Published October 19, 2015
Research
Corn Stalk Lodging: A Forensic
Engineering Approach Provides Insights
into Failure Patterns and Mechanisms
Daniel J. Robertson, Margaret Julias, Brian W. Gardunia, Ty Barten, and Douglas D. Cook*
Abstract
Stalk lodging is essentially a structural failure. It
was therefore hypothesized that application of
structural and forensic engineering principles
would provide novel insights into the problem
of late-season stalk lodging of maize (Zea mays
L.). This study presents results from a structural
engineering failure analysis of corn stalk lodging,
involving detailed inspection and measurements
of lodged stalks and a multidimensional imaging
study to assess stalk architecture based on structural engineering principles. This work involved infield observation of >20 varieties of lodged corn
stalk in eight international locations and detailed
geometric analysis of four varieties. Analysis of
collected data revealed very strong, yet previously
unreported, patterns in corn stalk lodging. Corn
stalks predominantly fail (break) by creasing, fall
in the direction of the minor diameter of the cross
section, and break within 4 cm of a node. These
failure patterns, across a broad sampling of varieties and environments, suggest a consistent weakness in maize stalk architecture, indicating that a
common solution might be identified to strengthen
maize stalks. Structural engineering analysis of
stalk architecture and morphology revealed that
several geometric stress concentrators (features
known from engineering theory to increase local
stresses) occur in the predominant failure region of
corn stalk. Identified stress concentrators include
surface irregularities, sharp changes in diameter,
and voids occurring in the stalk pith. Each of these
stalk features persist across different international
locations, environmental conditions, and hybrid
varieties. These findings support the use of new
selective breeding approaches that focus on stalk
morphology and structural engineering analysis of
corn stalk architecture to develop lodging resistant varieties of maize.
D. Robertson, M. Julias, and D.D. Cook, Dep. of Mechanical Engineering,
New York Univ–Abu Dhabi, PO BOX 129188, Abu Dhabi, United Arab
Emirates; B, Gardunia, and T. Barten, Monsanto Corporation, 1551
Highway 210, Huxley, Iowa. Received: 8 Jan. 2015. Accepted 22 Apr.
2015. *Corresponding author ([email protected]).
Abbreviations: CT, computed tomography.
A
lmost 100 yr ago, Garber and Olson (1919) observed that
“lodging in cereals is dependent on so many factors of
unequal value in the different sorts that no one factor seems to
be correlated closely enough with lodging to be of much value
as a selection index.” Lodging has historically been measured by
visual counting of lodged stalks at harvest. Unfortunately, this
approach is notoriously unreliable because it is confounded by
several environmental factors, including disease, pest damage,
wind, and rain (Flint-Garcia et al., 2003a).
In regards to maize, three-quarters of a century ago, Hunter
and Dalbey (1937) looked forward to a time when laboratory or
other controlled methods could provide more reliable data on
lodging. Since then, many approaches to stalk lodging have been
explored. Some of these have included bending tests, rind penetration, and anatomical measurements (Davis and Crane, 1976;
Martin and Russell, 1984; Zuber et al., 1980). However, the
genetic relationships for these methods have proven to be weak to
moderate (Flint-Garcia et al., 2003a,b,c; Peiffer et al., 2013; Hu
et al., 2013). After nearly 100 yr of research on lodging, visual
counts of lodging are still the primary method used to quantify
lodging resistance (Hu et al., 2013; Lian et al., 2014).
Lodging is clearly a highly complex, elusive problem. Perhaps
one reason for the lack of progress has been the fact that this
Published in Crop Sci. 55:2833–2841 (2015).
doi: 10.2135/cropsci2015.01.0010
Freely available online through the author-supported open-access option.
© Crop Science Society of America | 5585 Guilford Rd., Madison, WI 53711 USA
All rights reserved.
crop science, vol. 55, november– december 2015 www.crops.org2833
problem has primarily been investigated from a biological
perspective. Because lodging is fundamentally a structural
issue, we hypothesized that the principles, skills, and techniques of forensic and structural engineering might provide unique insights and suggest new solution approaches
for this stubborn problem.
The primary purpose of forensic engineering is to
determine causes of accidents, device failures, or other
unintended events. Forensic engineering principles could
therefore potentially identify new patterns or features that
are not only correlated with stalk lodging, but that are
causally related to stalk strength or weakness. A forensic
engineering analysis of lodged corn stalk was therefore
conducted, including (i) an observational study of naturally
lodged corn stalks in the field, and (ii) a detailed assessment
of the corn stalk architecture from a structural engineering
perspective. To our knowledge, this is the first time that
this approach has been applied to the study of crop lodging.
Several goals and objectives were identified at the onset
of this study. First, we wished to assess the value of applying forensic engineering to the problem of stalk lodging.
Second, we sought to identify patterns (if any) in lodged
corn stalks (e.g., modes of failure, location of failure, etc.).
Third, we sought to identify features of the stalk that
potentially influence lodging. Identification of these features could prove to be valuable for future breeding studies.
METHODS
This study consists of two distinct components. The first component was an observational study to obtain new information
on the failure patterns of corn stalks in their natural environment. The purpose of this first component was to produce a
comprehensive description of corn stalk failure, including types
of failure (and rates for each type), locations of failure, and other
important information.
The second component of this study involves an assessment
and analysis of the corn stalk structure (especially the corn stalk
geometry) based on engineering principles and expertise. Several
imaging modalities were used to collect relevant information
on the corn stalk geometry. Informed by novel failure data and
geometric information, engineering theory and expertise were
applied to generate new hypotheses of corn stalk failure that are
consistent with each of these sources of information.
Observational Study of Naturally
Lodged Stalks
Forensic engineering failure analysis begins with detailed
examinations of failed specimens. Naturally lodged corn
stalks were examined at several locations in Iowa during the
fall of 2010 to determine the range of possible failure modes
(or types) and to design a methodology for collecting relevant
data. This methodology was then used to collect lodging data
at four locations in Iowa during the 2012 growing season (fields
near Shelby, Ellsworth, Boone, and Williams, IA) and at four
locations in South Africa (Rigdar, Migdol, Viljoenskroon,
2834
Table 1. Environmental conditions in Iowa locations, 2012
growing season.
Location
Soil type†
Boone
Ellsworth
Shelby
Williams
Nicollet–Clarion loam
Nicollet loam
Marshall silty clay loam
Canisteo silty clay loam
†
Average
Rain
rainfall
Average
events
per
temperature of more
month
(high, low) >25 mm
mm
55.7
46.4
65.3
54.9
C
28.6, 14.1
28.1, 13.7
28.3, 13.8
27.9, 13.5
2
1
2
1
Nicollet loam (fine-loamy, mixed, superactive, mesic Aquic Hapludolls); Clarion
loam (fine-loamy, mixed, superactive, mesic Typic Hapludolls; Marshall silty clay
loam (fine-silty, mixed, superactive, mesic Typic Hapludolls); Canisteo silty clay
loam (fine-loamy, mixed, superactive, calcareous, mesic Typic Endoaquolls).
Table 2. Environmental conditions in South Africa locations,
2014 growing season.
Location
Migdol
Rammalutzii
Rigdar
Viljoenskroon
Average
rainfall per
month
Average
temperature
(high, low)
Rain events
of more
>25 mm
mm
44.8
–
–
48.1
C
28.7, 15.0
–
–
27.9, 14.7
3
–
–
0
and Rammalutzii, SA) during the 2013–2014 growing season.
Environmental conditions from those sites are summarized in
Tables 1 and 2. Daily weather data was measured in all four
Iowa locations. Weather data for South Africa locations was
only available in Migdol and Viljoenskroon. Soil classifications
were not taken from South African locations. Drought conditions were prevalent across all four Iowa locations in 2012
with higher temperatures and reduced rainfall being recorded.
For example, in 2011 in Boone, IA, average rainfall during the
growing season was 109.4 mm mo−1, while in 2012 it was only
55.7 mm. In 2011 there were three large rain events >25 mm
in Boone, IA, but only two in 2012. In 2011, Williams, IA, had
six strong storms but only one in 2012. Even though drought
conditions persisted throughout the growing season, timely
rains helped produce average yields (13.9–15.7 m 3 ha−1).
To reduce the number of potential confounding factors,
only stalks with no visible presence of disease or pest damage
were included in this study. Additionally, the study focused only
on late-season stalk lodging, which occurs between physiological maturity of the grain and the time of harvest. Because the
purpose of this research was to identify trends affecting corn
in general (not specific varieties), a broad sampling approach
was employed that included over 20 commercial and precommercial varieties of dent corn and sweet corn. Data recorded
for lodged stalk specimens included distance from the site of
tissue failure to the nearest node line, the internode section in
which the failure occurred, descriptive photographs of failure
patterns, presence of disease or pest damage, a written description of the failure region, etc.
www.crops.org
crop science, vol. 55, november– december 2015
Figure 1. Representative photographs of the three naturally occurring stalk failure patterns observed in this study.
Engineering Analysis of Corn
Stalk Architecture
Structural analysis of corn stalk architecture was performed on
intact (unbroken) corn stalk specimens. Sampling for this component of the research was distinct from the observational analysis
described above. Stalk samples were collected immediately before
harvest at Monsanto testing sites near Greenville, IA, in 2011.
Stalks were sampled from two replicates of four commercially
available DeKalb hybrids seeded at four plant densities (74, 89,
104, and 119 1000-plants ha−1) in a completely randomized design.
To prevent spoilage and increase shelf life of the specimens, stalks
were placed on a forced-air dryer for 2 d to reduce stalk moisture to a stable level (between 10 and 13%). All specimens were
evaluated for disease and insect damage; only fully intact, uncompromised stalk sections were included in this study.
Analysis included laboratory inspection and measurement
of stalk tissues using high-resolution X-ray computed tomography (CT) scanning, stereomicroscopy, and scanning electron
microscopy. The CT scans were performed on stalks using an
X5000 scanner (NorthStar Imaging). Sets of five stalks were
scanned together at 7.5 frames per second with each scan consisting of 1200, two-dimensional radiographs. Scan data was
calibrated using the density of two known reference objects.
Reconstructed three-dimensional scans had a spatial resolution
of 90 µm voxel−1 and a 16-bit intensity range. At this resolution,
individual vesicles within each stalk are clearly visible.
Geometric analysis of CT scan data was performed using
custom computer software developed for this project. This
software automatically identifies the location of each node in
the digital data set and then uses the nodal locations as reference points for the measurement of numerous geometric and
structural engineering attributes including interior and exterior
boundaries, rind thickness, maximum and minimum diameter
of each cross section, eccentricity of the cross-sectional area,
moment of inertia, etc.
Physical examination of corn stalk samples revealed that
voids (i.e., holes or cracks) sometimes occur within the pith.
Manual inspection of CT data was used to determine the presence and location of voids in scanned stalks. A manual approach
was chosen because of the low contrast between the void space
and low-density pith tissue, which limits the effectiveness of
automatic edge-detection algorithms. This manual analysis
produced binary data (presence vs. absence of visible voids) at 10
regularly spaced sample points along each internode section of
each stalk. Stalks were inspected using both a stereo microscope
crop science, vol. 55, november– december 2015 and a scanning electron microscope to observe corn stalk cross
sections and tissue organization.
Finally, geometric information was combined with failure
pattern data to form hypotheses of stalk failure that were consistent
with principles of structural failure from the field of engineering.
RESULTS
In-Field Observational Study
Manual inspection of naturally lodged corn stalk specimens revealed that stalks typically broke in one of three
ways (i.e., three primary failure modes). These failure
modes included snapping, splitting, and creasing. Photographs of each failure mode are shown in Fig. 1.
Overall, 91% of specimens failed as a result of creasing. Stalks that fail in this mode typically display either
one or two creases, which are oriented perpendicular to
the apical-basal axis of the stalk. Splitting of stalks appears
to primarily be a secondary failure mode (i.e., creasing
occurs first and, as the crease becomes more pronounced,
splitting of the rind occurs). This explanation of failure
progression is supported by two observations: (i) manual
straightening of failed stalks results in the closure of split
tissue, and (ii) laboratory three-point bending experiments (Robertson et al., 2014) also produce splitting as
a secondary failure following creasing. Snapping failures
were the second most common failure mode, occurring
in 7% of examined specimens. Splitting was rarely seen
in the absence of creasing (2% of examined specimens).
These results are summarized in Fig. 2. Confidence intervals were calculated by applying a standard interval estimation approach (Brown et al., 2001). To calculate confidence intervals proportions were reduced to two: creasing
and noncreasing. The resulting 99% confidence intervals
are provided in Fig. 2.
Crease failures are closely related to plant physiology.
Creases were generally aligned with the major diameter of
the stalk cross section and the plants generally fell in the
direction of the minor stalk diameter. The direction of primary failure (direction in which the plant fell) was determined using a stalk-based angular coordinate system, with
the origin at the center of the stalk cross section and the
www.crops.org2835
Figure 2. Distributions of failure types and failure directions. Failure direction is defined relative to the leaf groove of each stalk, with the
leaf groove as 0. Due to symmetry of the stalk about the 0 and 90 lines, all angles are presented between 0 and 90. Failure direction
data collected at four South Africa locations only (hence a reduced sample size). Whiskers represent 99% confidence intervals of the
reported percentages.
Figure 3. Histogram of failure locations relative to the closest node. The node line is represented by the vertical dashed line. Histogram
bins have a width of 0.5 cm.
zero-degree direction in the direction of the leaf groove.
Using the (approximate) symmetry of the stalk about the
0 and 90 directions, only angles between 0 and 90 were
recorded (i.e., an angle of 135 and 180 were recorded as
45 and 0 respectively). Failure angles were grouped into
bins centered at 0, 45, and 90, as shown in Fig. 2. The
predominant direction of failure was in the 90 direction,
which is the direction of the minor diameter of the corn
stalk. Overall, 85% of stalks failed in this direction.
Engineering bending analysis of the typical stalk cross
section predicts that stalks are stiffer and stronger when bent
in the direction of the major cross-sectional dimension, but
weaker when bent in the minor diameter direction. While
2836
further research is needed on this subject, these differences
in stiffness and strength likely explain the propensity of
stalk failure in the direction of the minor diameter.
Finally, the primary failure location of lodged stalks is
also closely related to stalk physiology. Failure most commonly occurs in close proximity to the node and the meristematic tissue immediately above the node. As shown in
Fig. 3, the distribution of failure location is centered about 1
cm above the node, with failure frequency rapidly decreasing with distance from this point. Overall, 89% of specimens failed within 3 cm of the distribution center. Failure
was observed less frequently just below the node, where the
stalk exhibits dense, tough tissue and a thick rind.
www.crops.org
crop science, vol. 55, november– december 2015
Figure 4. Stereo (top) and scanning electron microscope (bottom) images illustrating differences between internodal and nodal tissue
organization. Internodal tissues (right) are highly organized and smooth in appearance while nodal tissues are irregular. Surface
irregularities occurring near the node including epidermal wrinkling and divots from undeveloped brace roots can act as engineering
stress concentrators thereby lowering the bending strength of corn stalk.
Engineering Analysis
Close inspection of stalk morphology from the macro to
micro scales reveals several geometric features that may act
as stress concentrators. A stress concentrator is a material
or geometric feature of a structure that results in increased
structural stresses. Because failure always results from
stresses that exceed the ultimate strength of the associated
material, stress concentrators are often closely related to
failure. For example, the small holes in perforated notepad
paper act as stress concentrators to ensure that the paper
rips along a given line. Several stress concentrators of corn
stalk were identified in this study and are described below.
Maximum stresses always occur on the outermost surfaces of structures that are loaded in bending. Gross observations of the exterior surface of corn stalks were therefore
performed. These observations revealed that internodal
tissues are smoother, more organized, and more regular than tissues near the node. Basal nodes, in particular,
often exhibit divots (crater like depressions in the stalk)
that are associated with undeveloped brace roots. A significant amount of wrinkling in the epidermal tissues was also
crop science, vol. 55, november– december 2015 observed to occur near nodes. Figure 4 consists of images
obtained from stereo and scanning electron microscopes
that illustrate the differences between internodal and nodal
tissue organization. Divots and wrinkling occurring near
the node were found to be nonuniform in terms of shape
and size. Surface irregularities such as these typically act as
stress concentrators and can substantially increase the propensity of structural buckling (Simitses, 1986).
Sudden changes in any geometric feature can act as a
stress concentrator. The CT scan data was therefore analyzed to quantify sudden geometric changes in corn stalk.
Analysis of this data revealed that major changes in virtually all geometric features occur in and around the nodal
and meristematic tissue (i.e., the predominant failure
region of naturally lodged corn stalk). Figure 5 displays
graphs of several geometric features of a representative
stalk which demonstrate this trend. Such patterns in stalk
morphology and geometry were typical of all maize stalks
investigated in the study.
Cross-sectional changes in diameter, such as those
observed to occur at the node of corn stalk (Fig. 6), are some
www.crops.org2837
Figure 5. Geometric features of corn stalks undergo dramatic changes in the nodal and intercalary meristem regions. Several geometric
features measured in the current study are plotted above for a typical stalk. These changes can act as engineering stress concentrators.
Tick marks on the horizontal axis indicate nodes. A computed tomography image of the corresponding stalk is shown at the bottom of
the figure for physiological reference.
of the most well-known stress concentrators (Shigley et al.,
2003). The average reduction (and standard deviation) in
major and minor diameters at the node was 13.8% (4.1%)
and 10.2% (6.5%), respectively. More gradual changes in
diameter were also observed in the intercalary meristem
region. As seen in Fig. 5, the diameter typically tapers just
after the node line and then becomes relatively constant
in the internodal region. While the changes in diameter
occurring in the intercalary meristem are less drastic than
those at the node, they can still act as stress concentrators.
Finally, we observed drastic changes in rind thickness
and tissue density near the node. Rind tissue is thinnest
in the internodal region, with thickness increasing rapidly
near the nodes. At the node itself, corn stalk is nearly solid.
The area moment of inertia—a quantity that is related to
structural stiffness under bending as well as stress distribution—was also observed to change rapidly near each node.
2838
The CT scans revealed that virtually all (99%) of
the stalks had at least one sizeable void in the pith tissue.
Examples of such voids are shown in Fig. 7. Rapid drying
during senescence may cause the pith tissue (parenchyma)
to crack or split as moisture leaves the stalk too quickly,
similar to the splitting that occurs when wood dries rapidly. The presence of sharp corners and uneven pith tissue
distribution can act as stress concentrators.
DISCUSSION
One goal of the current study was to catalog failure modes
and patterns of naturally lodged corn stalk. Consistent failure patterns suggest a consistent structural weakness or a
design flaw. Well-designed structures do not have a consistent weakness and therefore they either (i) do not fail at all,
or (ii) they fail at a broad distribution of points and in a variety of failure modalities. In the latter case, failure is caused
www.crops.org
crop science, vol. 55, november– december 2015
Figure 6. Sudden cross-sectional changes in geometry are observed to occur at the node line in corn stalk. Such changes are well known
engineering stress concentrators. These changes artificially increase local stresses and likely lower the structural bending strength of
corn stalk.
Figure 7. Computed tomography scanning (left) revealed that pith voids (i.e., holes or cracks) occurred in nearly every stalk investigated
in the current study. A photograph of one such void is shown at right. These voids are thought to act as stress concentrators, artificially
increasing local stresses, and propensity of rind buckling.
by randomly distributed local imperfections in material or
geometry, not by a systemic weakness in the design.
The application of forensic engineering techniques
revealed strong patterns of corn stalk failure in failure
modes, direction, and failure location. Considering the large
amount of genetic and environmental diversity inherit in
the results (eight international locations, >20 varieties, and
multiple years), the consistency of failure patterns was quite
surprising. We found that stalk failure almost always occurs
within a few centimeters of a node, involves a crease in
the direction of the major diameter, and results in the stalk
falling predominantly in the direction of the minor diameter. Failures away from the node are very rare in healthy,
intact corn stalks, even though rind tissue is thinnest in the
crop science, vol. 55, november– december 2015 internodal regions. Such consistent failure patterns indicate
that corn stalk possesses a systemic structural weakness.
Furthermore the fact that failures occur in regions of high
geometric variation imply a causal relationship between
geometric stress concentrators and failure that is strongly
supported by engineering theory.
Three predominant stress concentrators (pith voids,
surface irregularities, and sharp changes in diameter) are
colocated with the predominant failure region of naturally
lodged corn stalk. These features appear to be common
in many different corn varieties regardless of corn type,
growing location, or planting densities. Mechanical threepoint bending tests conducted in our lab have revealed
that failure frequently initiates at or near one of these stress
www.crops.org2839
concentrators (Robertson et al., 2014). A brief discussion
of each stress concentrator, including its potential effects
on stalk bending strength, is given below.
It is well accepted in engineering theory and practice
that hollow or foam filled tubes are particularly susceptible
to surface imperfections such as the divots seen on the left of
Fig. 4 (Simitses, 1986). Geometric imperfections of approximately the same size as undeveloped brace root divots have
been reported to decrease the buckling load of anisotropic,
hollow tubes by up to 50% (Tennyson, 1968). However, to
our knowledge, the scientific literature on plant physiology does not currently contain any mention of how surface
irregularities may affect structural strength of plant stems.
While numerous engineering studies have addressed
stress concentrations as a result of changes in diameter, a
review of the corn and plant literature revealed no mention of this topic. The equations of Shigley et al. (2003)
were used to estimate the effect of diameter changes (see
Fig. 6) on corn stalk bending strength. These calculations,
based on an isotropic rod in pure bending with the average cross-sectional parameters reported in this study, indicate a 300% increase in stress near the node. The anisotropic material properties of corn stalk likely reduce this
stress concentration, and further research is required to
achieve a more precise prediction. Thus, while nodes are
known to increase stalk strength by preventing cross-sectional ovalization (i.e., microbuckling) (Niklas, 1992), the
sharp changes in cross-sectional geometry occurring near
the node can also act as a stress concentrator. Minimizing
or smoothing the sharp changes in geometry that occur
near the node might be one avenue for improving stalk
strength. Indeed, a recent computational study (Von
Forell et al., 2015) has shown that reducing the geometric
variation of the surface of corn stalks (i.e., smoothing the
surface geometry) causes reductions in stress levels when
all other factors are held constant.
The pith of corn stalk is known to act as a mechanical brace thus increasing the bending strength of the stalk
(Niklas, 1992; Lu et al., 1987). It prevents the stalk cross
section from ovalizing (Niklas, 1998) and resists inward
buckling of the rind tissue. Voids in the pith tissue inhibit
it from performing these mechanical functions. Surgical
destruction of transverse diaphragms of grass stems that
serve the same mechanical purpose as the pith tissue (i.e.,
transverse braces) has been shown to reduce bending
strength by up to 20% (Niklas, 1992). Similar results have
also been shown in foam-filled, thin-walled tubes constructed from typical engineering materials (Kim et al.,
2004; Reid et al., 1986; Chen et al., 2002).
Rather than being a stress concentrator, the increase
in rind thickness occurring near the node is likely nature’s
approach to handling greater stresses that arise near the
node. By increasing rind thickness, more tissue is available on which to distribute bending forces. However, since
2840
failures are highly concentrated near the node, the effect of
increased rind thickness appears to be insufficient to overcome the stress concentrations imposed by other features.
Future studies involving structural finite element modeling of corn stalk may provide more insight on the role of
the rind thickness on stress distributions near the node.
Limitations
Observational studies such as this one are always limited to
observable (and often uncontrollable) events. Stalk lodging is often associated with strong wind and rain storms
that strike during the senescence period. Particularly
strong storms occasionally flatten entire crop fields.
This study relied on natural occurrences of stalk lodging, which could potentially bias the study because the
fallen stalks were perhaps unnaturally weak, malformed,
or affected by the undetected presence of disease or pests.
On the other hand, the patterns revealed in this study
persist across many types and varieties of corn, growing
locations, growing seasons, and climate, suggesting that
plant architecture plays a central role in stalk strength and
failure. Since initially reporting these results, several crop
scientists have contacted the authors and confirmed the
failure patterns reported in this study.
Finally, while the stress concentrators identified in this
study are colocated with failure occurrences, and similar
geometric features are known in the engineering literature
to be stress concentrators, we must always bear in mind
that correlation does not imply causality. Engineering
theory provides compelling explanations for how geometric features near stalk nodes could affect (and perhaps
cause) failure. More focused research will be needed test
and confirm or disprove these hypotheses. If confirmed,
these geometric features would enable new approaches for
improving stalk strength through selective breeding.
CONCLUSION
Corn stalk lodging exhibits strong patterns of failure
modes (predominantly creasing), failure orientation (predominantly aligned with the major diameter of the stalk),
and failure location (89% of failures originate in a narrow
region near the node). These results imply that many varieties of corn suffer from a consistent structural weakness.
Several stress concentrators (features known from engineering theory to increase local stresses) are spatially correlated with the predominant failure location of lodged
corn stalk. Identified stress concentrators included pith
voids, surface irregularities, and changes in cross-sectional
geometry. It seems likely that one or more of these factors
contribute to the failure patterns observed in this study
and may influence maize stalk strength.
Forensic engineering has proven to be a useful
approach for investigating lodging of corn and other agricultural crops. In addition, recent engineering analyses of
www.crops.org
crop science, vol. 55, november– december 2015
bending test methodologies (Robertson et al., 2014, 2015)
have shown that many prior studies have used inappropriate techniques for assessing bending strength. This may
explain the weak to moderate connection between bending strength and lodging in prior studies.
The results of this study suggest that stalk strength is
closely related to relatively small and previously unstudied
morphological features. It is conceivable that modification of these features (stress concentrators) would require
minimal diversion of bioenergy and biomass and, therefore, would not have a detrimental effect on yield. This
approach could lead to substantially different results than
previous selective breeding studies, which have almost
exclusively focused on gross internodal geometries and
material properties. In summary, engineering methods
and tools appear to hold great promise in addressing the
problem of stalk lodging.
Acknowledgments
This research was supported by the Core Technology Platform
Resources at New York University Abu Dhabi and funding
from the National Science Foundation, award #1400973.
References
Brown, L.D., T.T. Cai, and A. DasGupta. 2001. Interval estimation
for a binomial proportion. Stat. Sci. 16:101–133.
Chen, W.T., T. Wierzbicki, and S. Santosa. 2002. Bending collapse of thin-walled beams with ultralight filler: Numerical
simulation and weight optimization. Acta Mech. 153:183–206.
doi:10.1007/BF01177451
Davis, S.M., and P.L. Crane. 1976. Recurrent selection for rind
thickness in maize and its relationship with yield, lodging, and
other plant characteristics. Crop Sci. 16:53–55. doi:10.2135/cro
psci1976.0011183X001600010013x
Flint-Garcia, S.A., J. Chaba, L.L. Darrah, and M.D. McMullen.
2003a. Quantitative trait locus analysis of stalk strength in
four maize populations. Crop Sci. 43:13–22. doi:10.2135/cropsci2003.0013
Flint-Garcia, S.A., L.L. Darrah, M.D. McMullen, and B.E. Hibbard.
2003b. Phenotypic versus marker-assisted selection for stalk
strength and second-generation European corn borer resistance
in maize. Theor. Appl. Genet. 107:1331–1336. doi:10.1007/
s00122-003-1387-9
Flint-Garcia, S.A., M.D. McMullen, and L.L. Darrah. 2003c.
Genetic relationship of stalk strength and ear height in maize.
Crop Sci. 43:23–31. doi:10.2135/cropsci2003.0023
Garber, R.J., and P.J. Olson. 1919. A study on the relation of some
morphological characters to lodging in cereals. J. Am. Soc.
Agron. 11:173–186. doi:10.2134/agronj1919.00021962001100
050001x
crop science, vol. 55, november– december 2015 Hu, H., W. Liu, L. Homann, F. Technow, H. Wang, C. Song, S. Li,
A.E. Melchinger, and S. Chen. 2013. QTL mapping of stalk
bending strength in a recombinant inbred line maize population. Theor. Appl. Genet. 126:2257–2266. doi:10.1007/s00122013-2132-7
Hunter, J.W., and N.E. Dalbey. 1937. A histological study of stalkbreaking in maize. Am. J. Bot. 24:492–494. doi:10.2307/2437068
Kim, A.K., S.S. Cheon, M.A. Hasan, and S.S. Cho. 2004. Bending
behavior of thin-walled cylindrical tube filled with aluminum
alloy foam. Key Eng. Mater. 270:46–51. doi:10.4028/www.scientific.net/KEM.270-273.46
Lian, L., A. Jacobson, S. Zhong, and R. Bernardo. 2014. Genomewide
prediction accuracy within 969 maize biparental population.
Crop Sci. 54:1514–1522. doi:10.2135/cropsci2013.12.0856
Lu, R.F., J.A. Bartsch, and A. Ruina. 1987. Structural stability of
the corn stalk. ASAE. Paper No. 87-6066:1–16.
Martin, M.J., and W.A. Russell. 1984. Response of a maize synthetic to recurrent selection for stalk quality. Crop Sci. 24:331–
337. doi:10.2135/cropsci1984.0011183X002400020029x
Niklas, K.J. 1992. Plant biomechanics: An engineering approach to
plant form and function. University of Chicago Press, Chicago.
Niklas, K.J. 1998. Modes of mechanical failure of hollow, septate stems. Ann. Bot. (Oxford, UK) 81:11–21. doi:10.1006/
anbo.1997.0505
Peiffer, J.A., S.A. Flint-Garcia, N. De Leon, M.D. McMullen, S.M.
Kaeppler, and E.S. Buckler. 2013. The genetic architecture of
maize stalk strength. PLoS ONE 8:1–14.
Reid, S., T. Reddy, and M. Gray. 1986. Static and dynamic axial
crushing of foam-filled sheet metal tubes. Int. J. Mech. Sci.
28:295–322. doi:10.1016/0020-7403(86)90043-3
Robertson, D., S. Smith, B. Gardunia, and D. Cook. 2014.
An improved method for accurate phenotyping of corn
stalk strength. Crop Sci. 54:2038–2044. doi:10.2135/cropsci2013.11.0794
Robertson, D., S. Smith, and D.D. Cook. 2015. On measuring the
bending strength of septate grass stems. Am. J. Bot. 102:5–11.
doi:10.3732/ajb.1400183
Shigley, J.E., C.R. Mischke, and R.G. Budynas. 2003. Mechanical
engineering design. 7th ed. McGraw-Hill Higher Education,
New York.
Simitses, G.J. 1986. Buckling and postbuckling of imperfect cylindrical shells: A review. Appl. Mech. Rev. 39:1517–1524.
doi:10.1115/1.3149506
Tennyson, R. 1968. The effects of unreinforced circular cutouts on
the buckling of circular cylindrical shells under axial compression. J. Eng. Ind. 90:541–546. doi:10.1115/1.3604686
Von Forell, G., D. Robertson, S.Y. Lee, and D.D. Cook. 2015.
Preventing lodging in bioenergy crops: A biomechanical analysis of maize stalks suggests a new approach. J. Exp. Bot. (in
press) doi:10.1093/jxb/erv108
Zuber, M.S., T.R. Colbert, and L.L. Darrah. 1980. Effect of recurrent selection for crushing strength on several components of
maize. Crop Sci. 20:711–717. doi:10.2135/cropsci1980.001118
3X002000060009x
www.crops.org2841