker & Sons, of London, the WALDFLOTE, 8 FT., is almost invari-
ably inserted in their Great divisions. In several important
German Organs, the stop of 2 ft. pitch is to be found, as on the
Second Manual of the Organ in the Cathedral of Riga, and on the
First Manual of the Organ in the Cathedral of Ulm. We have not
been able to find an instance of the insertion of the stop of i ft. pitch,
although Seidel and others say that it is made of that high pitch.
* * * Vox RETUSA., 8' von Zinn ist em FlStenregister, welches im obersten Manual der Domor-
gel zu Lund in Schweden vornanden ist. Retusa ist eine veraltete Benennung welche eine ge-
dampfte Stimme anzeigt. Em ahnlicher, aber richtiger Ausdruck ist das schon erwahnte Ob-
tusa. " Seidel.
t " Vox VINOLATA, 8' von Metall, spitz auf warts, von enger Menstir und schwacher Intona-
tion, ist ein Plotenwerk im 3 (Qber-) Manual der Domorgel zu Lund. Wie der Ausdruck vi-
nolata zu einem Orgelregister passt, durfte schwer zu entratseln sein." Seidel.
THEIR ARTISTIC REGISTRATION 289
SCALE AND FORMATION. The WALDFLOTE is not only made of different pitches
but also of different forms and materials, and, necessarily, of different scales.
Both German and French authorities describe it of large scale.* In German and
French Organs the stop is to be found either of metal or wood; and, when of- the
former, of either a cylindrical or conical form. An example of the latter form exists
on the Third Manual of the Schulze Organ in the Marienkirche, Lubeck. It is of
tin and of 2 ft. pitch. In English Organs, the WALDFLOTE, of 8 ft. or 4 ft. pitch, is
invariably of wood and of the usual quadrangular form, deeper than wide. It is
understood to have been first made in England by William Hill, of London, in
1841, and since then has been commonly made by all English organ-builders, its
tonal value having been fully recognized. In the 8 ft. stop, the bass and tenor
octaves were commonly of covered pipes; but in later and better examples the
covered pipes are confined to the bass octave. In the 4 ft. stop no covered pipes
should be introduced. The mouth of the English WALDFLOTE pipe is inverted,
placed on the narrow side, and usually cut up about one-third its width in height.
In some examples the upper lip is cut thin, while in others it is somewhat thick,
and carefully rounded and burnished, preferably with plumbago. In the best
examples the block is depressed below the lower lip, the distance being equal to
one- third the width of the mouth; this distance varies and influences the tonality
to some extent. The front of the pipe, in which the inverted mouth is cut, should
be of either close-grained mahogany or maple. The pipe is tuned by a metal shade
at top, but it should be so accurately cut to length as to require fine tuning only by
means of the shade. The entire stop must be very carefully regulated.
TONE AND REGISTRATION. As the WALDFLOTE is made of differ-
ent forms and materials, it naturally follows that there are consider-
able differences in its tonality. These differences are marked be-
tween the stops of German and English formation and voicing.
Seidel describes the tone as nothing peculiar, being rather broad,
woody, and hollow. This is not descriptive of the voice of the proper
English WALDFLOTE, which is of unimitative flute-tone, peculiarly
sweet, and inclining, in some fine examples, to a horn-like coloring.
It is this compound tonality which renders the stop so valuable in
combination and artistic registration. When artistically voiced, the
stop, in both its 8 ft. and 4 ft. pitches, is valuable in solo passages
if they are not too lengthy. In combination with the softer-toned
lingual and the string-toned stops it produces beautiful and varied
colorings of great artistic value and expression. Such being the case,
it would seem very desirable to insert the stop in some soft and ex-
pressive division of the Organ: yet it is strange to find it inserted,
* " WALDFLQTE, WALDPFEIFE, Tibia sylvestris, ist em offenes weit xnensuriertes FlStenwerk
von Zinn, Metall, zuweilen auch. von Holz, 8, 4, 2 und i'. Die Intonation dieser Sttmme is nicht
sonderlich, denn sie ist breit-,holzern-, und hohl-klmgend. " Seidel.
"La FU&TE DES Bois ( Waldflcete),&z grosse taille, en stoffe, quelquefois meme en bois, sefait
depuishuit-pieds jusqu'a un pied. Elle ne se distingue guere de la Hohlflcete ni de IzHolzflcete,
dontellea-tour a tour et pirfois tout ensemble la double caractere. C'est peut-etre parce
qu'elle n'a rien d'assez tranche qu'elle devient chaque jour plus rare." Regnier.
290 ORGAN-STOPS
apparently without exception, in the unexpressive Greats of Walker
& Sons' important Organs, including that in York Minster, recon-
structed in 1903. In the Concert-room Organ, WALDFLOTES, of 8 ft.
and 4 ft. pitch, would prove of great value in the expressive division
chiefly devoted to the stops representing the brass-wind forces of the
grand orchestra. In such a situation they would have to be of large
scale and full intonation, voiced on wind of not less than 6 inches'
pressure. To the voice of the HORN that of the WALDFLOTE will
impart richness, smoothness, and an increase of orchestral character :
while, in registration with the other necessary labial stops of the
division, it will create numerous valuable compound tonalities
and colorings with or without the impressive lingual stops properly
apportioned to the division.
WALDHORN, Ger. The name given to a lingual stop, the
voice of which is intended to imitate in its tonality that of the old
Hunting Horn. It seems to have been made of 8 ft., 4 ft., and 2 ft.
pitch;* but it appears that the imitative quality was confined to the
8 ft. stop or only the lower portion of its compass. On the Third
Manual of the Organ in the Cathedral of Lund, Sweden, we find the
stop labeled WALDHORN, BASS CLARINETTE, DISCANT, 8 FT. This
would clearly indicate that in this stop the WALDHORM portion is of a
Bass Clarinet tonality: accordingly, the voice of the stop can have
no relation to that of the Hunting Horn or Cor de Chasse. What-
ever the tonal value of the Lund Organ stop may be; it may be safely
said that a strictly imitative WALDHORN would not be desirable in
the Organ of to-day.
WALDQUINTE, Ger. This stop is mentioned by Seidel, Reg-
nier, and Schlimbach. It has been made of 5% ft., 2^ ft., and i J^
ft. pitch; but Seidel says that in Organs constructed in his time it
was very seldom inserted. This would appear to have been the case,
for we have failed to find a single record of the insertion of a WALD-
QUINTE in any Organ. Wolfram (1815) and Schlimbach describe
the stop as similar in form and tonality to the WALDFLOTE.
WEIDENPLOTE, WEIDENPFEIPE, Ger. Literally Willow
*" WALDHORN, Cornetto di Caccia, Corau par force, C. sylvestre, Cors de chasse, sind Be-
nennungen ernes seltenen Rohrwerkes, welches zu 8, 4 und 2' im Pedal und Manual vorkommt
und den Ton des gleichnamigen Blasinstruments nacnahmen soil. Bis jetzt durfte diese Stimme
wohl noch keinem Orgelbauer gelungen sein. Musikdirektor Wilke ist der Meinung, dass der
Charakter des Waldhorns eher durch Labialpfeifen als durch Zungenstimmen ZM erzielen sei
und zwar durch die r in der Karrkirche zu Neu-Ruppin stehende Stimme FLUTTUAN, deren
Klang dem Waldhornton sehr nahe kommt, " Seidel-Kothe.
THEIR ARTISTIC REGISTRATION 291
Flute. The name that has been used to designate a metal labial
stop of extremely small scale and soft intonation. Its name would
imply a flute-tone, which was probably somewhat indeterminate in
its character. Such a stop would be suitable for a true Chamber
Organ, or for insertion in an Echo or Ancillary Aerial Organ.
WIENERFLOTE, Ger. Literally Vienna Flute. Locher says
this stop "is one of the most charming wood flutes, intonated rather
brighter than the PLAUTO DOLCE. As a rule, it occurs on one of the
upper manuals as an 8 ft. or 4 ft. solo stop, more particularly in Swiss
Organs. . . . The denomination 'WIENERFLOTE' lacks all ety-
mological or historical foundation. In the new [1888] Votiv Organ,
although this stands in Vienna itself, there is not a single WIENER-
FLOTE amongst its sixty-one speaking stops." On the Third Manual
of the Walcker Organ in the Cathedral of Vienna there is a WIENER-
FLOTE, 8 FT. Another example exists on the Third Manual of the
Organ in the Cathedral of Riga.
Carl Locher, Chief Organist of the Catholic Church, at Berne,
who had favorable opportunities of judging the tonal value of the
stop, remarks : * ' WIENERFLOTE is one of the most useful stops on the
upper manuals, not only as a solo, but also for combination with
any other stop. I found it particularly beautiful in combination
with the OBOE and FLAUTO TRA VERSO." It has been generally
understood that the tone of the WIENERFLOTE closely resembled the
imitative tone of the FLAUTO TRA VERSO; but Locher's example of
effective combination would seem to indicate a different tonality;
unless the combination of the labial stops merely increased the imi-
tative flute-tone.
The pipes of the WIENERFLOTE are open, quadrangular, and have
inverted circular or semicircular mouths, partly over or against
which are adjusted sloping caps. Such a formation would point to
the production of an imitative quality of flute-tone, but not so pro-
nounced as that of the harmonic FLAUTO TRA VERSO, nor so valu-
able in solo effects or registration.
x
XYLOPHONE. A percussion musical instrument recently in-
troduced as a stop in certain Organs. Pine examples are made in
this country by the Kohler-Liebich Co., and by J. C. Deagan, both
of Chicago. The XYLOPHONE, made by the former firm, is con-
structed of four octaves, chromatic, of Rosewood bars, graduated
292 ORGAN-STOPS
in size, and adjusted over properly tuned cylindrical metal resona-
tors. The percussion action is electrically operated from the clavier
of any Organ in which the stop is placed. Stops of this class are
not suited for the dignified Church Organ,
ZARTFLOTE, Ger. The stop bearing this name was, according
to Seidel, invented by the organ-builder Friedr. Turley, who first
called it a GAMBA; but as its tone was of a soft and refined fluty
quality rather than of a string character," Musikdirektor Wilke ad-
vised its inventor to adopt the more expressive name ZARTFLOTE.
The stop is formed of small-scaled open pipes, usually of wood,
voiced to yield an extremely tender flute-tone; hence its name. It
has been made of both 8 ft. and 4 ft. pitch. As a stop of 8 ft. pitch
it is inserted in the Swell of the Organ in the Marienkirche, Wismar;
and as a wood stop, of 4 ft. pitch, it is to be found in the Echo of
the Schulze Organ in the Marienkirche, Lubeck. In the Echo of the
fine Organ in the Church of St. Bartholomew, Annjey, Yorkshire,
built by Edmund Schulze, there is a ZARTFLOTE, 4 FT., formed of
conical metal pipes from tenor C to c 4 . The bass octave is of small-
scaled wood pipes. This stop speaks on wind of i J^ inches, and has
a voice of an extremely soft and refined flute-tone. We do not find
the ZARTFLOTE in any of Walcker's important Organs. A ZART-
FLOTE, 8 FT., the pipes of which are of pine and pear-tree, exists on
the Fourth Manual of the Ladegast Organ in the Cathedral of
Schwerin. The stop is to be found in some English Organs; usually
of 4 ft. pitch, and inserted in the Choir or Swell.
The name ZARTFLOTE has been used by John W. Whitely to
designate a beautiful stop invented by him in 1896. This stop, of
8 ft. and 4 ft. pitch, is formed of covered metal pipes with project-
ing ears carrying cylindrical aluminium harmonic-bridges. The
stop is a modification of the QUINTATEN, with every trace of coarse-
ness removed.* Its voice is a light, bright, flute-tone, with sufficient
reedy quality to impart to it a distinctive tonal coloring. A stop of
this beautiful tonality would be of the greatest value in the soft
divisions of the Organ; contributing largely to refined and artistic
registration.
ZARTGEDECKT, Ger. The name that has been used to desig-
nate a covered stop of wood or metal, of 8 ft. pitch, yielding an unimi-
* Full particulars, with illustration, of this stop are given in " The Art of Organ-Building "
Vol.II., p. 549 ; and in "The Organ of the Twentieth Century, ** pages 391-2.
THEIR ARTISTIC REGISTRATION 293
tative flute-tone softer and sweeter than that of the LIEBLICH-
GEDECKT, 8 FT. The stop requires, for the production of its charac-
teristic voice free of any pronounced harmonic to be voiced on
wind of very low pressure, preferably i J/ inches. This stop is in all
essentials similar to the STILLGEDECKT (g. v.}.
ZAUBERFLOTE, Ger. Literally Magic Flute. A covered
harmonic stop, invented by William Thynne, of London, and first
introduced by him in the Organ installed in the Inventions Exhibi-
tion, at South Kensington, London; and which was afterwards
erected in Tewkesbury Abbey. It is of 4 ft. pitch, and is inserted in
the Choir of the Organ. In the ZAUBERFLOTE, 4 FT., in the Organ
in the Church of St. John, Richmond, Surrey, the harmonic pipes
commence at tenor C; and, as they are covered, they speak their
first harmonic, or the second upper partial tone the Twelfth. To
prevent its speaking the prime tone, or that proper to its length,
each pipe is pierced with a small hole, as in the case of the open pipes
of the FLUTE HARMONIQUE. The ZAUBERFLOTE was made of both
8 ft. and 4 ft. pitch. When of the former pitch, the bass octave was
formed of covered wood pipes, voiced to carry down the characteris-
tic tone of the harmonic portion of the stop as closely as possible,
but by no means satisfactorily, as can be supposed.*
TONE AND REGISTRATION. The tone of the ZAUBERFLOTE, as
voiced by its inventor, is sui generis and of a very refined and sym-
pathetic character, highly appreciated by those endowed with ears
sensitive to tonal values. In artistic registration it is of great value;
and as an OCTAVE, it lends itself to the production of a series of
effective tonal colorings in combination with the softer string-toned
labial and reed-toned lingual stops; imparting a distinctive harmonic
structure to all dual combinations in which it is introduced, and to
which it gives a special brilliancy and vivid coloring. The stop is,
accordingly, of the greatest value in its 4 ft. pitch; and this fact its
inventor realized, as we know. He was for several years our valued
friend.
ZINK, ZINKEN, Ger. A lingual stop made by the old German
organ-builders, and intended to imitate the tone of the obsolete
wind instrument known as the Zinken, or by the Italian name Cor-
netto Curvo and the French name Cornet-a-Bouquin. Describing
* Pull particulars, with illustration, of the formation of this stop are given in "The Art of
Organ-Building," Vol. II., pp. 546-7; and in "The Organ of the Twentieth Century /' pages
389-90.
294 ORGAN-STOPS
the instrument, Carl Engel says: "Although the Zinken is blown
through a mouth-tube somewhat similar to that of a trumpet, it has
finger-holes like a flute. Its sound is harsh, and would be unpleasant
in a room, but the Zinken was intended for the open air, and for
performing chorales on the towers of churches, so that all the people
in the town could hear the solemn music. They were, in fact, com-
pelled to hear, for it vibrated through the air over their heads like
the church bells themselves. Thus the Zinken may have served its
purpose well in olden time, notwithstanding its harshness of sound."
The organ ZINKEN was, apparently, a lingual stop, of 2 ft. pitch,
having a strident voice, and finding its place in the Pedal Organ.
Under the Dutch term, CINQ, 2 FT., stops of the class exist in the
Pedals of the Organs in the Cathedral of St. Bavon, Haarlem, the
Cathedral of St. Lawrence, Rotterdam, and other important
Churches in Holland. In the Pedal of the Organ in the principal
Protestant Church in Utrecht, built in 1826, there is, in addition tc
the CINQ, 2 FT., a CLARION, i FT.
A A
651