THE DOUBLE REED

59

The Physical Properties and the Anatomy

of Arundo Donax: Reed Cane
Dr. Kirsten M. Boldt-Neurohr
Ellensburg, Washington

region is the subtropical countries found along the

Mediterranean Sea. From this area it has spread
widely to most other sub-tropical regions. The giant reed transplants well and now flourishes in some
unsuspected areas. For instance, although Arundo d.
grows well along warm coastal elevations, it has been
seen growing in the Himalayans at an elevation of
8,000 feet.5 Now widespread, Arundo d. has been successfully cultivated in France, China, Southern Africa, Australia, Brazil, Italy, Argentina and the United
States. In the US, it grows as far north as Washington,
D.C. and Portland, Oregon, although no known useable cane for reeds has been grown in the D.C. area.6
It predominantly grows in California, Texas and the
southwestern states.
Arundo d. habitually shows a tall stalk, or culm,
and grows to the height of between eighteen and
thirty feet with nodal divisions, roughly fifteen to
twenty-two nodes per stalk. A node is recognizable
as a solid thickening of the cane wall, creating an alternating pattern with the hollow internodes (node,
internode, node, etc.) for the length of the stalk. Each
internode ranges from five to fifteen inches in length.
The diameter of the culms ranges from seven millimeters to thirty-five millimeters; the optimal diameter for bassoon cane is twenty-four to twenty-seven
millimeters. (see Illustration 1 and 2 following page)
The hollow culm has an outer wall that ranges from
two to seven millimeters thick.
Young stalks come up in early spring and are
green in color. The diameter of the individual stalk
does not alter once sprouted. A minute primary
thickening of the walls does occur. (Primary thickening is a hardening produced by new layers of cells
grown in addition to the first year cells.) However,
the additional vertical layers of the primary thickening add little or no actual circumference to the
culms.7 The culm grows very fast, in the first month
of growth approximately two feet a week.
In the first weeks of growth a cane crop can look
much like a corn patch. The wide leaves, similar in
appearance to corn leaves, grow from each node (See
Ill. 3 and 4 following page). A husk-like fiber also
grows up from each node, wrapping tightly around

ARTICLES

s a cane enthusiast I have taken my obsession to the next level, I have become a grower of reed cane. This only fueled my enthusiasm and desire to go even further and research the
origins, the physiology and the detailed anatomy of
Arundo donax. Basically, I want to understand the
inner workings of this organic matter upon which we
double reeds base our success, happiness and sanity.
This is a compilation of my research and that of many
of my esteemed cane enthusiast predecessors; I hope
you find it as interesting as I do.
The Giant reed, Arundo donax L.1 Gramineae, is
a member of the large grass family that is indigenous
to the countries that border the Mediterranean Sea.
Cane is often mistakenly synonymous with bamboo
but they are only distantly related in the Grass family,
which in itself holds thousands of species.2 Cane has
been used in the making of musical instruments and
has provided the material for the vibrating reed for
at least 5,000 years.3 Despite the ready availability of
relatively good Arundo d., musicians still find it difficult to obtain exactly the right cane for the reeds
they desire. Each individual musician has particular
preferences for cane resistance, hardness and flexibility that are particular to their instrument and playing
style.
Each harvest yields a low percentage of useable
cane, according to documents from the Rico cane
industry; an acre of land will support approximately
30,000 poles of Arundo d., of which roughly 25% will
be the correct diameter and age for use for reeds.
After the curing and selection process only about
1,000 pounds of tubes from the 7,500 poles will be
distributed for sale, often much less. The discarded
cane, unsuitable for musicians use, is used to make
fishing poles, broom sticks, weaved mats, fences and
furniture.4
Arundo donax L. is a fibrous perennial plant, a
member of the grass family: Genus Arundo, Family
Gramineae, Tribe Festuceae. This tribe has six different species, including Arundo d. Some common
names that are found in reference to Arundo d. are:
Giant Reed, cane, reed cane and Arundo. Arundo
d. has a connective, familial root system. Its native

60

THE PHYSICAL PROPERTIES AND THE ANATOMY OF ARUNDO DONAX: REED CANE

Ill. 2. Node-internode, of both 1st and 2nd year culms

(2nd yr, on far right)

Ill. 3. Young Arundo d.

Ill. 4. First year Arundo d.

ARTICLES

Ill. 1. Measuring the internode of rst year Arundo d.

Back to Table of Contents

THE DOUBLE REED

Ill. 5. Arundo d. Rhizome

the stalk. In the first year of growth this husk is bright

green but it eventually dies and becomes brown and
papery. Many have thought that the decay of the husk
was the cause of the brown spotting sometimes found
on the epidermis of the cured cane. Donald Cassadontes 1995 research has since proven that this is not
the case. The discolored spotting on the surface is actually the result of an increased concentration of tannin, a fatty acid substance, stored as an energy reserve
for plant metabolism, found on the exterior of many
plants.8 According to Cassadonte the rotting of the
leaves could not be the cause of the spotting because
tannins are also present under the epidermis surface.

This manifests as thin dark lines

in the pith of the cane, or, in rare
instances, the lines appear as deep
as the third layer of parenchyma.
The presence of this tannin is not
indicative of good cane nor does it
affect the cane, contrary to popular belief.9
As the plant matures, the culm
becomes yellow-green in color. In
the second season of growth, small
secondary branches about five
millimeters in diameter emerge
from many of the nodes and also
leaf out. Some mature plants are
topped by large terminal plumelike panicles, thirty to sixty centimeters long.10 In twelve years
of growth, my plants have never
flowered. Although the plumes
are often thought to produce seeds
this is a myth, because Arundo d.
can only grow from a rhizome or a
rooted stalk.11
The root system in Arundo d.
is a large fleshy bulb, called a rhizome, which lies horizontally an
inch to a foot below the surface
of the soil (See Ill. 5 and 6). From
this rhizome thin, tough, fibrous
roots, called adventitious roots
penetrate deep into the soil.12 By
reaching the deepest water table
these adventitious roots allow the
plant to access continuous water
even in times of drought.13 It is
from this rhizome that multiple
stalks sprout vertically, while the
rhizome continuously and simultaneously grows horizontally under the soil, sending
up new shoots in a growth pattern similar to the connective root system of common grass. A hardy root,
the rhizome can withstand multiple abuses. Even
after being cut into many pieces and transplanted, it
continues to prosper.
It is because of this determination of growth from
the rhizome that the California Invasive Plant council has labeled it as an invasive species (comprising
over 68% of the riparian vegetation in the Santa Ana
River), thus launching an active movement to eradicate Arundo d. from the stream banks of its state.14

ARTICLES

Ill. 6. Arundo d. Rhizome

61

62

THE PHYSICAL PROPERTIES AND THE ANATOMY OF ARUNDO DONAX: REED CANE

ANATOMY OF ARUNDO DONAX

It is the anatomy of Arundo d. that has made it an
unsurpassable material for musical reeds. No other
material, man-made or organic, has been found that
so perfectly balances the properties needed for musical reeds.

ARTICLES

According to some experimental work done

by Benade together with Walter Worman and
Daniel Wright, the particular virtue of Arundo
Donax for reed making lies in its ratio of longitudinal stiness to density, together with great
transverse exibility at appropriate thickness.15
Cane provides the desired stiffness in the throat or
tube of the bassoon reed, which is further supported
by the hardened epidermis of the cane. Additionally, its unique flexibility allows it to vibrate freely
as needed in the flatter blades of a reed. Cassadonte
explains that unlike most woods where there is a single dominant plane of strength; Arundo d. has three
dominant planes of equal strength.16 Of all the musical reeds, the bassoon reed puts possibly the most
stress on the cane. The process begins by taking a
half moon curved piece of twelve-inch cane which
is gouged down to a preferred thickness. One half of
the piece is manipulated into a round tube and the
other half scraped into an almost flat, freely vibrating
blade, of a greater degree of flatness than the original curve of the culm.17 It is because of these requirements that Arundo d. is perfectly suited to the task.18
The equal planes of flexibility allow the vibration to
be predictable and even, whereas another wood material with only one plane of flexibility would result
in chaotic and unstable vibrations, if any vibrations
would even occur.19
Another attribute of Arundo d. that contributes
to superior reeds is the lignin content found in cane.
Lignin, found in all woody plants, resists excessive
absorption of water in the cells which would otherwise cause decay.
The major chemical component (20% to 40%)
of Arundo Donax is lignin. This highly complex
gummy ller substance is formed from dead
plant cells and usually occurs as a result of the
aging process of the plant. Lignin protects older
plants from pathogens and water, stabilizes the
cell wall matrix and gives viscous shock protection and some plasticity to the plant.The
presence of rather amorphous lignin in relatively

Back to Table of Contents

exible cell wall structure of Arundo may be responsible for changing the plant stem from one
in which there is only one naturally dominant
plane of material strength into one in which
each x,y,z planes are of equal strength.20
Since lignin increases as the plant matures, it is more
likely to be found in the older sections of the culm.
Lignin is hydrophobic, or water repelling. The hydrophobic qualities of lignin and the high proportion of
it found in Arundo d. are among of the factors that
make cane undisputed as the optimal material for
woodwind reeds. Lignin acts as a deterrent to moisture absorption and reed material with low amounts
of it tend to become saturated after short periods of
use.21 Because of the way in which musical reeds are
used, they are continuously immersed in water and
then allowed to dry or partially dry before immersion again. This constant wet-dry treatment would
break down most natural substances quickly, whereas Arundo d. resists decay because of the high lignin
content.22
Anatomically, Arundo d. is made of three layers. The hard waxy epidermis, referred to as bark by
reed makers, is what gives stability to the tube of a
bassoon reed. The second layer is a thick sclerified,
or hardened, fiber band, which is the dense supporting tissue of the stalk. The third layer is a thick inner
cortex comprised of the parenchyma or the soft thin
walled inner cells of the cane, which many call the
pith of the cane.23 The two inner layers hold the vascular bundles, which are bundles of vessels that carry
the food and water through the stalk of the cane.24
Vascular bundles are distributed freely throughout
the parenchyma.25
As stated by Peter Kolesik in his 1998 Annals of
Botany article, Arundo d. is comprised of three concentric circles: 1) the hard waxy epidermis (bark) and
outer cortical cells, 2) the thick sclerified fiber band,
and the 3) thick inner cortex, or pith.26 Alternatively,
Cassadonte loosely compares the three layersthe
dermal, the ground (supportive tissue) and the vascularto skin, muscle and bone.27

EPIDERMIS
The epidermis is an almost impenetrable single hard
layer covered by a waxy cuticle. This dermal tissue is
initially green in color but turns golden in maturity.
The epidermis is rigid because of sub-epidermal sclerenchyma cells, which are initially water conducting
cells in the tall tubes. As the stalk grows the tubes

THE DOUBLE REED

become more load bearing. Once the stalk ceases

elongation the cells die, developing harder thicker
walls and smaller pit cells.28 It is because of these
small strong cells that Arundo d. can grow to such
unusually tall heights for a grass stem. Heinrich gives
a visual explanation that the relationship of sclerenchyma and parenchyma cells is similar to reinforced
concrete. The parenchyma is analogous to the concrete, and the sclerenchyma is analogous to the metal
reinforcement.29 Cassadonte clarified that sclerenchyma is a generic term for thick-walled supportive
cells of a plant stem. The long slender cells are often
refers to as fibers.30

SUPPORTIVE TISSUE OR FIBER BAND

INNER CORTEX
Along with the fiber band of the inner layer are vessels which supply food and water from the roots to
the stem. The inner cortex comprises the bulk of the
stem tissue, made up of a mix of vascular tissue and
parenchyma cells. Vascular bundles are distributed
freely throughout the cross section of parenchyma.
They are collateral, surrounded by one or more layers
of thick walled strongly lignified fibers.34 Lignin is
highly present in the parenchyma cells of second year
growth mature culms.35 The cell walls of parenchyma
are essentially elastic giving a reed its springiness.
The vascular tissue is somewhere between a viscous
and elastic tissue.36 The inner cortex also contains the

compound hemicellulose, found in most woods in

general to aid in the absorption of water.37
Multiple studies by Kolesik, 38 Veselack,39 JeanMarie Heinrich,40 and a scientific team lead by Dr.
H.C. Spatz41 have found that the favorable characteristics which lead to the performance qualities of
good reeds are all found in the inner cortex: all
characteristics are attributed to the same anatomical
feature, the vascular bundles.42 As Veselack states,
Good reeds are found to contain well-developed
cells and tissues associated with the vascular bundles.43 The vascular tissue is vital; it is the sheathing of the vascular bundles that at a certain maturity,
contains a fiber that gives the stiffness and allows vibration to occur on all planes.44
This is an amazing plant, which it seems Mother
Nature specially suited for the purpose of a double
reed. Through a vast amount of research from scientists and musicians alike we have ascertained physiologically why Arundo d. is unmatched in its ideal
qualities for reeds. However, I still have the question
of why some cane is undeniably better than other
cane? More so, why is some cane deemed absolutely
marvelous for one bassoonist and turns out horrific
reeds for another? Naturally, we all have different
instrument set-ups, embouchures and physical attributes that alter the way we play from the person next
to us. But as I have gained more knowledge on this
great plant, I am sure that more goes into the differences found in cane. This has led me to the research I
am doing on the affects of climate during the cultivation of Arundo d. How does the climate during cultivation ultimately affect the resulting reed? Like fine
wine, would knowledge of the climate and growing
conditions of a certain cane harvest help in obtaining cane perfectly suited for an individuals taste? My
dissertation research has produced some interesting
results, of which I promise to share with you, my fellow cane enthusiasts.

ENDNOTES
1

L. is the accepted abbreviation for the father

of binomial nomenclature, Carl Linnaeus, an
18th c. Swedish scientist and botanist. A proper
botanical name always includes the author or
the authority of the species. (Dr. Frank Lang,
personal communication, June 30, 2009).
Philip Bate, The Oboe: An Outline of its History,
Development and Construction 3rd ed. (New
York: W.W. Norton inc.,1956/1975), 20.

ARTICLES

The ground or supporting sclerified tissue is composed of long chains of lignified parenchyma cells.
This is often recognized when looking at the end
grain of a gouged piece of cane; it is seen as a thin
brown reddish band below the epidermis, made more
visible when the cane is wet. Parenchyma are thin
walled cells of the ground tissue which make up the
bulk of the plant. Marilyn Vasleck calls these sclerenchyma cells a fiber band in her 1981 article in The
Clarinet. The fiber band also includes very small vascular bundles, which are more numerous in the fiber
band towards the periphery of the stem than the vascular bundles found the inner layers.31 Vascular bundles are found in both the fiber band and the inner
cortex similar to the pattern found in corn and other
members of the grass family. They are what woodwind musicians often refer to as the grain or fiber of
the cane.32 Veselack reports that in a young threemonth culm there is no fiber band present, proving
that the ring develops with maturity of the plant. 33

63

64
3

4
5
6
7

8
9
10
11
12
13
14

ARTICLES

15

16
17

18
19
20
21

22
23

24
25
26
27
28
29

30
31

THE PHYSICAL PROPERTIES AND THE ANATOMY OF ARUNDO DONAX: REED CANE

Robert E. Perdue, Jr., Arundo donax-Source of

Musical Reeds and Industrial Cellulose, Economic Botany, vol. 12, No. 4 (October-December,
1958), 376. and Rico Reeds, Origin of the Cane
Reed (Los Angeles, 1982), 2.
Rico, Origin of the Cane Reed, 9
Perdue, Arundo donax Source, 368.
Ibid.
Donald Jay Cassadonte The Clarinet Reed:
an Introduction to its biology, chemistry and
physics (DMA diss., The Ohio State University,
1995), 22.
Cassadonte, The Clarinet Reed, 71, 84.
Ibid, 71.
Ibid.
Rico, Origin of the Cane Reed, 4.
Perdue, Arundo donax Source, 352.
Cassadonte, The Clarinet Reed, 52.
California Invasive plant Council [cited Sept. 21,
2010] Available from http://www.cal-ipc.org/ip/
management
Philip Bate, The Oboe: An Outline of its History,
Development and Construction 3rd ed. (New
York: W.W. Norton inc.,1956/1975), 21.
Cassadonte, The Clarinet Reed, 181.
James B. Kopp, Counting the Virtues of Bassoon Reed Cane, International Double Reed
Journal vol. 26, no. 4 (2004): 50.
Ibid
Ibid
Cassadonte, The Clarinet Reed, 181.
Marilyn S. Veselack and Jerry J. Nisbet, Cell
and Tissue Differences in good and usable
woodwind reeds from Arundo donax L. The
Clarinet Vol. 8, no 4. (1981): 10.
Cassadonte, The Clarinet Reed, 175, 181.
Peter Kolesik, A. Mills and M Sedgley, Anatomical Characteristics Affecting the Musical
Performance of Clarinet Reeds Made from
Arundo donax L.(Gramineae), Annals of Botany
81 (1998):153.
Veselack, Cell and tissue Differences, 9.
Perdue, Arundo donax Source, 372.
Kolesik, Anatomical Characteristics, 153.
Cassadonte, The Clarinet Reed, 82.
Ibid, 89.
J.M. Heinrich, The Bassoon Reed: An Analysis
of its Construction Aesthetic Machanical and
Botanical Aspects, Journal of the Double Reed 7
(1979): 24.
Cassadonte, The Clarinet Reed, 90.
Perdue, Arundo donax Source, 372.

Back to Table of Contents

32
33
34
35
36
37
38
39
40
41

42
43
44

Veslack, Cell and Tissue Differences, 9.

Ibid
Perdue, Arundo donax Source, 372.
Veselack, Cell and tissue Differences, 9.
Cassadonte, The Clarinet Reed, 106.
Ibid 163. and Kopp, Virtues of Bassoon Reed
Cane, 8.
Kolesik, Anatomical Characteristics, 153.
Veselack, Cell and Tissue Differences, 9.
Heinrich, The Bassoon Reed, 25
C.H. Spatz, H. Beismann, F. Brchert, A.
Emanns, Biomechanics of the Giant Reed
Arundo Donax, Philosophical Transactions of
the Royal Society of London 29, vol. 352, no. 1349
(January 1997) 5.
Kolesik, Anatomical Characteristics, 153.
Veselack, Cell and Tissue Differences, 12.
Kopp, Virtues of Bassoon Reed Cane, 4.

THE DOUBLE REED

27

An Overview of the Rise of Arundo donax in Oboe Reeds

Zachary Blais
Worcester, Massachusetts

A Bassoonists dream, that ONE REED, that nondescript entity which constitutes a reed with life, zeal,
a resilience present only in a rubber band, life as
long as an elephant.
A Reed that will NOT change with the very slightest
temperature fluctuations.
The very act of playing you, Reed,
is Destructive.
But, WHAT CAN I SAY?
-Michael Trentacosti,
WHAT CAN I SAY?1

Figure 1: Two photographs of a typical dried Arundo

donax cross section. [N.B. The upper image is sharpened to enhance detail.]

iant cane has existed easily as far back as

the writing of the Bible, where cane is mentioned frequently.2 The plant itself is actually very common; in fact, some areas of New Zealand
are so inundated with it that it is considered an unwanted organism.3 Giant cane is also known to have
escaped cultivation in California, where it is reported
to do damage to established species.4 Given its commonplace status, it may not be surprising that it has
been so important in the history of Western music.
(Its price, however, may be surprising, but well save
that for later.)

HISTORY OF ARUNDO DONAX:

Initially, Arundo donax grew naturally near the Mediterranean Sea.5 The plant is noteworthy for its hardness, which Veselack attributes to silica (as in sand)
in its bark; she notes that the plant will spark if struck
with an axe. Giant cane is known to grow anywhere
from 2 to 8 meters (roughly 6 to 26 feet) in height, with
diameters of 1 to 4 centimeters (about to 4 inches)
giving finished reeds their distinctive curves. A smaller diameter yields a greater curvature of the finished
cane, so smaller cane is generally used for something
like an oboe reed while larger cane might be used for
saxophone or clarinet reeds. (Figure 16)
Arundo donax was utilized in early musical instruments as far back as five thousand years ago.2 Perdue
demonstrates that man used bone to create flutes in
the Stone Age, and further suggests that a much more
satisfactory instrument could be fashioned from a
tube of vegetable material.2 These cane flutes quickly
evolved into the pan pipe, or syrinx, which is simply
a tuned set of these tubes placed in a row. While the
common name pan pipe is derived from the instruments association with the Greek god Panand so we
know that these pipes were in use in ancient Greece,
in approximately 300 BCpan pipe expert Douglas
Bishop states that syrinxes dating as far back as 4200
BC have been discovered in North America, some of
these also being made of cane.7 (see Figure 2,2 following page)
Soon, pipe crafters began to carve holes into the

28

AN OVERVIEW OF THE RISE OF ARUNDO DONAX IN OBOE REEDS

Figure 2: A modern pan pipe made from Arundo

donax cane.

side of their pipes, removing the need for several pipes

of different lengths. With the increased effort invested
in these instruments it was no longer viable to produce
a new instrument every time the mouthpiece failed,
and so the replaceable reed soon followed.2
It should be noted at this point that although
many primitive groups were able to discover the reed
mechanism, further progress to an actual instrument
was limited to only the Middle East, and specifically
to Egypt and southwestern Asia. This coincides with
the fact that, as noted earlier, Arundo donax was first
found naturally near Egypt along the Mediterranean
Sea. Perdue attributes this limitation of development
to the fact that donax cane was the ideal material for
producing reed instruments, based on its responsiveness and resilience to repeated soaking.2 Indeed, it
seems likely that the presence of donax cane was what
propelled instrument development in these regions, as
other areas with structurally similar plants (e.g. bamboo) were not able to independently develop double
reed instruments.
Unfortunately, there is very little information on
how musicians first discovered the mechanism of the
double reed. The surna, which is generally considered
to be the first European double reed instrument, is
believed to have first surfaced around the eighth century.8 It did not use donax cane; rather, the material
used was usually some sort of leaf or grass which was
much thinner. The player could not use any sort of
embouchure, as this would cause the thin material to
stop vibrating altogether, and so instead he would enclose the entire reed in his mouth. The instrument had
a metal disk on its body, called the pirouette, and the
player would place his lips on the disk to enclose the
reed.9 (Figure 3: 10)
The instrument that evolved from this, the shawm,

Back to Table of Contents

Figure 3: Two
modern surnas.
The one on the left
is Turkish, and the
one on the right is
Syrian. Note the
reeds, which are
independent of
the instrument.

Figure 4: Comparison of a standard

treble shawm
(left) and a bass
shawm (right),
drawn to scale.
The ruler to the
far left measures
feet.

came into existence during the twelfth and thirteenth

centuries.9 The shawm was different in several ways
from the surna; these differences included both a longer instrument body and a longer, narrower reed. In
addition, for the first time the reed was made from
Arundo donax, which was much harder than the
grasses and leaves used before in the surna. This, coupled with a pirouette located closer to the base of the
reed, allowed players to place their lips directly onto
the base of the reed, giving them more control over
the instruments sound; in Figure 411, the pirouette is

THE DOUBLE REED

Figure 5: A modern shawm.

Note its longer shape and
slightly longer reed.

Figure 6: Four oboe

reeds, dating from
the late 18th century
to today, all made of
Arundo donax cane.

Figure 7: An 18th-century reed from a bourdon,

an organ pipe similar in
mechanics to the oboe.
The reed itself is said
to resemble that of the
early oboe.

in the form of a wooden cup enclosing the reed.9 Baines notes that by the fifteenth century, the shawm was
produced as a bass instrument, measuring six feet in
length, as well as the more common tenor (shown in
Figure 58) and treble versions.8 In addition, a small
discant version measuring less than two feet and a
great-bass shawm measuring roughly ten feet both existed in the sixteenth century.

29

The shawm flourished for a few centuries, but was

soon replaced by a new French invention: the oboe,
first crafted by a group of Parisian instrument makers
led by Jean Hotteterre. While the transition from surna
to oboe marked some important structural changes
the use of joints instead of a one-piece instrument, the
inclusion of ornamentation to strengthen these joints,
and the (initially) uneven internal bore8probably the
most important change was the removal of pirouette
entirely, with the performer now using a full embouchure to control the reed.9 At first this embouchure
was likely more vital than ever; the earliest reeds were
about 30 millimeters longer and 3 millimeters wider
than todays average reed, similar to the reed seen in
Figure 712. 9 This new instrument is believed to have
debuted publicly in 1657, where it was extremely successful and immediately spread from France.8
The reeds used in the early oboe, as noted above,
were drastically different in size; however, the material that was used is the same Arundo donax that musicians are so familiar with today (see Figure 68). In
an article on 18th century reedmaking from 1962, and
acting on samples gathered before and after, Thomas
Warner proposes that the change from this original
larger size to the smaller modern size occurred around
1780.13 At the very latest, this change would have to
have occurred by 1783, when Warner notes that Christophe Delusse had begun to make modern-style oboes,
which would require modern-style reeds.13,8
From this point forward, the development of the
oboe is much more gradual. Ledet summarizes the instruments changes in a few sentences by saying:
Basically, over the years the oboe bore has become narrower, the reed has become controllable by the lips after being freed from the pirouette, and reeds have become narrower as the
bore has gotten smaller. More keys have been
added to facilitate technique and intonation.
Heavier wood is used for the resonating body.
Thanks to, in part, the principles of Boehm,
Helmholtz, and the advancing science of acoustics, tone holes have undergone much development in spacing, size, and undercutting, which
has helped intonation and response.9
These changes have only brought about minor changes in the construction of the reed. Reed 3 in Figure 6,
which was made circa 1870, utilizes the same style of
scrape (i.e. V- instead of U-shaped) maintained to this
day in France, with only the length of the scrape modified to compensate for changes in the instrument.8

30

AN OVERVIEW OF THE RISE OF ARUNDO DONAX IN OBOE REEDS

Even with this difference, Reeds 3 and 4 are strikingly

similar, and could plausibly be confused by a player
who has not studied them carefully.

ANALYSIS OF ARUNDO DONAX

AS A MATERIAL:
This brings us to today, an age where plastics and artificial materials reign. Having read this brief history, it
can be seen that the cane that is so widespread today is
not necessarily the best material for the job, but rather
was the most usable material available to those who
first started making instruments. The same material
has been used all alongafter all, why fix it if it isnt
broken?and for that reason it is used to this day.
However, ask any oboist today and he or she can
almost certainly name something that is broken
about the traditional oboe reed. At the end of December 2009, I conducted a survey of oboists in the Boston Musicians Association, asking for their opinions
on items such as the dimensions, material, lifespan,
required embouchure, and required soaking of a reed.
The survey also asked about the players thoughts on
artificial reeds, and asked the question, If you could
change one thing about the traditional oboe reed, what
would it be? Why?
This survey established that several aspects of the
traditional cane reed are particularly troublesome:
1. Both with pre-processed cane and completed reeds, inconsistency is a significant problem. As any double reed player
knows, slight changes in weather, humidity, and temperature can cause significant
changes in the quality of a reed; and, as
reedmakers know, cane itself is a highly
variable material because of its organic
nature. A piece of cane taken from a pile is
often just as likely to produce a good reed
as it is to produce garbage, and it can be
difficult to tell which will occur beforehand.
2. Reedmaking is a long and involved process, which makes inconsistency even
more of a problem. A reed that will last
only a handful of playing hours can take
an hour just to be made, which is an enormous time commitment for oboists who
do not perform for a living.
3. Even a good reed will be extremely fragile.
All it takes to lose the best reed ever made
is brushing it on a stand or chipping it on a
tooth. It is also usually impossible to bring

Back to Table of Contents

the reed back to a working state once this

has happened, as many players know.
4. Reeds are extremely expensive. A professional who makes his or her own reeds
gains a slight reprieve here, but students
and casual players are forced to spend a
fortune on reeds, which can be prohibitive.
All of these problems are direct results of the fact
that cane is a naturally-grown, organic product. Its
structure, as seen in Figure 814, is composed of many
plant fibers held together with lignin, which acts as
a glue of sorts. This configuration gives cane a much
higher strength axially (that is, moving with the fiber)
than equatorially (that is, moving against the fiber).

Figure 8: A vectorization of an Arundo donax cross

section, looking down the length of the bers. (A) is
the ber itself, and (B) is the weaker lignin holding each
ber together.

This difference in directional strength causes reeds

to generally fail by cracking vertically down the reed.
The internal forces holding each individual fiber together are larger than those holding separate fibers
together, and so when the cane weakens enough to fail
it does so by splitting the fibers apart. This weakening
is a result of fatigue, which is a natural process in any
part which goes through repetitive stresses.
To understand how fatigue comes about, it is necessary to know some basic information about how the
oboe reed functions mechanically. When at rest, the
oboe reed is just as it looks when held in the hand: the
blades of the reed sit nicely on each other. When air
begins to pass through the reed, air flows through the
inside of the reed but not the outside. This fact sounds
obvious, but is absolutely vital to buzzing on the reed.
Because of this difference in pressure, caused by sitting
air outside and flowing air inside, the blades of the reed
will compress together. (This phenomenon is known

31

THE DOUBLE REED

as Bernoullis Principle, and is the same principle that

allows aircraft to lift off even though they are much
heavier than the air around them.) The compressed
blades decrease the opening size of the reed, which reduces air flow and increases the inside pressure.
At this point, the reed is closed, and the edges of
the blades are acting like two compressed springs due
to the hardness of the cane. This compression, combined with the now-higher internal pressure, causes
the blades to fly open, nearly separating at their farthest. Now the wider opening increases air speed,
decreasing internal pressure, and the blades act like
stretched springs. This new tension drives the blades
back to their normal state, but because they are already
moving together they quickly become compressed
once again. While the blades are vibrating, the air
pressure coming out of the staple is constantly changing. When the blades close the lower air flow causes a
lower exiting air pressure, and when the blades open
the higher air flow causes a higher exiting air pressure. This modulation of pressure is what the listeners
ears sense when he or she hears the reed buzzing. If
the reed is tuned to a perfect A of 440 Hz, the listener
senses exactly 440 of these cycles every second.
These cycles, while they can produce beautiful music when put into the oboe, also do enormous amounts
of damage to the reed. Every time the blades compress
or stretch, very small cracks begin to form on the surface of the cane. These cracks are initially on the order
of microns or smaller. (For those not accustomed to
such a small measurement: in each millimeter, there
are exactly one thousand microns. A human hair is
about 100 microns wide, and an oboe reed measuring
72 millimeters is 72,000 micrometers long.) Each time
the blade flexes, these cracks are pulled apart by tension in the surface of the cane, and so every cycle the
reed generates will add a little more depth. As above, a
reed playing at 440 Hz goes through 440 cycles every
second of playing; a higher pitch will be at a higher
frequency, and will require more cycles per second.
This gradual process of cracking is the cause of
fatigue. After long enough, these cracks will weaken
the structure of the blades, causing them to flatten and
eventually stop sounding. If one were to continue to
flex the blades after they stopped producing sound, the
cracks would form more quickly between fibersthat
is, lengthwisethan they would across fibers, and at
some point the micro-cracks would likely join to cause
a catastrophic failure where the cane splits visibly
and permanently down the grain.
Going back to the fibrous nature of the plant, it can
be seen that the difference in fiber and lignin strengths

Figure 9: A diagram of the reed blades during vibration.

Red arrows show compression and blue arrows show
tension.

is also the cause of (1) above. When first preparing

cane, it is impossible to tell whether or not the cane
is strong enough to withstand playing for a long time.
Because cane is grown naturally, it is expected that
there will be a great deal of variations from piece to
piece; like a fingerprint, no two canes should be identical. A particular cane tube could naturally be extremely poor for reedmaking, but the reedmaker himself or
herself may not know it until he or she is unable to
produce a working reed. This unreliability also exasperates (2), as all the effort expended in making this
doomed reed will ultimately be for nothing, which is
both a waste of time and a frustration for the reedmaker. The expenditure of effort, then, leads to (4); if a
reedmaker is spending an hour of skilled labor on each
reed, and a significant number of reeds cant even be
sold, it makes sense that reeds are so expensive. Add
the fact that the highest quality cane often comes from
France and must be shipped internationally to reedmakers, and the price only becomes steeper.
A plausible solution might be to replace the expensive Arundo cane with a cheaper material such as
bamboo. Unfortunately, this is not a viable solution,
and it is possible to compare several properties of
these two materials to explain why. A brief definition
of these parameters:

32

AN OVERVIEW OF THE RISE OF ARUNDO DONAX IN OBOE REEDS

c. The curve for cane is largely straight,

while the curve for bamboo is jagged.
This shows that the cane material is likely
to form a brittle, sudden fracture when it
fails, while bamboo will fail partially at
several points before separating fully. The
implication of this difference is that there
is a significant difference between the two
materials structures. Coupled with the
greater tensile strength of bamboo over
cane, this suggests that there is some sort
of adhesion present in the bamboo structure holding it intact.
From these observations, it should be obvious that
even though bamboo and Arundo donax may look
similar at first glance, bamboo is simply too different
to substitute for cane. Even if scraping bamboo into the
correct shape and tying it into a reed is possible, it will
be far too brittle to vibrate without cracking quickly.
It seems, then, that the solution to many of the oboe
reeds problems is to get rid of the variable nature of
the starting material not by finding a natural replacement, but rather by using a new
material chosen to match the properties of
cane. Some of the more prominent double
reed companies have tackled the problem
by producing synthetic reeds over the past
several years. Unfortunately, they are generally panned by users due to their wildly
differing characteristics from natural cane.
Figure 10: Stress-strain curves for cane and bamboo. Note the dierOne company, Lgre Reeds, has successing scales; if the curves were overlaid, the cane curve would be much
fully
created a mass-produced single reed
longer and shorter than that of bamboo.
for the clarinet and saxophone made of
specially-processed polypropylene (think
yogurt cups). They boast an impressive roster of proSeveral important things can be gathered from
fessionals from around the world using these reeds.
these plots:
However, they have been unable so far to create a sata. Note that the maximum stress, or tenisfactory double reed; as they say themselves, it is a
sile strength, of cane is only 5 MPa,
lot harder than we originally thought it would be!18
while bamboo can withstand stress of
up to 300 MPa. As stated in the definiInstead of simply looking at reeds from the same
tion of stress above, this gives bamboo a
viewpoint that we look at plastic toys or toolsthat
strength roughly half that of steel (which
is, molding them into the correct shape and sending
explains why it is often used for constructhem on their waywe must look carefully at the
tion purposes). In contrast, cane only has
characteristics of the materials we choose. New matea strength of 5 MPa, which makes it less
rials must be able to replicate the desirable qualities of
resilient than a similarly-sized piece of
caneits flexibility and hardness, density, and acousLDPE plastic.
tic properties, for examplewhile improving its weaknesses, such as its tendency to split easily. This may be
b. Also note the maximum strain of each
as simple as finding what is needed in a catalog, or it
material. Bamboo has a maximum elonmay be as difficult as developing a new plastic resin
gation of 1%, which makes it less ductile
with the correct properties.
than most metallic alloys. On the other
hand, cane can elongate by up to 11%.
t

4USFTT is the force being applied per unit
area, similar to a tensile (or negative)
pressure on the material. Here, it is measured in megapascals, or newtons per
square millimeter. The higher the force
that is applied to the cane, the higher the
stress. (For comparison, the maximum
stress before failure of a typical steel is
about 700 MPa. LDPE, the plastic used
to make grocery bags, tears at roughly 10
MPa.15
t

4USBJO is how much longer the material
gets under a given stress. If a material is
initially 100 millimeters long and is elongated to 110 millimeters, it is under a
strain of 10%. Brittle materials such as
glass cannot undergo high strain, but ductile materials like rubbers can.
These properties, when plotted against each
other, give a picture of the physical properties of the
material16, 17:

Back to Table of Contents

THE DOUBLE REED

33

SUMMARY

ACKNOWLEDGEMENTS

Before pulling out the plastic catalogs, it is important

to quantitatively measure the characteristics of the
reed. Any oboist can tell you that the reed usually
weakens after a few playing hours, but in order to solve
the problem we need to take a closer look at the reason
behind this. By taking samples from worn reeds and
analyzing them under a high-powered microscope, it
should be possible to examine fatigue patterns and see
exactly which regions of the reed need to be strengthened. Knowing this puts a reedmaker in an excellent
position to correct the problem: for example, if the
heart of the reed is where the most damage occurs,
then it may be possible to apply a particular processing technique to, for example, add surface compression to the heart and prevent cracks from forming or
enlarging.
As a way to determine the wear pattern of a reed
over its playing life, I have proposed an experiment to
be completed over this summer (2010). The experiment will call for an artificial mouth with material to
mimic an oboists lips, powered by an air compression pump to provide constant airflow. When operating, the mouth is intended to hold on to a reed
and play it for several hours without stopping. This
way, we can accurately measure the wear the reed is
subjected to over its lifetime, and can further focus
our efforts on strengthening those aspects of the
reed. This research is similar to that of John Gillette,
who examined the conditions under which a reed
will perform (see Some Measurements of Pitch and
Air Pressure of Bassoon Reeds in The Double Reed
2004 for more information).
Overall, the quantitative study of the reed at work
is something that could easily be groundbreaking for
the double reed player. There are mountains of anecdotal evidence for different problems with the traditional cane reed, but to date very little has been done
to measure them accurately and apply these measurements to new approaches to the reed. The fact that Lgre has been able to make a globally-acclaimed synthetic single reed out of such a readily available plastic
is very promising, and is a very good sign that the
same is possible with a double reed. If oboists are able
to contribute to studies on the oboe reedor even better, perform their ownit will provide a much-needed
push toward an improved modern reed.

This work was conducted as part of an undergraduate Interactive Qualifying Project, or IQP, necessary to
fulfill the undergraduate degree requirements at WPI.
As such, I wish to sincerely thank Prof. Satya Shivkumar and Mr. Joe Halko for their work as advisors on
this paper.

ENDNOTES
1. TRENTACOSTI, Michael. WHAT CAN I SAY?.
The Double Reed, 1978, vol. 1, no. 1 [cited 6 Mar
2010]. Available from <http://www.idrs.org/publications/DR/DR1.1/what.html> International
Double Reed Society.
2. PERDUE, Robert E., Jr. Arundo Donax: Source
of Musical Reeds and Industrial Cellulose. Economic Botany, Oct. - Dec., 1958, vol. 12, no. 4. pp.
368-404. ISSN 00130001.
3. Biosecurity New Zealand. Giant Reed. 20 Mar
2009, 2009Available from:<http://www.biosecurity.govt.nz/pests/giant-reed>.
4. DIGIORGIO, Marguerite. Reeds as Weeds. Science News, 22 Feb, 1992, vol. 141, no. 8. pp. 127.
ISSN 00368423.
5. VESELACK, Marilyn S. Arundo Donax: The
Source of Natural Woodwind Reed. The Double
Reed, 1979, vol. 2, no. 1 [cited 6 Mar 2010]. Available from <http://www.idrs.org/publications/
DR/DR2.1/arundo.html> International Double
Reed Society.
6. SNIECKUS, Kurt. Cross Section of Arundo
Donax. , 2010 [cited 12 Apr 2010].
7. BISHOP, Douglas. A Worldwide History of the
Panflute. , 2003Available from:<http://www.panflutejedi.com/pan-flute-history-main.html>.
8. BAINES, Anthony. Woodwind Instruments and
their History. 3rd ed. Mineola, NY: Dover Publications, 1991. ISBN 0486268853.
9. LEDET, David A. Oboe Reed Styles: Theory and
Practice. Bloomington, IN: Indiana University
Press, 1981. ISBN 0-253-37891-5.
10. BARDANT, Jean-Xavier. Zournas. Wikimedia,
2006 [cited 7 Apr 2010]. Available from <http://
en.wikipedia.org/wiki/File:Zournas.jpg>.
11. LAZAR, Bill. Shawm, Alto in G, Copy of Charles
Collier. , 2010 [cited 7 Apr 2010]. Available from
<ht t p://w w w.la z a rsea rly music.com/UsedInstruments/images/Lazar/CollierCopyAltoShawm.jpg>.

34

AN OVERVIEW OF THE RISE OF ARUNDO DONAX IN OBOE REEDS

12. WOLDBYE, Ole. A Register of Early Reeds (GSJ

XXX, p. 150). The Galpin Society Journal, May,
1978, vol. 31 [cited 19 Apr 2010]. pp. 144. Available from <http://www.jstor.org/stable/841196>
JSTOR. ISSN 00720127.
13. WARNER, Thomas. Two Late Eighteenth-Century Instructions for Making Double Reeds. The
Galpin Society Journal, 1962, vol. 15 [cited 14 Apr
2010]. pp. 25. Available from <http://www.jstor.
org/stable/842036?seq=4> JSTOR.
14. FROST, Eberhard. Arundo Donax Cross Section.
www.the-clarinets.net [cited 19 Apr 2010]. Available from <http://www.the-clarinets.net/images/
arundo_schnitt_mikroskop.png>.
15. CALLISTER JR., William D. Materials Science
and Engineering: An Introduction. 7th ed. New
York, NY: John Wiley & Sons, Inc., 2007.
16. AMADA, Shigeyasu; and UNTAO, Sun. Fracture Properties of Bamboo. Composites Part B:
Engineering, 2001, vol. 32, no. 5. pp. 451-459.
Available
from
<http://www.sciencedirect.
com/science/article/B6TWK-4372WGD-6/2/
ISSN
a8a24e2d612b5776145dfdab5c745682>.
1359-8368.
17. SPECK, O.; and SPATZ, H. C. Mechanical Properties of the Rhizome of Arundo Donax L. Plant
Biology, 2008, vol. 5, no. 6 [cited 29 Apr 2010]. pp.
661-661-669. Available from <http://www3.interscience.wiley.com/journal/120130826/abstract>
Wiley InterScience.
18. Lgre Double Reeds Ltd. Lgre - Double Reeds.,
2009. Available from:<http://www.legere.com/
index.php?page=double-reed>.

Back to Table of Contents

THEDOUBLEREED

35

Th e Tradition of Harvest and Curing of Arundo Donax L.

for Reed Cane,Part 2
Kirsten M.Boldt-Neurohr
Ellensburg, Washington

eeds for bassoon and other woodwind instruments are made from the Giant Reed,
Arundo donax L. Unfortunately cane that is
commercially available varies considerably in quality and consistency. Multiple studies have been done
on the various factors of Arundo donax that create inconsistencies in cane samples, on the ways by
which musicians can obtain more consistent results
from cane, as well as studies on the ways in which to
choose cane that display the playing characteristics
that the individual desires. However, no research has
been done on the eff ect of the micro-climate during
the cultivation of Arundo d. As it is, growers use a
variety of techniques to harvest and select the internodes used for reeds. Th e internodes shipped to
musicians are selected by diameter, color and other
variables; but the resulting batch is inconsistent from
piece to piece. Consequently, many internodes are
thrown away and wasted because they do not meet
the personal preferences of the musician.
Th is frustration could be alleviated by some ad ditional knowledge of the conditions in which the
cane was grown. Like grapes for wine, the climatic
patterns of the region where the cane is cultivated affect the structure of the cured product that musicians
use for reeds. Th e variation in weather that occurs
from year to year aff ects each harvest diff erently, as
in wine vintages. I have done a study which assesses,
through objective and subjective ways, the eff ect that
climate has on Arundo donax during cultivation, and
the resultant eff ects that the yearly variable weather
patterns have on the cane.
Four harvests from the same patch of Arundo
donax were used for the study. Located in Ashland,
Oregon, the ten year-old crop has been maintained
and carefully documented by the author and by the
head grower, Darrell A. Boldt. Th e irrigation, fer tilization, harvesting and curing procedures have
remained constant. Th e only variable has been the
natural weather patterns throughout the years. Th e
variations of certain parameters are then correlated
to the weather patterns of the specifi c region in which
the cane is grown, so as to determine the results that
diff erent changes that weather conditions have on

cane.
Th e fi rst step in the production of cane is the
growing and harvesting techniques employed by the
growers. Th is is a great tradition passed on through
generations. Although there are some variables from
grower to grower, the basic procedure remains the
same.
Th e main sources for commercial cane are
France, Spain, Italy and California. More recently
quality cane has been grown in China, Argentina
and Australia. However, the area in the South of
France along the Var River is widely held as the premiere location for growing cane. Th e Var River runs
for seventy-fi ve miles before entering the Mediter ranean Sea four miles South of Nice. Predominantly
the towns Hyeres [43 7 0 N, 6 7 0 E] and Frjus
[4325 60 N 6 43 60 E] have oft en been thought of
as the premiere locations for the growing of cane because of the suitable growing conditions created by
the unique micro-climate. Th is is characterized by
light rainfall, average sun days and proximity to the
cool sea breezes, but not too close because of the salt
content carried on such breezes. Additionally the region has a rich alluvial soil high in silicone, nitrogen
and bauxite (an aluminum ore resultant from volcanic activity) content.1
While it is true that climate and soil both have
made this area highly suitable for the growing of
Arundo d., contrary to what was once believed, it is
not true that the high quality cane can only be produced from a certain strain found in the Southern
region of France.2 Today, cane from many geographic
locations is successfully used to make reeds; many
double reed players now prefer cane from locations
other than France, such as Argentina, Australia and
China. To the authors knowledge no comprehensive
study has been conducted on either the climate or the
soil to determine why the Var region produces such
fi ne cane. Robert Perdue states his opinion on the
matter in his article Arundo donax-Source of Musical Reeds and Industrial Cellulose:
A careful study of atmospheric and soil conditions under which cane is produced in Southeastern
France should point the way to the most satisfactory

36

THE TRADITION OF HARVEST AND CURING OF ARUNDO DONAX L. FOR REED CANE, PART 2

American environment [to grow cane]. Such an environment appears to be in the Southwestern or western United States. Careful study of the botanical and
agronomic characteristics of the plant, in connection
with the detailed study of the physical and musical
quality of the cane could be made.3
Th e belief that only the French can produce the
best cane was strongly encouraged by the French
growers. It has historically been the largest producer
of quality material; this monopoly was not contested prior to World War I. Also, in the 18th and 19th
century many woodwind instruments gained great
mechanical advancements. At this time the majority
of the premiere woodwind manufacturers were located in France, leading to the domestic propagation
of French cane. Post World War I there was a large
cane shortage because of the damage done to crops
during the battles in France.4 As a result of this wartime shortage, musicians made eff orts to grow cane
in parts of the United States as well as other countries, thus spreading the propagation of cane to other
geographic locations.5 Granted climate and soil have
both made the Var area in the south of France ideal
for the growing of Arundo d., but it is no longer true
that high quality cane is only produced from a particular strain found in the Southern region of France.

THE CULTIVATION OF ARUNDO DONAX FOR

MUSICALREEDS
Arundo donax is not sprouted from seeds, but from
vegetative propagation, the dividing and transplanting of the rhizome. Th e rhizomes are planted six to
eight feet apart and covered with soil up to a depth
of ten centimeters,6 and can be simply cultivated by
maintaining weeds, watering if necessary, adding
fertilization and removing culms that are of undesirable diameters. A less common form of propagation
is horizontally laying a culm under the surface of
the soil. Th is will eventually send resultant adventi tious roots down into the water base, while sending
small culms up.7 Th is procedure will eventually cre ate a rhizomatous matter that continually produces
culms. It is because of these adaptive growth patterns
that Arundo d. is considered an invasive pest and the
California Department of Agriculture spends a great
deal of research on how to eradicate the plant.
Arundo d. is adaptable to a variety of soil conditions. In the wild, Arundo d. has been able to fl ourish
in soils that are apparently very infertile, even soils
that do not sustain other crops. Perdue states that
in cultivation the plant responds well to soil that has

Back to Table of Contents

been improved, particularly soil that is high in nitrogen.8 Th is has also been reiterated by Mark Eubanks,
founder of Arundo reeds and well-known expert on
the growing of Arundo d. According to a scientifi c
botany database, Arundo d. prefers fertile soil in
a sunny sheltered position [is] best developed in
poor sandy soil[and] said to tolerate all types of
soils, from heavy clays to loose sands and gravelly
soils.9 However, the information in this site was not
dedicated specifi cally to the growing conditions of
Arundo d. for the use of musical reeds.
Some growers hold the opinion that clay soils
are optimal while others say that only quality cane is
grown in loose sandy soils.10 It appears that the most
suitable habitat for the production of cane is a light
soil which provides adequate moisture for maintenance and continuous growth. A site near to the sea
is regarded as preferable to a more inland location.11
Annual thinning of smaller diameter culms encourages the rhizomes to produce larger diameter culms,
provides adequate room for the inside culms to prosper without crowding, and ultimately increases the
density of the stand.

HARVEST TECHNIQUES
A new plantation of Arundo d. requires up to fi ve
years growth before the fi rst successful harvest of
usable cane. Harvest occurs when the plant is dormant so that there is little or no sap movement in
the plant. If cane is harvested while the sap is in full
movement the cellular structure is not hardened and
the resulting cane is greenish in color which does not
abate with sun curing. Th is results in cane that is
considered too soft and unsatisfactory. 12 Th e key to
a successful harvest is the temperature of the soil at
the time of harvest, which is ideally 40F or below.
It is also preferable to have conditions in which the
weather has been cold for several weeks, with extended periods of lower temperatures so the sap is completely settled in the lowest internodes of the culms,
below soil level.13 Th erefore, harvest time should oc cur in the winter months of December to February
in the northern hemisphere. Harvest should also be
completed during the phase of the new moon to also
ensure little sap fl ow. As with the tides of the sea, sap
fl ow in plants are directly aff ected by the phase of the
moon.14
Only poles in their second year of growth should
be harvested, as the cellular walls in the fi rst year
have not hardened and will subsequently result in
cane that is considered too soft for reed making.

THEDOUBLEREED

Various scientists have given two important reasons

for why the culms need to reach a certain maturity
for development, both of which concern the growth
and thickening of cell walls that occur in Arundo d.
aft er the fi rst year of growth. 15
1.) Lignin is the molecule that gives cane the
stiff ness and vibration needed for reeds,
and it increases over time; Ligninis
present in the cortex region of the twoand half year culms.16 Lignin is formed
from dead cells and gives water protection to more vulnerable cells, acting as a
pathogen barrier and a parasite growth
inhibitor. Th e older the section of plant
the more likely lignin is to be found.17
According to scientist J.P. Joseleau in his
1975 article, there is a direct correlation
between age and lignin; the concentration of lignin increases about ten-fold as
one goes from the top (younger portion)
of the stem to the bottom (older portion).18 Lignin helps cells resist excessive
water absorption that would otherwise
create decay.
2.) Hemicellulose is hydrophilic, in that it
aids in the absorption of water. Th is al lows the cane to absorb moisture and store
the water when the reed is soaked.19 Th e
concentration of hemicellulose decreases
with the maturity of the culm.20 Later
harvesting of Arundo d. leads to stable
concentration of [hemicellulose] in the
plant stem. Alternatively, younger stems
sectioned near the base of the plant (the
older portion) show the same trend.21 As
James Kopp adequately explains, underdeveloped sections of cane may have too
much hemicellulose, which would lead
to excessive water absorption of the reed,
creating a reed that is too soft . 22
Th e anatomy of Arundo d. changes with maturity. In her dissertation, Clare Lawton explains that
Moisture content, distance from the bottom of the
plant, and proximity of the sclerenchyma to the epidermis all appear to aff ect the strength of the culm. 23
She continues, It is probable that anatomical features that infl uence the mechanical properties of liv ing plant material, also aff ects the tonal qualities of
musical instrument reeds.24 Lawton states that cane
harvested aft er two years will have more scleren chyma with thicker walls than a one year old plant.
Sclerenchyma cells are composed of both the lignin

37

which hardens the cell walls and hemicellulose.25 Favorable reeds with preferred tonal quality are generally found to have a larger portion of sclerenchyma.26
Marilyn Veselack and Donald Cassadonte also conclude in their independent studies that with maturity
the more favorable levels of lignin and hemicellulose
are present.
Poles of second year growth are easy to spot by
experienced growers because the small branches of
a culm only emerge aft er the fi rst season (See Ill. 1
and 2, following page). Also, the newest stalks grow
around the parameter of the cane patch, although if
the center of the patch has been suffi ciently maintained and thinned of stalks of undesirable diameter,
new shoots will also emerge in the center. For this
reason it is best to look for the additional branches or,
alternatively, mark the fi rst year culms as they come
up for proper identifi cation the following seasons.
Like a tree that puts out small shoots or suckers from its base of the tree, Arundo d. also sprouts
smaller culms aft er a larger culm is cut. To ensure
that the rhizome will send up additional culms of
the desired diameter for bassoon reeds, the harvest
technique is to cut the culms at the rhizome, several
inches below dirt level. If this is properly done, the
rhizome will sprout another large diameter for the
next season.

DRYING AND CURING PROCESS

Th e treatment of the harvested cane varies greatly
from plantation to plantation, depending upon the
individual grower and on the preference of the reed
manufacturer for whom the cane is grown. Typically, once cut the culms are oft en stripped of the
branches and husked. However, some growers keep
the husk and branches intact while the stalks are
tied in small bundles and vertically stored in a cool
storage area with good circulation, to dry for fi ve to
eighteen months. Many French growers will bundle
in a crab, approximately twenty stalks placed leaning towards each other, resembling an Indian teepee.
Th e teepee is then left outside to dry for the remain der of the mild southern French winter.27 Certain
growers prefer to initiate the process of curing the
cane under the shade of a tree or an open-sided shed
to provide protection from the sun.28 No matter the
preference, the initial drying requires at least four to
eighteen months, during which the moisture in the
cane is evaporated out.
Th e culms are then stripped of the leaves, branch es and husk, if not done previously, and are sun-cured

38

THE TRADITION OF HARVEST AND CURING OF ARUNDO DONAX L. FOR REED CANE, PART 2

for a minimum of three weeks29 up to as long as two

years.30 Th e curing process requires the stalks to have
at least three weeks of southern sun exposure in order
to prevent cracking and to prevent the cane turning
a reddish color from sun overexposure.31 Th e stalks
are then suspended horizontally or laid along a rail to
cure. In at least one method, the cut internodes were
strung on a string and laid in a large fi eld. 32 Th e stalks
or cut internodes, are turned everyday or as oft en as
the grower deems necessary for even sun exposure.
Sunning is essential to stabilize the reed structure.
During this period the chlorophyll in the cell walls
is changed to chromophores (a color producing compound), which is what produces the golden yellow
color that is desirable in quality cane. Th e rigidity of
the cell walls of the cane also stabilizes. If curing does
not occur, or if it is not cured for suffi cient time (as
can be told be a green tinge in the cane) the cane will
be too pliable for reed use.33

CANE USED FORSTUDY

Ill.1. New second year branch

Ill.2.Second year branches

Back to Table of Contents

Th e crop used in my study of the aff ects of climate

during cultivation on reed cane was planted in Ash-

THEDOUBLEREED

39

Ill.3. Cane crop July,2004

land, Oregon in 1996. Th e rhizome and sprout used

to start the crop was given to the author by Mark
Eubanks, the principal bassoonist of the Oregon
Symphony (1978-2007) and former owner of Arundo
Reeds. Eubanks has been growing cane for reeds experimentally since 1982, in two separate locations in
Northern Oregon. Eubanks originally received his
starting cut of cane from Bob Stevens in California,
who received the initial rhizome for his cane directly
from a plantation in the Var region in France.
Th e plot in Ashland, [42 194 N, 122 78W] has
a climate of hot, dry summers and cool, wet winters.
Th ere were three original rhizomes and sprouts that
were planted in three locations. Th e fi rst was on a
hill in the back of the growers property, adjacent to
the vegetable garden where the soil is high in silicon
content and well fertilized with organic compost. Th e
second sprout was planted in the front of the property which is a slope of clay soil overlaying fi ssured
sandstone rich in silicone. Th e third was planted on
the bank of the Rogue River on family property outside of Eagle Point, Oregon. Th e latter was unfortu nately uprooted and swept away in the fl ood of January, 1997; therefore it was not included in this study.
Because growth on the back patch was more vigorous, the second plot was transplanted to augment the

fi rst patch because the underlying sandstone made it

diffi cult to send out the small adventitious roots for
moisture.
Once the fi rst stalk or rhizome is planted in the
ground, it takes two to fi ve years for the stand to
gain enough maturity for harvest. In the fi rst year
of growth only fi ve to ten young stalks came up.
Although thinning the stand does not decrease the
growth of new stalks, the author did not thin the
stand until three years aft er initial planting. Th e fi rst
harvest was in January 1999, at which time fi ft een
useable bassoon diameter culms were cut for the drying and curing process. At the same time we minimally thinned the unusable diameters. In the winter
of 2000, there were not enough stalks in the correct
bassoon diameter for a viable harvest. Th e winter of
2001 had a good harvest but was unusable for this
study because of a lack of consistent fertilization.
Th erefore the fi rst fi ve harvests of the cane were un usable for this study for reasons of inconsistent fertilization and too little or no useable diameters.

PARAMETERSFOR CHOOSING
FOUR HARVESTS
Th e harvests for the years 2004, 2005, 2006 and 2007

40

THE TRADITION OF HARVEST AND CURING OF ARUNDO DONAX L. FOR REED CANE, PART 2

Ill.4. Dorm ant caneJanuary 2,2004

Ill.5. Harvest,January 2,2004,the author and the head

grower

were chosen for this study because they had consistent fertilization and irrigation schedule; they had a
signifi cant count of usable culms at harvest time; the
thinning procedures were regulated and the weather
data was consistently recorded. Beginning in 2002
the crop was fertilized once monthly during the
warm growing months with high nitrogen fertilizer
with phosphorous and iron. Because the author chose
these growth years to study because of consistent record of irrigation, fertilization and weather, the only
variation in measurable parameters was the climate.
In the study by Peter Kolesik, musicians and scientists compared cane from two diff erent locations
in Australia. One was a plantation with regulated
fertilization and irrigation, the other from a windbreak growing in a wild location. Th e average stiff ness for the reeds grown in the regulated plantation
was higher. [T]he plantation [cane] had a signifi cant
higher proportion of cortial vascular bundles with a
continuous fi ber ring, and had bundles with a higher
proportion and area of fi ber. 34 Because of this fi nd ing, the use of only the harvests that had the same

Back to Table of Contents

fertilization was paramount for the validity of this

study.
Irrigation is maintained by both a soaker hose
and a non-oscillating sprinkler during the dry summer months in addition to the precipitation provided by the recorded natural weather patterns. Th e
cane was watered regularly during the typically dry
months of July, August and September for all six of
the years leading to the four harvests chosen. A soaker hose that was run through the entire patch of cane
was used for two to two and a half hours two to three
times a week. Once a month the grower would use a
non-oscillating sprinkler for one and a half hours in
the center of the patch to ensure the newly laid fertilizer would penetrate the soil.
Th e soil of the plot is high in silicon and clay. Lo cated adjacent to a vegetable patch that is fertilized by
natural compost and manure once a year, the Arundo
d. shoots oft en sprout into the garden, a natural in vasive characteristic. However the sprouts that are
found in the garden are consistently greater in diameter compared to the rest of the patch, indicating a
further benefi t from additional organic fertilization.
Beginning in the summer of 2002 the plot was fertilized with 25 pounds of high nitrogen fertilizer once
a month during the summer months. Th e contents
are 23% nitrogen, 3% phosphate, 13% potash, 9% sulfur, and 2.5% iron, derived from ammonium sulfate,
urea, ammonium phosphate, potassium chloride,
ferrous sulfate and sulfur coated urea. Th e cane is
mulched with straw and leaves twice a year.
Th e harvest of the crops used for study occurs
every December or January at or within a day of the
new moon. An all day process, at least two harvesters are used: one to identify the second year culms
while the other digs deep into the soil at the base of
the culm in order to fi nd the Rhizome. (See Ill. 6) We
use a reciprocating saw to cut the culm directly at
the rhizome below the surface of the soil in order to
ensure another large diameter culm grows from the
cut. Once felled, the culm is measured to determine
which internodes are usable. Th e culms are then
stripped of branches and the dry husk then bundled
in groups of ten to fi ft een culms for drying (See Ill. 7).

DRYING AND CURING PROCEDURES

Th e bundled culms are stored in the downstairs stor age room of the growers house. Since it is not connected to a heat source, it is a cool room with ample
circulation. As mentioned above, many growers dry
the cane for only a few months in crabs outside. I have

THEDOUBLEREED

Ill.6. HarvestJanuary,2005

Ill.7. H2005,bundled and readyfor cold storage

41

42

THE TRADITION OF HARVEST AND CURING OF ARUNDO DONAX L. FOR REED CANE, PART 2

Ill.8. Culm bundles drying vertically in cold storage

Ill.9. Cutting the culms after sun curing

Back to Table of Contents

varied this technique for spatial reasons in the storage room and because the wet winters common to
Oregon would make drying outside impossible. Cane
that is grown in China is oft en shipped to another
more southern location for warm outside curing.35
Th e bundles of ten to fi ft een culms are stored verti cally leaning against a wall, so that any moisture will
not collect to one side of the culm but slowly drain to
the bottom internode which will be discarded. Th e
cane is kept in the cool dry storage for 18-21 months
before sun curing (see Ill.8).
Once dried, cane must be cured in the sun to ensure that all of the moisture is evaporated from the
stalk and that the chemical composite of the cane
stabilizes. Cassadonte explains the scientifi c reason
for the change in color from green to yellow, As cane
ages, the relative amount of xyanthophyll (which
is yellow) increases and the amount of chlorophyll
(which is green) decreases.36 Th e poles are laid on
a fl at surface with Southern facing sun exposure to
ensure that they get a complete days worth of sun exposure, but do not get the intense heat of either East
or West exposure which could over harden the cell
structure of the cane. Th e poles are rotated a quar ter turn once daily, to ensure even exposure for four
weeks. At the end of the
curing process, the poles
are cut into internodes,
selecting only those of the
correct diameter and put
in ventilated boxes for storage until testing.
Th e art of harvesting
and properly curing reed
cane has been passed on
for centuries. If done incorrectly the cane is unusable by even the least picky
players. However, despite
the meticulous treatment
of cane by all growers, reed
makers around the globe
encounter a plethora of unsatisfactory or simply poor
cane. I am now convinced
this is because varying
weather conditions that
occur year to year during
the cultivation of the cane
strongly aff ects the fi nal
product. Th e following cor -

THEDOUBLEREED

43

responding articles will report the initial study and its

fi ndings. I hope this will eventually encourage grow ers to indicate harvest years on the cane and the resulting characteristics of that harvest year. Th erefore,
like when we buy a bottle of nice wine dependant on
the year, we will have the same luxury with our cane.

the Gouger, Journal of the International Double

Reed Society, no. 11. (1983): 17. Also, Mark Eubanks, op. Cit..
14 Eubanks, op. cit.
15 Marilyn S. Veselack and Jerry J. Nisbet, Cell

Dr. Kirsten Boldt-Neurohr holds a Bachelors of Music

in Education and Performance from Ithaca College, A
Masters in Music performance from the San Francisco
Conservatory of Music and Doctoral of Musical Arts
Degree from University of Colorado. In addition to
her degree studies she spent a semester studying at the
Royal College of Music in London, England. Previously
the bassoon instructor at Metropolitan State College in
Denver, Dr. Boldt-Neurohr presently lives in Washington
where she teaches privately and actively freelances,
she also still holds the second bassoon position in the
Steamboat Springs Orchestra in Colorado. Th is coming
spring, Dr. Neurohr will teach music theory at Central
Washington University. Her principal teachers have been
Mark Eubanks, Lee Goodhew-Romm, Martin Gatt,
Rob Weir, Stephan Paulsen and Yoshiyuki Ishikawa.

16 Ibid.
17 Cassadonte, Th e Clarinet Reed, 106.
18 J.P. Joseleau and F. Barnoud, Hemicellulose of
Arundo donax at diff erent stages of Maturity,
Phytochemistry vol. 14, issue 1 (January 1975) 56.
19 Cassadonte, Th e Clarinet Reed, 163
20 Joseleau, Hemicellulose of Arundo donax, 74.
21 Cassadonte, Th e Clarinet Reed, 160.
22 James B. Kopp, Counting the Virtues of Bassoon
Reed Cane, International Double Reed Journal
vol. 26, no. 4 (2004): 8.
23 Clare Elizabeth Lawton, A Study of Variation
in the quality of Oboe Reeds made from Arundo
donax L. Gramineae, (PhD. Diss. University of
Reading, 1998), 68.
24 Ibid
25 Kopp, Virtues of Bassoon Reed Cane, 5.
26 Lawton, A Study of Variation, 141.
27 Peter Hedrick, Notes on Cane Growing from the
South of France Woodwind World vol. 13 issue
1(1974): 15.
28 Perdue, Arundo donax Source, 382.
29 Ibid. 383.
30 Cassadonte, Th e Clarinet Reed, 30.
31 Perdue, Arundo donax Source, 381.
32 Ibid.
33 Cassadonte, Th e Clarinet Reed, 31.
34 Peter Kolesik, A. Mills and M. Sedgley, Anatomical Characteristics Aff ecting the Musical
Performance of Clarinet Reeds Made from Arundo donax L.(Gramineae) Annals of Botany 81,
(1998):154.
35 Eubanks, personal communication, Aug. 2007.
36 Cassadonte Th e Clarinet Reed, 84, 89.

ENDNOTES
1

3
4
5
6
7
8
9

10
11
12
13

Robert E. Perdue, Jr., Arundo donax-Source of

Musical Reeds and Industrial Cellulose, Economic Botany, vol. 12, No. 4 (October-December,
1958). 391.
Donald Jay Cassadonte Th e Clarinet Reed: an
Introduction to its biology, chemistry and physics (DMA diss., Th e Ohio State University, 1995),
22.
Perdue, Arundo donax-Source of Musical Reeds
and Industrial Cellulose, 391.
Ezra Kotzin, Th e Reed Problem Woodwind
Magazine, vol. 1 no. 1 (Nov. 1948): 4
Perdue, Arundo donax Source, 384.
Ibid
Ibid, 381.
Ibid, 371.
PlantsforaFuture.org, Database [website] (Plants
for a Future [cited Sept. 6, 2008]); available from
h t t p : / / w w w. i b i b l i o . o r g / p f a f / c g i - b i n /
arr_html?Arundo+donax
Perdue, Arundo donax Source, 380.
Ibid.
Mark Eubanks, personal communication, December 6, 2006.
John W. Reid, Cane Selectivity from the Field to

and Tissue Diff e r ences in good and usable

woodwind reeds from Arundo donax L. Th e
Clarinet Vol. 8, no 4. (1981): 9.

THE DOUBLE REED

99

What Makes a Good Reed? An Explanation of Flexibility, Density

and Hardness of Arundo Donax L. Part III.
Kirsten M. Boldt-Neurohr
Ellensburg, Washington

the specific region in which the cane was grown. (The

results of which will be explained in a future article.)
From this I was able to deduce the effects that different climatic changes have on cane. A quick side
note: each harvest from my study is referred to by
its harvest year (ie: H2004). Because the cane is harvested after the second year growth, the climate data
was collected for the two years prior to the harvest.
Therefore, any mention of the climate data for an individual harvest is referring to the two years prior to
the harvest date. For instance the climate data for the
2004 harvest (H2004) is referring to the years 2002
and 2003. Each harvest was cut at the new moon in
January of the harvest year.

HARDNESS
The hardness of cane is often cited as the most desired
quality in cane for reeds. Random House Dictionary
defines it as the resistance of metal or other materials
to indentation by an indenter of fixed shape and size
under a static load or to scratching.4 Karl Almenraeder in his 1842 treatise on making bassoon reeds
suggests using a fingernail test. The tester should run
his thumb nail along the epidermis of the cane, following its curvature to make a faint mark; if the resulting mark is as deep as it would be in a pencil then
the cane is too soft.5 He also states that if no mark is
left, the cane is too hard. However it has been noted
that Almenraeder liked fairly soft cane, so an absence
of a mark is not necessarily an indication of overly
hard cane but dependant upon personal preference.6
Donald Cassadonte7, Kopp, and John Reid all refer
to this method of testing hardness in cane. Reid indicates in his 1983 article Cane Selectivity from the
Field to the Gouger, that the cane should be hard:
the mark should be difficult to make and should not
go very deeply into the surface.8 As with all qualities
of cane the reed maker must find the degree of hardness that he or she prefers. Lawrence Intravaia found
in his study that cane from the same growth tended
to test approximately the same hardness.9 However,
he does not note if it is from the same harvest year or
simply the same cane source. In my research, hard-

Back to Table of Contents

ARTICLES

hat makes a good reed? If we had that

answer there would never be another
unhappy double reed player again. However, everyday oboists, bassoonists and some scientists (sometimes a lovely combination of these professions) come closer to narrowing the factors that do
in fact make a good reed. Or at least a reed that is
more consistently agreeable. During my research on
the effects of climate on Arundo Donax, I needed to
set parameters that were tangible and easily tested to
determine the variations in different cane. Ultimately, I came to agreement with the previous research of
many other double reed players. There are three vital
characteristics that, when tested and the preferred
balance found, can greatly enhance the chances of
getting a great reed: these are the flexibility, the density and the hardness of the cane.
The three testing parameters chosen for my study
on the effects of climate during cultivation of Arundo
Donax, are considered by a majority of reed makers
as the most important virtues to be found in quality
cane. These are characteristics that can be described
quantitatively with numerical precision. Edwin Lacy
in his article Testing the Specific Gravity of Bassoon
Cane lists some qualities that contribute to the vibratory characteristics in cane. The three qualities he
lists, hardness, resilience (or flexibility) and specific
gravity are not only important but can be measured
and quantified.1 James Kopp in his article Counting
the Virtues of Bassoon Reed Cane stated the same
three characteristics as worthwhile tests to perform
on cane to create quality reeds. He maintains that the
three most important characteristic of A. donax are
a high ration of stiffness to density, equal flexibility
in all three planes of vibration and quick damping.2
In addition Kopp included the color and the resonant
frequency of cane.3
I took these studies and others like them as the
basis of my research. A study was conducted on
four different harvests of the same crop of A. donax,
grown on a plot in Southern Oregon, the only variable was the natural weather patterns throughout the
year. The variations in the parameters discussed below were then correlated to the weather patterns of

ARTICLES

100

WHAT MAKES A GOOD REED? AN EXPLANATION OF FLEXIBILITY, DENSITY AND HARDNESS OF ARUNDO DONAX L. PART III.

ness is relatively consistent throughout an individual

(H2006) was the softest of the four harvests with an
harvest year but varies between different harvests. I
average hardness when dry at 0.199 while H2004 was
will comment on this in more detail below and in my
the hardest, with an average hardness of 0.168 when
following article.
dry. This is a 20.4% difference in hardness between
There are various meters that are commercially
the two harvests which is statistically significant.
available to test the hardness of cane. All use a spring
Equally significant, H2006 had an average hardloaded probe that penetrates the inner surface of the
ness of 0.336 while wet and H2004 had an average
gouged cane. The spring loaded probe test is more
of 0.275 while wet, also a 20% difference between the
conclusive and the results are more consistent than a
harvests. All harvest were at least 50% harder when
fingernail test. Since the dial indicator gives a numertested dry than when wet.
ical reading, the hardness
can be logged and a reed
40
maker can consequently de30
termine the exact hardness
Hardness Wet
that they prefer in cane.
20
Hardness Dry
Intravaia used a meter that tested the outside
10
of the cane because [of]
0
the extreme softness of the
10
H07
H06
H05
H04
pitch [pith] of the cane.
However, Kopp explains the
Fig. 2: Comparison of average hardness between harvests
possibility that the testing
equipment that was available at the time of his study prior to 1973, a Rockwell
DENSITY OR SPECIFIC GRAVITY
Superficial Hardness Tester, had a diamond cone indicator with a minimum load of 15 kilograms, and
The specific gravity of cane is different than the denwas most likely not sensitive enough to test the inner
sity of cane, even though some use the terms syncortex of the bassoon reed cane, as more modern maonymously. The reason is because the test of specific
chines are able to do.11
gravity is testing the ratio of the density of a specific
solid material to the density of water.12 A substance
I used an analogue hardness tester by Forrests
with specific gravity that is greater than one is densProducts for my recent study. The hardness tester
er than water and will sink in water; however those
device measures, in millimeters, the penetration of a
with a specific gravity of less than one are less dense
small, spring loaded ball bearing into the inner surthan water and will float when immersed. As Lacy
face of gouged cane; the same pressure is applied each
states, it [specific gravity] is the weight of material
time. The principal of the readings is based on the inin a specific volume, expressed as a relationship to
crement depth due to the incremental load. A larger
the density of water.13 The density of a material is
reading equals a deeper penetration, thus indicating
softer cane. The resulting reading on the analogue
the measurement of the mass per unit volume.14 In
meter is the amount that the ball bearing can penLacys floatation test, equal sized, dry gouged cane
etrate the cane in increments of 0.01 millimeters. The
was floated vertically in a water filled jar, with the
reported numbers are therefore recorded in millimeresult that some sank further than others. This ultiters. I tested the cane both wet and dry, testing the
mately tests the specific gravity of the cane not the
inner cortex surface, which will become the blades
actual density; how dense the piece of cane is gives us
of the reed. Characteristically, wet cane gives a softer
the numerical value of its specific gravity.
reading than that of dry cane. Although some of the
Bassoonists and oboists alike have devised methhardest samples when tested wet (0.22) were also the
ods of evaluating the density of quality cane prior to
hardest reading when dry (0.11), this was not always
the reed making process. The most popular of these
the case. There were a few that tested at 0.11 when dry
is the dynamic test that both Lacy and Kopp refer to
and were as high as 0.26 when wet.
in which a dry piece of gouged cane is dropped from
The hardness test produced the most interesta certain distance onto a hard surface.15 The density
ing and statistically significant results of the three
of each piece of cane affects its resonance frequency,
tests performed for my research. The 2006 harvest
when dropped; the resultant timbre or pitch heard

THE DOUBLE REED

can indicate the density. Therefore, picking pieces

of cane that produce the same pitch when dropped
will give more consistent results. Kopp explains the
conventional wisdom that the higher the resonant
frequency, the denser the cane, and therefore, the
more desirable the cane.16 The floatation test which
Lacy devises is more accurate and still easy and
inexpensive.
In her 1998 dissertation Elizabeth Lawton explains that high density in Arundo d. is the result of a
higher percentage of sclerenchyma than parenchyma.
As determined from light micrographs, the
sclerenchyma is denser than the parenchyma
and therefore the more present the higher the
density of the sample, again confirming both
subjectively and objectively that a higher level of
sclerenchyma in oboe reed cane produce reeds
of better tonal quality.17

101

ner cortex), or the state of the cane when tested (wet

or dry).
The results of the density test the four harvests
that I tested for this study were comparatively similar
throughout. The largest difference was between the
2004 harvest and the 2006 harvest. The 2006 harvest (H2006) was the least dense with an average of
69.2%, while the 2004 harvest (H2004) was the most
dense with an average of 75.5%. As will be explained
in a future article the only differences between these
harvests was the climate; the two years leading to the
2004 harvest was the driest two years of the study
but H2004 also had the two years with the greatest
temperature variations, as well as the greatest number of nights with below freezing temperatures. However, the two years leading to the 2006 harvest had
the most precipitation of the five years of recorded
weather that applied to this study. Similar to Lacy, I
also found a direct correlation between the hardness
and the density of the cane which will be explained
later.

Back to Table of Contents

ARTICLES

As stated in a previous article The Physical Properties and the Anatomy of Arundo
Donax, vol. 33 no. 4 the level of
78
sclerenchyma relates to the ma76
turity of the culm at harvest. This
74
leads one to believe that a culm
that is harvested after two years
72
Density
maturity will yield a higher level
70
of sclerenchyma, and will there68
fore produce a denser piece of
66
cane.
H07
H06
H05
H04
As described above, Lacys
test involves dry pieces of gouged
Fig. 1: Comparison of the average density for each harvest.
cane of equal length, gouge thickness and width. The tester immerses each piece of cane briefly in water and then
FLEXIBILITY
wipes off the excess water, which removes the air
bubbles that adhere to the cane creating surface tenThe flexibility of a piece of cane has also been referred
sion. The water is in a straight sided jar approximateto as the resiliency of the cane to spring back to its
ly six inches tall. Second, the tester drops the cane
original shape, the elasticity or its stiffness in vibrainto the water, letting it rise naturally so as to find
tions.20 These terms are applicable to anything that
its own level. The tester then marks the cane with a
is pliable or capable of being bent. The term resilpencil at the water line. The length of the cane which
ience is also applicable: the ability to spring back to
was immersed in the water is then divided by the tooriginal form or position after being bent or comtal length of the cane; the resulting percentage is the
pressed.21 Elasticity lies somewhere between those
specific gravity for that piece of cane.18
terms referring to the ability to stretch, expand or
take new shape under pressure, usually with a return
Lacy found that cane with density measurements
to an original shape or position.22 The idea of a return
ranging from 78% to 82% make reeds with similar
playing characteristics. He also found a strong corto an original shape relates to the aspect of flexibility
relation between the hardness test and the density
and to the stiffness of the material. Intravaia labeled
test,19 though he does not state how he tested the
the elastic aspect of cane with what he called recovery. He flexed a piece of cane for five minutes and
hardness of the cane, on what side (epidermis or in-

ARTICLES

102

WHAT MAKES A GOOD REED? AN EXPLANATION OF FLEXIBILITY, DENSITY AND HARDNESS OF ARUNDO DONAX L. PART III.

measured its ability to return to its original shape.

is an important requirement that the test pieces be
To test the stiffness of his cane Intravaia clamped a
closely isomorphic; that is, of identical length, width,
gouged piece of cane at one end in a horizontal posithickness and curvature. Otherwise results will be
tion and applied a downward force, bending the cane
skewed by unintended variables. For my study each
and measuring the limit to which it bent. This is a
piece of cane was tested while wet directly after the
similar idea to that of the flexter machine (described
gouging process, meaning the cane had been soaking
below). The results of his test were that, generally, the
for a minimum of thirty-six hours. It was also tested
rank of stiffness was similar when tested both wet
when completely dry. It is not surprising that cane is
and dry. Interestingly, he found an exception: cergenerally less flexible when dry than wet. However,
tain pieces tested less flexible when wet than when
this is not always the case. A few pieces of cane tested
dry, even though all were obtained from the same
equally flexible both wet and dry. Also the most flexsource.23 In my research I only came across such
ible piece of wet cane of the 200 pieces tested was not
the most flexible piece of that particular batch when
results in two pieces of cane, they were slightly less
dry. The difference between the wet test and the dry
flexible when tested wet than dry. Both were very soft
test remained similar between all four harvests. The
pieces of cane that ultimately made poor reeds.
interesting result of the flexibility test was also beStiffness is a desirable trait in reed cane that retween the two harvests H2004 and H2006. When
sults from a scientific anatomical feature: the fiber
tested wet the 2004 harvest was the most flexible with
sheathing of the vascular cell bundles in the inner
an average of 39.69 . In contrast, and this is what I
cortex.24 Kopp also confirms that this anatomical
find interesting, when tested dry the 2004 harvest
feature has been linked to specific agricultural pracwas one of the two least flexible harvests; average
tices, which include the spacing of the crop, irriga32.92 , and the 2006 harvest was the most flexible
tion, fertilization, climate, the age of the stem and the
when tested dry, an average of 34.03. You may also
age of the internode within the stem.25 My research
remember that H2004 was overall the hardest haralso verifies this.
vest of cane while H2006 was the softest.
Double reed makers tend to focus on hardness
and/or flexibility while many
scientific studies have focused
39.8
on stiffness. The three con39.6
cepts are separable but equally
39.4
indicative of the quality of the
39.2
cane. However, it must be noted
Flexibility Wet
39
that the preference of each indi38.8
vidual are unique. This is dem38.6
38.4
onstrated when comparing the
38.2
results referenced in James Poes
H07
H06
H05
H04
2003 article with Intravaias.
Poe concluded that cane that
Fig. 10: Comparison of the average flexibility when wet for each harvest.
is softer requires less flexibility
to make a good reed whereas
harder cane requires greater
flexibility.26 I agree with this
34.5
assessment. He also stated that
34
as the hardness of cane decreas33.5
es, the flexibility increases, but
33
softer cane can be found with
32.5
Flexibility Dry
less flexibility.27 Again, my find32
ings agree with this statement.
31.5
Intravaias conclusions were the
31
30.5
exact opposite, Cane may be
H07
H06
H05
H04
discarded if it tests either stiff
and soft or hard and flexible.28
Fig. 11: Comparison of the average flexibility when dry for each harvest.
For all of the above tests it

THE DOUBLE REED

103

The physical characteristics that compose a favorable

H2007
H2006
H2005
H2004
reed cannot be attributed to
Hardness-Dry
0.1752
0.1998
0.1882
0.1628
one single characteristic but
to a combination of all three
Hardness-Wet
0.3032
0.3366
0.315
0.2752
characteristics
tested: flexFlexibility-Dry
31.88
34.03
33.11
32.93
ibility, hardness and density.
Flexibility-Wet
38.87
39.226
38.752
39.69
The most significant correDensity
74.78%
69.20%
73.09%
75.50%
lations found were between
the reeds that were rated
Table 1: Cane test results; averages for each harvest.
by their hardness when wet
and the reeds sorted by their
As mentioned above I found a direct correlaflexibility when dry. I prefer reeds in the range of
tion between the density and the hardness of the
0.27-0.33 hardness (tested when wet), and for flexcane tested. H2004 is exceptional in the results of
ibility when dry a range of 26.7-31.1. However, a
hardness. This harvest was found to be the hardest
reed found favorable in both of those characteristics
of all four harvests with a 20% difference from the
would not have a preferred tone quality or response
Cane tests - Averages

Back to Table of Contents

ARTICLES

The Flexter machine is an original invention by

least hard (H2006), and a 7% difference in hardness
James Poe, a retired automotive engineer and basfrom H2007, the second hardest harvest. H2004 had
soonist. In his article Cane Hardness and Flexian average hardness of 0.275 when wet and 0.1628
bilty,29 Poe described a machine he had invented
when tested dry. H2004 was also the most dense of
to measure flexibility in a gouged piece of bassoon
the four harvests by 8.7% from the least dense, which
cane. Mr. Poe has since worked with Udo Heng of
was, again, H2006. This is a substantial difference in
Reeds n Stuff to create a digital model of his invendensity between these two harvests. It is also a logical
tion. The machine enables the tester to measure the
combination from a botanical view to find extreme
flexibility of a piece of gouged cane by securing the
hardness and high density in H2004. The greatest
cane on both ends. The machine has a simple layout:
contrast between harvests occurs between H2004
attached to a wooden base, are two vertical posts
and the H2006 in hardness, density and flexibility
with horizontal rods and clamps that are rounded
(see fig. 16).
for easy fit of the cane. The left post
and horizontal rod are stationary
while the right posts rod rotates.
H2006 and H2004 Comparison
The right post also has a battery operated digital counter at the top of
80
the post to give an accurate display
70
of the degree of flexibility. Once the
60
H06
piece of cane is secured on one end
50
the opposite end is clamped in a
H04
40
headstock which is allowed to rotate
30
when the fixed load is applied. The
fixed weight is a one pound weight
20
that is placed on a measuring hook
10
just below the counter on the right
0
post. The same weight is placed on
/HYKULZZ /HYKULZZ -SL_PIPSP[` -SL_PIPSP[` +LUZP[`
the hook for each test, thus giving
>L[
+Y`
>L[
+Y`
an equal and controlled twist to
the cane. The cane is flexed and the
Fig. 16: Comparison of physical characteristics tested between
counter displays the exact degree of
H2006 and H2004.
the twist. The Flexter machine used
in my study was purchased from
Reeds n Stuff.

ARTICLES

104

WHAT MAKES A GOOD REED? AN EXPLANATION OF FLEXIBILITY, DENSITY AND HARDNESS OF ARUNDO DONAX L. PART III.

if it also had too much or too little density. The presence of all three of these favorable characteristics is
necessary for a good performance reed. While the
numerical results of this test may seem to show small
differences, as I will demonstrate in a future article,
when the same cane was applied in a test of actual
reed performance, the results were quite measurable.
I have found, as with many circumstances, that
any extremes in cane are not optimum. For example
if a piece of cane is found to be very hard yet has high
flexibility then it will be an unbalanced reed. However, a piece of cane that is marginally considered
hard but has greater flexibility, this could produce a
nicely balanced reed. Again, the preference of each
individual reed maker is what matters, and finding
your own personal preference and making reeds constantly out of cane that meets your preferred criteria
will give you greater percentages of the perfect reed.
As I will demonstrate in the next article, these characteristics are directly correlated to the climate in
which A. donax is grown. I will also propose what
particular aspects of the climate during cultivation
have the greatest affect on different characteristics of
the cane. If we could purchase cane with the knowledge of the climate the harvest experienced, we could
potentially have more consistency in the cane we buy
and therefore, in each reed we make. X

ENDNOTES
1

3
4
5

6
7

Edwin Lacy, Testing the Density or Specific

Gravity of Bassoon Cane, The Double Reed,
vol.24 no. 4 (2001): 45.
James B. Kopp, Counting the Virtues of Bassoon
Reed Cane, Journal of the International Double
Reed Society, vol. 26, no. 4 (2004): 6.
Ibid., 16.
Jess Stein, ed., Random House College Dictionary
(New York: Random House, Inc., 1981): 602.
Karl Almenrder, On the Making of Bassoon
Reeds, Journal of the International Double Reed
Society, 8 (1980), 23 (trans. Ester Froese from
Carl Almenrder, Die Kunst des Fagottblasens
[Mainz: Scott, 1842]).
Kopp, Virtues of Bassoon Reed Cane, 57.
Donald Jay Cassadonte The Clarinet Reed: an Introduction to its biology, chemistry and physics
(DMA diss., The Ohio State University, 1995), 22.
John W. Reid, Cane Selectivity from the Field to
the Gouger, Journal of the International Double
Reed Society, no. 11 (1983) 18.

10
11
12
13

14
15
16
17

18
19
20
21
22
23
24
25
26

27
28
29

Lawrence J. Intravaia The Effects of Hardness

and Stiffness of Bassoon cane upon Performance
of the Reed, Journal of the International Double
Reed Society, no. 6 (1978) 17.
Ibid., 4.
Kopp, Virtues of Bassoon Reed Cane, 14.
Stein, Random House College Dictionary, 1262.
Edwin V. Lacy, Testing the Density or Specific
Gravity of Bassoon Cane, The Double Reed, vol.
24 no. 4 (2001) 45.
Stein, Random House College Dictionary, 355.
Lacy, Testing the Density, 45. Kopp, Virtues of
Bassoon Reed Cane, 16.
Kopp, Virtues of Bassoon Reed Cane, 16.
Clare Elizabeth Lawton, A Study of Variation
in the quality of Oboe Reeds made from Arundo
donax L. Gramineae, (PhD. Diss. University of
Reading, 1998), 141.
Lacy, Testing the Density, 45.
Lacy, Testing the Density, 46.
Intravaia, Effects of Hardness, 6.
Stein, Random House College Dictionary, 1123.
Ibid., 424.
Intravaia, Effects of hardness, 17.
Kopp, Virtues of Bassoon Reed Cane, 18.
Ibid.
James M. Poe, Cane Hardness and Flexibility:
Related Measurements Leading to better Bassoon
Reeds, The Double Reed, vol. 26 no. 2 (2003) 63.
Ibid., 62.
Intravaia, Effects of Hardness, 45.
Poe, Cane Hardness and Flexibility, 63.