вход по аккаунту



код для вставкиСкачать
Analysis of Free Forward Flight of Schistocerca
gregaria Employing Telemetric Transmission of
Muscle Potentials
Fakultät für Biologie, Universität, 78457 Konstanz, Germany
The free flight of mature female Schistocerca gregaria loaded with a transmitter
device has been video-monitored for short periods of time in the laboratory. Several flight parameters, such as body angles to horizontal and to flight path, ascent angle, and flight speed, have been
analysed. These flight parameters and their interrelationships coincide with those revealed during
steady free flight in the field (Baker et al. 1981. J Comp Physiol 141:233–237). Therefore, these short
free flights in the laboratory represent free autonomous flight performance, but offer the possibility
of controlling environmental conditions. During free flight, the firing patterns of four selected powerful depressor and elevator muscles of the fore- and hindwings were obtained by a transmitter device
fixed to the prothorax of the locust. Based on the recruitment of individual motor units, a correlation
of the muscle-specific activity with the aerodynamic flight parameters can be accomplished. Wingbeat frequency is positively correlated to ascent angle and to speed. The recruitment of the motor
units of the forewing muscles shows a correlation with some of the flight parameters, whereas the
hindwing muscles are activated in a rather immutable pattern. This argues for the importance of the
forewings for maneuvering, whereas the hindwings mainly serve to provide the basic power output
in flight. J. Exp. Zool. 284:119–129, 1999. © 1999 Wiley-Liss, Inc.
Locust flight has been extensively analysed for
its aerodynamic and physiological basis. Thus, two
main approaches have been chosen: (1) analysis
of freely flying locusts (e.g., Baker and Cooter, ’79;
Baker et al., ’81; Kutsch and Stevenson, ’81) and
(2) analysis of tethered flight (e.g., Weis-Fogh, ’56;
Gewecke, ’75; Wortmann and Zarnack, ’93). However, even the most sophisticated tethered arrangement (e.g., Baader et al., ’92; Spork and
Preiss, ’93) represents a rather static system missing the dynamics characterizing the natural closed
loop situation, in which the sensory inflow is unimpeded allowing an unbiased sensorimotor integration. Undoubtedly, flight behaviour is influenced by
tethering (Zarnack and Wortmann, ’89). For example, one of the consequences of such an open loop
paradigm (loss of adequate sensory feedback) is that
the wing-beat frequency is lower in tethered compared to free flight (Baker et al., ’81; Kutsch and
Stevenson, ’81). Even a recommendation to study
flight by suspending locusts “freely” by wires
(Wootten and Sawyer, ’54; Möhl, ’88) does not overcome the restriction impeded by the wires’ length.
In a further advance, a telemetric transmitter
mounted to the locust has been developed (Kutsch
et al., ’93) which allows the recording of physiological signals from freely flying animals. This
technique was used here to understand the recruitment of selected muscles (depressor and elevator muscles of both wing pairs) during forward
flight (horizontal, ascent, descent). Therefore, the
classical work of Wilson and Weis-Fogh (’62), then
employing tethered locusts, is extended to freely
flying locusts. For the first time we can correlate
aerodynamic parameters with motor patterns for
a better understanding of the neuronal control
underlying locust flight. This advance allows us
to study dynamic processes underlying natural
Animals and flight conditions
For our experiments, female desert locusts,
Schistocerca gregaria Forskål (adult age 20–25
days) were chosen, selected from a crowded culture (temperature, 42°C; rel. humidity, c. 75%;
light-dark cycle, 12 hr–12 hr). Immediately before
the flight test, the animals were separated in
small perspex cages and warmed to T = 35°C.
Grant sponsor: DFG; Grant number: Ku 240/17-1.
*Correspondence to: W. Kutsch, Fakultät für Biologie, Universität,
78457 Konstanz, Germany.
Received 6 May 1998; Accepted 5 October 1998.
The dimensions of the flight room were: length
10 m, width 5.7 m, height 2.2 m. The room was
illuminated by several rows of light tubes along
the ceiling with a mean light intensity of c. 1400
lux. The room temperature was 28–30°C and relative humidity was 65–70%. Flight activity of the
locust was induced by accelerating them by hand
with a rather constant launching angle (c. 35°)
and an initial speed (v) of c. 4 m/s. (For Schistocerca some other values for take-off velocity have
been published: v = 2.5 m/s [Katz and Gosline,
’93] or v = 3.2 m/s [Bennett-Clark, ’75]; however,
in our lab, take-off velocities, following a jump,
ranged from v = 4.0–4.4 m/s [Fischer, ’94].) A stationary video-camera was placed at 5 m beyond
the start point, with a distance to the opposite
wall of 4 m. To avoid possible wall effects (Stevenson et al., ’95) only forward flights within a corridor of 8 × 2 × 2 m were analysed. The focus width
of the camera covered an area of 2.8 × 2.0 m. The
distance from the start point to that of entering the
camera focus was sufficient to enable free, autonomous flight performance. The locusts could fly 2 m
further ahead after leaving the camera focus.
Telemetric device
The mass of the transmitter device, including
the battery, was m = 0.5 g, which is equivalent to
c. 20% of the locust’s mass. This additional load
does not impede the flight parameters (Fischer et
al., ’97). The transmitter was fixed to the pronotum by a beeswax-resin mixture.
Each wing is driven by about 12 muscles (Snodgrass, ’29; Albrecht, ’53). From these, we recorded
the activity of the powerful direct depressor muscles
(M99, M129) and indirect elevator muscles (M83/
M84 and M113) of the fore- and hindwing, respectively. The electrode (Manganin, Isabellenhütte,
Dillenburg; insulated to the tip, diameter 50 µm)
was inserted into the chosen muscle, and the reference electrode was inserted into the abdominal
haemocoel. Both electrodes were connected to the
modulator part of the transmitter (Kutsch et al.,
’93). The telemetric electromyograms (EMGs) were
detected by a HB9CV aerial and demodulated by
a wide-band receiver system (for details, see
Kutsch et al., ’93).
Data acquisition and analysis
The EMGs were displayed on an ocilloscope and
stored on tape. In parallel, the oscilloscope trace
was filmed by a second video camera. Both the
recordings of the flight (see above) and the oscilloscope camera were synchronously stored online
on a video mastertape. The frame frequency of
both cameras was 50 Hz (shutter opening time 1
ms), giving a frame interval of 20 ms.
The video mastertapes were analysed frame by
frame (Panasonic AG 7380 recorder, Sony Trinitron monitor). The silhouettes of the animals were
successively drawn onto clear overhead films. Using conventional analysis (e.g., Baker and Cooter,
’79; Kutsch and Stevenson, ’81) several flight parameters (see Fig. 1) could be determined. The
locust’s centre of gravity is midpoint between front
and abdominal tip (Weis-Fogh, ’56), which was not
shifted significantly by the transmitter (Fischer,
’94). All flight parameters are considered to be “instantaneous” with respect to the given time interval of 20 ms.
The video mastertapes allowed for the correlation of the instantaneous flight parameters with
EMGs. First of all, the cycle length (or its inverse,
the instantaneous wing-beat frequency, Hz) can
be determined. Given a mean cycle length (tc) of
43 ms (see Results), there are 2 or 3 individual
frames related to each wing-beat cycle; therefore,
for a correlation of the optical data and cycle
length, the instantaneous flight parameters had
to be averaged and associated to the concomitant
muscle cycle. The size of the viewing field allowed
an analysis of up to 15 consecutive wing cycles.
By a precise evaluation of the EMGs, the recruitment of motor units can be assessed, taking
into account the parameters of signal form, relative amplitude, and possible changes in relative
size within a burst (see Wilson and Weis-Fogh,
’62 and Fig. 2).
Flight parameters
The following flight parameters were analysed
(compare with Baker et al., ’81): ascent angle (α,
[°]); body angle to horizontal (ϕ, [°]); body angle
to flight path (δ, [°]); flight speed along the flight
path (v, [m/s]) (see Fig. 1). Altogether, 335 flights
of 75 animals were investigated and the parameters were correlated to each other.
Data analysis and conventional statistical procedures (e.g., Sachs, ’78) were employed. They
were aided by the computer programs Origin 3.0
and SAS for Windows. All data were tested for
normal distribution (P < 0.01) and homogeneity.
Analysis of correlation and linear regression was
based on the least square method. Binominal distribution and their confidence levels concerning
relative occurrence have been determined after
Fig. 1. Determination of flight parameters by analysing
consecutive video-frames. t, cumulative time interval (∆t =
20 ms) of successive frames; S, translatory path of the centre
of gravity, indicated by a black dot; α, ascent angle: angle
between horizontal plane and translatory path of S; ϕ, body
angle to horizontal: angle between horizontal plane and lon-
gitudinal axis of the locust; δ, body angle to flight path, calculated by the difference between ϕ and α; v, flight speed:
instantaneous translatory speed of the centre of gravity within
two frames; SL, silhouette length of the animal, the mean
natural length of the female population is 54 mm.
Clopper and Pearson (’34). Comparisons between
data were carried out by a one-tailed test.
Analysis of flight parameters
Fig. 2. Criteria for the discrimination of motor unit recruitment, derived from telemetric EMGs of M99. AI, AII:
Sequences of a continuous flight showing the activity of the
first unit (1), double discharges of this unit (1,1′), and the
additional recruitment of the second unit (2). For this example, the second unit is seen to either add up to the after
discharge of the first unit (1′/2) or to overlap completely with
the first single spike (1/2, see last potential). B: Sequence
from a different animal, exhibiting constant recruitment of
both motor units (1,2).
The ascent angle (α) varied from –20° to almost
40°, whereas for the body angle to horizontal (ϕ)
we found values spanning from –20 to 60° (Fig.
3A). The ascent angle (α) was positively correlated
with body angle to horizontal (ϕ) (Fig. 3A). On
average, an increase of the ascent angle of 10 degrees was paralleled by a 9 degree increase of ϕ.
During horizontal flight (α ± 0 degrees) the body
angle (ϕ) was in the range of 16 degrees, during
descent negative body angles to horizontal were
observed. To understand whether the position of
the animal’s longitudinal axis changed in relation
to the general flight direction (translatory movement of its centre of gravity), the ascent angle (α)
and body angle to flight path (ϕ) had to be compared (Fig. 3B). The negative regression line indicated that the steeper the animal’s ascent, the
more congruent was its longitudinal axis to the
flown flight path.
A significant relationship was found when com-
suppose that the general flight speed represents a
“scalar” flight parameter independent of the ascent
angle (α). The flight speed (v) of the animals was
in the range of 1.8 to 6.6 m/s (mean ± SD: v = 3.7 ±
1.0 m/s).
Correlation of flight parameters and
motor pattern
Altogether 1,610 wing-beat periods (tc) have
been analysed. The mean cycle length (±SD) was
43.4 ± 3.2 ms. This corresponded to a mean wingbeat frequency of 23 Hz (range: 18.2 to 31.3 Hz).
Analysis of correlation and linear regression
was performed for paired values: tc vs. α and tc
vs. v (Fig. 4). The observed cycle length (tc) was
significantly reduced with increasing ascent (α).
Depending on the regression, the wing-beat frequency (Fig. 4A) increased by c. 0.8 Hz with each
degree of ascent (α). At horizontal flight, wingbeat frequency was in the range of 22 Hz. When
flight speed was related to cycle length (tc), then
a decrease of tc was correlated with a reduction
in speed (Fig. 4B). A recalculation of cycle length
to wing-beat frequency indicated that flight speed
increased by 0.1 m/s when wing-beat frequency
increased by 1 Hz.
Recruitment of muscle units during flight
Fig. 3. Relationship between (A) ascent angle (α) and body
angle to horizontal (ϕ), (B) ascent angle (α) and body angle
to flight path (δ), and (C) ascent angle (α) and flight speed
(v); n = number of samples. All correlation coefficients (r) are
significant (P < 0.05).
paring instantaneous flight speed (v) to the ascent
angle (α) (Fig. 3C). A linear regression indicated a
reduction of the speed by 0.04 m/s when the ascent
angle (α) changed by 10 degrees. Due to this minimal change in speed related with a wide change in
α we assume that the statistically shown relationship is of low importance in free flight; we rather
An interpretation of the relevant EMGs corroborates previous data regarding the number of motor units involved in flight (e.g., Wilson and
Weis-Fogh, ’62; Tyrer and Altman, ’74; Zarnack
and Möhl, ’77). For both depressor muscles (M99,
M129) two (fast) motor units are reported. For the
elevator muscle of the hindwing (M113) only one
unit is apparent. Both forewing elevator muscles
(M83/M84) consist of only one unit, and since a
difference in the recruitment of both muscles was
not seen, the activity of these two elevator muscles
is lumped together.
To allow a correlation of muscle recruitment
with the kinematic flight parameters (see above),
data for the ascent angle (α) were grouped into
the following classes: –15° to –6° → descent, –5°
to 4° → horizontal flight, 5° to 14° → slight ascent, above 15° → steep ascent (see Fig. 6). Since
data analysis of EMGs vs. body angle to horizontal (ϕ) revealed similar trends to that shown for
the ascent angle (α) (see also the positive correlation of α vs. ϕ, Fig. 3A), we did not duplicate our
analysis for ϕ. Flight speed (v) and cycle duration (tc) were also grouped into four and five
classes, respectively, by partitioning the observed
range of data into equal subdivisions (see Figs.
Fig. 4. Relationship between (A) cycle length (tc) and ascent angle (α) and (B) cycle length (tc) and speed (v), accumulated for all muscles. Both flight parameters decrease with
increasing cycle length; n = samples. Both correlation coefficients (r) are significant (P < 0.05).
7, 8). The activation of the different muscle units
as well as their mode of recruitment is given as
their relative occurrence within the relevant
classes. Complete omission of potentials for individual wing-beat cycles within a longer series for
any complete total occurred only rarely, for such
a case an effect upon the instantaneous flight performance was not obvious.
Recruitment of M99 (forewing depressor)
During flight both motor units of this muscle
(also known as subalar muscle) could be recruited
(see Figs. 2, 5A), which, based on their temporal
recruitment, were termed first and second unit.
Firing of these two units depended on the flight
path. During descent flight (instantaneous values
–15°< α <–5°), both motor units were activated
(Fig. 5A) with a slight preponderance of the first
unit (Fig. 6A). Thereby the following modes of activation became obvious with almost equal prob-
Fig. 5. Examples of telemetric EMGs of two forewing
muscles, right side (r): depressor (M99r) (A, B) and elevator
(M83/M84r) (C, D). For both muscles a change of muscle activity is correlated with different ascent angles. A: Recruitment of both motor units, varying from close succession to
complete overlap; flight data: α = –6°, v = 3.6 m/s, f = 24 Hz.
B: Same animal as A, exhibiting only single spikes of one
unit with steep ascent; flight data: α = 16°, v = 3.2 m/s, f =
24 Hz. C: Single firing of one unit of this muscle complex;
flight data: α = 8°, v = 3.5 m/s, f = 22 Hz. D: Same animal as
C, exhibiting an afterdischarge per each cycle with steep ascent; flight data: α = 19°, v = 3.0 m/s, f = 24 Hz.
ability (Fig. 6B): only the first unit was recruited
with a single spike; the first unit discharged
twice; both units fired once whereby both could
summate completely. Occasionally the first unit
showed a double discharge together with a single
activation of the second unit. With a transition
from horizontal to steep flight path (α > 15°), the
relative activation of the second unit diminished
(Fig. 6A). This was paralleled by a reduction in
double discharges of the first unit. Therefore, a
steep ascent was usually characterized by a single
discharge of the first unit of M99 (Figs. 5B, 6B).
There was no significant change in the activation of the forewing subalar muscle in correlation
with flight speed (Fig. 7). For all classes, at least
when above 2.5 m/s, both motor units were recruited, with the typical preponderance of the first
unit (Fig. 7A). Thereby, the mode of activation,
whether employing both motor units or after-discharges of the first unit, did not show a significant change with increasing flight speed (Fig. 7B).
For the lowest instantaneous flight speed class (v
< 2.5 m/s), we registered only a small number of
values (n = 10); therefore, the apparent reduction
Fig. 6. Analysis of motor unit recruitment of the forewing
depressor muscle (M99) in relation to ascent angle (α). A:
Indication of the overall activity of both motor units. B: Individual recruitment of both units. The number (n) of values
per class, the relative occurrence per class and the 95% confidence interval (bars) (P < 0.05) are shown; * indication of
groups which are not significantly different (P > 0.05). (This
presentation of data is valid also for Figs. 7, 8, 10, 11.)
in repetitive discharges for low speeds was of no
statistical significance.
Similarly, no trend was observed in the recruitment of the subalar’s motor units with respect to
cycle length (tc) (Fig. 8). There was a more or less
constant relation in the activity of both units with
a superiority of the first unit (Fig. 8A). The grade
of activation of both units remained almost constant
independent of the cycle length (Fig. 8B). Therefore, wing-beat frequency and grade of subalar recruitment were not related to each other.
Recruitment of M129 (hindwing depressor)
Due to the size of this muscle, also known as
the hindwing subalar muscle, it became difficult
to separate reliably the two known motor units
in a common EMG. We therefore, restricted our
analysis to the recruitment of the rostral unit, be-
Fig. 7. Analysis of motor unit recruitment of M99 in relation to instantaneous flight speed (v). There is a preponderance of the first motor unit, and there is no change of motor
unit recruitment related to a change of speed. For the presentation of data, see Fig. 6.
cause this unit is reported to be more active than
the caudal one (Stevenson, ’88).
The rostral unit fired mostly once per cycle (Fig.
9A, B), while double firing was rare (mean: 15%)
(Fig. 10). Apparently, each animal had its “individual” activation pattern exhibiting either single
or double firings, likewise covering the broad spectrum of flight parameters. This is in agreement with
the independence of the mode of activation of these
motor units versus flight kinematics (Fig. 10).
Recruitment of M83/M84 (forewing elevators)
Each muscle of this elevator group, M83/M84,
comprised only one motor unit (see Fig. 5C, D)
for which a tight coupling was reported (Baker,
’79b). Our recordings allowed the interpretation
of only one unit, prohibiting an association to either of the two muscles. When the ascent angle
(α) changed from descending to a steeply ascending flight, the overall relation of single vs. double
discharges changed (Fig. 11A). While during de-
Fig. 8. Analysis of motor unit recruitment of M99 in relation to cycle length (tc). The first unit is activated with a
significantly higher relative occurrence compared to the second unit (P < 0.05), except for the small group representing
a short cycle length (30–34 ms). For the presentation of data,
see Fig. 6.
scent, a single spike usually occurred (Fig. 5C), and
when the locust steeply ascended, there was a
strong preponderance of double discharges (Fig. 5D).
When the activity of this muscle group was correlated with speed (v), the relative occurrence of single
and double discharges did not alter significantly
(Fig. 11B). However, muscle recruitment changed
when considering cycle length. At short wing-beat
cycles (tc) there was a preponderance of two spikes,
whereas longer cycles were almost exclusively correlated with single excitations (Fig. 11C).
Recruitment of M113 (hindwing elevator)
Muscle M113 consists of only one unit. This unit
discharged usually only once per wing beat. Consequently, there was no correlation between the
recruitment of this motor unit and the instantaneous flight parameters (e.g., see changes in ascent angle, Fig. 9C, D).
Fig. 9. Examples of telemetric EMGs of two hindwing
muscles, left side (l): depressor (M129l) (A, B) and elevator
(M113l) (C, D). For both muscles a constant (single) firing of
a motor unit is exhibited independent of ascent angle. Flight
parameters: A, α = –2°, v = 4.5 m/s, f = 21 Hz; B, α = 12°, v =
4.2 m/s, f = 20.5 Hz (same animal as A); C, α = –8°, v = 4.4
m/s, f = 24.5 Hz; D, α = 20°, v = 4.5 m/s, f = 25 Hz (same
animal as C).
Flight parameters
Only a few reports are available in which flight
parameters of freely flying individual locusts have
been analysed. For S. gregaria, Waloff (’72) determined mean airspeeds ranging from 3.2 to 8.1
m/s, whereas for freely flying L. migratoria in the
field, Baker et al. (’81) gave a distribution of the
flight speed spanning 3.3 to 6.1 m/s. In the
present study of freely flying S. gregaria under
laboratory conditions, the range of flight speeds
coincides with that observed in the field. As observed for L. migratoria (Baker et al., ’81) there
is a positive correlation of ascent angle (α) with
body angle to horizontal (ϕ). Furthermore, the
body angle to flight path (δ) is negatively correlated with ascent angle (α), strengthening a slight
tendency as mentioned by Baker et al. (’81).
Whether this body adjustment is passive or actively controlled by changes in wing-stroke parameters needs to be investigated. Altogether, our
restricted conditions representing short flights in
the laboratory (analysis of flights over a distance
Fig. 10. Analysis of the recruitment of the rostral motor
unit of the hindwing depressor muscle (M129) in relation to
A, ascent angle (α); B, speed (v); and C, cycle length (tc). For
this motor unit a correlation of its recruitment with any of
the flight parameters is not obvious. For the presentation of
data, see Fig. 6.
of 2–2.5 m, within about 1 sec after release, low
light intensity) do not result in a completely
changed flight performance when compared to a
flight in the field.
Cycle length (wing-beat frequency) and its
correlation to other flight parameters
Wing-beat frequency in locusts has often been
determined under different conditions. For teth-
Fig. 11. Analysis of motor unit recruitment of the forewing elevator muscles (M83/M84) in relation to A, ascent
angle (α); B, speed (v); and C, cycle length (tc). There is an
intensification of double discharges correlated with an increment of the ascent angle (α), no correlation of muscle recruitment related with speed (v), and a decrease of the relative
occurrence of double discharges with lengthening of the wingbeat cycle length (tc). For the presentation of data, see Fig. 6.
ered female S. gregaria a value of 16.8 Hz is given
(Weis-Fogh, ’56). Furthermore, it is known that
wing-beat frequency decreases after flight start,
at least in tethered flight (Weis-Fogh, ’56; Gewecke and Kutsch, ’79) and the wing-beat frequency is higher in free flight compared to
tethered flight (Baker et al., ’81; Kutsch and
Stevenson, ’87). Therefore, the presently measured mean wing-beat frequency of 23 Hz for female S. gregaria seems to represent the frequency
of free flight soon after take-off.
The major aspects of wing movements are the
production of two dynamic forces, lift and thrust.
While Baker et al. (’81) did not find a correlation
of wing-beat frequency with any of the flight variables (except flight speed, see below), the present
investigation points to a positive correlation of
wing-beat frequency with ascent angle (α). This
observation corresponds with the notion that in
tethered flight lift and wing-beat frequency are
positively correlated (Weis-Fogh, ’56; Kutsch and
Gewecke, ’79). However, under tethered flight conditions, animals produced a lift carrying only
about 70% of their own body weight (Gewecke,’75;
Kutsch and Gewecke, ’79; see Zarnack and Wortmann, ’89). Extrapolating the previously found relationship of wing-beat frequency to lift, the body
weight is almost fully compensated by the higher
wing-beat frequency in free flight—almost 23 Hz
independent of additional loading; however, when
additional loading surpasses a value of about 20%
of the animal’s weight, active flight (horizontal or
ascent) is impeded (unpublished results). Probably
in combination with other wing-stroke parameters,
increase in wing-beat frequency will result in a
steeper ascent. A positive correlation of wing-beat
frequency with speed agrees with previous results
for tethered (Gewecke, ’75) and natural flight (Baker
et al., ’81). The agreement of our present findings
with those of previous investigators, therefore, allows us to consider that the flight performance of
Schistocerca carrying a transmitter is comparable
to that of a freely flying animal.
Recruitment of muscles
Wing mechanics controls power production in
flight (e.g., Jensen, ’56; Brodsky, ’94). Previous
studies of tethered locusts have shown that by
varying certain wing-stroke parameters, flight performance can be adjusted (e.g., Weis-Fogh, ’56;
Gewecke, ’75; Zarnack, ’83, ’88; Waldmann and
Zarnack, ’88). Modulation of wing-stroke parameters is also the main principle for flight steering, when the theory of unsteady aerodynamics
is applied to insect flight (Send ’92; Dickinson and
Götz, ’93; Zarnack and Send, ’94; see also Ellington
et al., ’96).
The present advance allows us to study muscle
activities accompanying free adjustment of flight
parameters previously impeded by the tethered
approach. Hence, it becomes possible to extend the
study of insect flight from a rather “steady” to a
“dynamic” approach. Aerodynamic forces can be
adjusted by wing-beat frequency and/or by modulation of the angle of attack, stroke amplitude,
and stroke plane angle. During flight both motor
units of the forewing depressor (M99) are usually
recruited but the first unit is more active than
the second one (Figs. 6–8). However, there is no
change in the relative activation of the two motor
units, whether single or double discharges, with
respect to flight speed (Fig. 7). Speed-related activity changes are also not observed for the identified (rostral) unit of the hindwing depressor
(M129) (Fig. 10B), the forewing elevator muscle
complex (M83/M84; Fig. 11B), or the hindwing elevator (M113). Modification of flight speed, therefore, appears not to depend on a modification of
the recruitment of the motor units serving these
important flight muscles. Consequently, the higher
speed, generated at start (e.g., Kutsch and Gewecke, ’79) cannot be explained by a higher activity of the (depressor) muscles (e.g., Wilson and
Weis-Fogh, ’62; Gettrup, ’66).
In free flight, cycle length and recruitment of
the different motor units seem not to be coupled
(except M83/M84). This is in agreement with a
previous statement for tethered flight (Wilson and
Weis-Fogh, ’62; for the forewing subalar muscle),
whereas Waldron (’67) found both a positive and
a negative correlation of the two parameters, depending on the stimulus conditions. Only in the
forewing elevators (M83/M84) is a decrease in activity correlated with a lengthening of the wingbeat cycle (Fig. 11C). Therefore, in free forward
flight no specific neuronal mechanism is involved
to control the grade of muscle recruitment during
different wing-beat frequencies.
In tethered flight, body angle to horizontal (ϕ)
is correlated with the angle of attack of the beating wings (e.g., Weis-Fogh, ’56; Wortmann and
Zarnack, ’93). Measurements of the ascent angle
(α), however, require a free flight approach. A higher
angle of attack of the wings, equivalent to an enlarged pronation with respect to the animal’s flight
path, must result in a growing lift production
(Jensen, ’56). Assuming conditions are otherwise
constant, this will cause a progressive ascent angle.
Among other kinematic parameters not yet
studied, such as stroke amplitude and stroke
plane, muscle activity controls the angle of attack. The pleuroalar muscle (M85/M114) has been
argued to influence the angular setting of the
wing (Pfau and Nachtigall, ’81; Wolf, ’90) during
tethered flight; whether this muscle functions
similarly in free flight requires further study.
In tethered flight higher lift production seems
to depend mainly on the activity of the forewing basalar muscles (M97/M98) in relation to
the ipsisegmental subalar muscle (M99); for the
hindwings (representing 67.5% of the total wing
area; Weis-Fogh, ’52)—which cover almost 70%
of the produced lift (Jensen, ’56)—such differences in the recruitment of both segmental depressors are not apparent (see Thüring, ’86;
Zarnack, ’88).
In the present study of free flight, there is a
clear correlation of the recruitment of the two motor units of the forewing subalar muscle (M99)
with respect to the flight angles, whether it be
ascent angle (α) (Fig. 6) or body angle to horizontal (ϕ) (not shown here). Gettrup (’66) associated
a reduction of M99 activity with an effect on relative lift. Given the original idea that lift is dependent on the wings’ pronation/supination during
downstroke (Wilson, ’62; Wilson and Weis-Fogh,
’62), the present result indicates that a steeper
ascent (more lift) may be associated with decreased supination (increased pronation). Because
both angles (α and ϕ) are positively related to each
other (Fig. 3A), it becomes difficult to decide
whether wing kinematic during ascent requires
an active regulation of body position or whether
this angle is passive.
The presently applied video monitoring did not
give detailed information with regard to an analysis of wing position, stroke angle, or stroke plane.
Such measurements are urgently needed to fully
understand the aerodynamics of freely flying locusts and its underlying neuronal control.
There is no unequivocal trend in recruitment of
the hindwing subalar muscle (M129) units in relation to flight angles. This conforms to the observation that pronation/ supination is more pronounced
in forewings than in hindwings (Baker, ’79a; see
also Möhl and Zarnack, ’77).
In regard to the elevators, Wilson and WeisFogh (’62) mentioned a variable activity depending on lift, while Zarnack (’88) reports a switching
off of M83 resulting in a reduction of upstroke.
Our study of free flight shows that the recruitment of the forewing elevator group (M83/M84)
is strengthened with steeper ascent (Fig. 11A).
How the total power output can be effected by this
muscle group is unknown, since wing elevation
in tethered flight is only 20 % of the amount required for carrying the body weight (Jensen, ’56).
No consistent change of the equivalent hindwing
muscle (M113) was observed with respect to the
instantaneous flight parameters.
In free forward flight the chosen flight muscles
are routinely recruited independent of the wingbeat frequency or flight speed. However, the flight
path (horizontal, ascent, descent) seems to be
mainly regulated by the activity of the forewing
muscles. But for a complete description of the motor pattern, which must also consider rolling and
yawing maneuvers, other muscles have to be
tested. And not only the relative strength of
muscle recruitment but also the relative timing
of the muscles (e.g., Zarnack and Möhl, ’77; Möhl
and Zarnack, ’77; Thüring, ’86; Waldmann and
Zarnack, ’88) need consideration. We hope that
this can be tackled by employing a newly developed 2-channel recording unit (Fischer et al., ’96).
We thank Dipl.-Ing. H. Kautz for technical assistance. The helpful suggestions of two anonymous referees are appreciated.
Albrecht FO. 1953. The anatomy of the migratory locust. London: Athlone University Press.
Baader A, Schäfer M, Rowell CHF. 1992. The perception of
the visual flow field by flying locusts: a behavioural neuronal analysis. J Exp Biol 165:137–160.
Baker PS. 1979a. The wing movements of flying locusts during steering behaviour. J Comp Physiol 131:49–58.
Baker PS. 1979b. The role of forewing muscles in the control
of direction in flying locusts. J Comp Physiol 131:59–66.
Baker PS, Cooter RJ. 1979. The natural flight of the migratory locust, Locusta migratoria L: I. wing movements. J
Comp Physiol 131:79–87.
Baker PS, Gewecke M, Cooter RJ. 1981. The natural flight of
the migratory locust, Locusta migratoria L: III. wing-beat
frequency, flight speed and attitude. J Comp Physiol
Bennett-Clark HC. 1975. The energetics of the jump of the
locust Schistocerca gregaria. J Exp Biol 63:53–83.
Brodsky AK. 1994. The evolution of insect flight. Oxford: Oxford University Press.
Clopper CJ, Pearson ES. 1934. The use of confidence or fiducial limits illustrated in the case of the binomial. Biometrika
Dickinson MH, Götz KG. 1993. Unsteady aerodynamic performance of model wings at low Reynolds numbers. J Exp
Biol 174:45–64.
Ellington CP, van den Berg C, Willmott AP, Thomas ALR. 1996.
Leading-edge vortices in insect flight. Nature 384:626–630.
Fischer H. 1994. Analyse des freien Fluges der Wanderheu
schrecke Schistocerca gregaria mit optischen und radio
telemetrischen Methoden. Diploma thesis. Konstanz: University Konstanz.
Fischer H, Ebert E, Salva J, Kutsch W. 1997. Adjustments of
the free flight system of desert locust (Schistocerca gregaria)
to changes in body mass. In: Elsner N, Wässle H, editors.
Göttingen neurobiology report. Proc 25th Göttingen Neurobiol Conf., vol II. Stuttgart: Thieme. p 270.
Fischer H, Kautz H, Kutsch W. 1996. A radiotelemetric 2channel unit for transmission of muscle potentials during
free flight of the desert locust, Schistocerca gregaria. J
Neurosci Methods 64:39–45.
Gettrup E. 1966. Sensory regulation of wing twisting in locusts. J Exp Biol 44:1–16.
Gewecke M. 1975. The influence of the air-current sense organs on the flight behaviour of Locusta migratoria. J Comp
Physiol 103:79–95.
Gewecke M, Kutsch W. 1979. Development of flight behaviour in maturing adults of Locusta migratoria: I. flight
performance and wing-stroke parameters. J Insect Physiol
Jensen M. 1956. Biology and physics of locust flight: III. the
aerodynamics of locust flight. Philos Trans R Soc Lond Biol
Katz SL, Gosline JM. 1993. Ontogenetic scaling of jump performance in the African desert locust (Schistocerca gregaria).
J Exp Biol 177:81–111.
Kutsch W, Gewecke M. 1979. Development of flight behaviour
in maturing adults of Locusta migratoria: II. aerodynamic
parameters. J Insect Physiol 25:299–304.
Kutsch W, Schwarz G, Fischer H, Kautz H. 1993. Wireless
transmission of muscle potentials during free flight of a locust. J Exp Biol 185:367–373.
Kutsch W, Stevenson P. 1981. Time-correlated flights of juvenile and mature locusts: a comparison between free and
tethered animals. J Insect Physiol 27:455–459.
Möhl B. 1988. Short-term learning during flight control in
Locusta migratoria. J Comp Physiol 163:803–812.
Möhl B, Zarnack W. 1977. Activity of direct down-stroke flight
muscles of Locusta migratoria (L.) during steering behaviour
in flight: II. dynamics of the time shift and changes in burst
length. J Comp Physiol 118:235–247.
Pfau HK, Nachtigall W. 1981. Der Vorderflügel groβer
Heuschrecken als Luftkrafterzeuger. II. Zusammenspiel von
Muskeln und Gelenkmechanik bei der Einstellung der
Flügelgeometrie. J Comp Physiol 148:135–140.
Sachs L. 1978. Angewandte Statistik. Berlin: Springer.
Send W. 1992. The mean power of forces and moments in
unsteady aerodynamics. Z Angew Math Mech 72:113–132.
Snodgrass RE. 1929. The thoracic mechanism of a grasshopper, and its antecedents. Smithson Misc Coll 82:1–122.
Spork P, Preiss R. 1993. Control of flight by means of lateral
visual stimuli in gregarious desert locusts, Schistocerca
gregaria. Physiol Entom 18:195–203.
Stevenson P. 1988. Central nervous aspects of flight pattern
generation and its ontogenesis in the migratory locust. PhD
thesis. Konstanz: University Konstanz.
Stevenson RD, Corbo K, Baca LB, Le QD. 1995. Cage size
and flight speed of the tobacco hawkmoth Manduca sexta.
J Exp Biol 198:1665–1672.
Thüring DA. 1986. Variability of motor output during flight
steering in locusts. J Comp Physiol [A] 158:653–664.
Tyrer NM, Altman JS. 1974. Motor and sensory flight
neurones in a locust demonstrated using cobalt chloride. J
Comp Neurol 157:117–138.
Waldmann B, Zarnack W. 1988. Forewing movements and
motor activity during roll manoeuvers in flying desert locusts. Biol Cybern 59:325–335.
Waldron I. 1967. Neural mechanisms by which controlling
inputs influence motor output in the flying locust. J Exp
Biol 47:213–228.
Waloff Z. 1972. Observations on the airspeeds of freely flying
locusts. Anim Behav 20:367–372.
Weis-Fogh T. 1952. Fat combustion and metabolic rate of flying locusts (Schistocerca gregaria Forskål). Philos Trans R
Soc Lond Biol 237:1–36.
Weis-Fogh T. 1956. Biology and physics of locust flight: II.
flight performance of the desert locust (Schistocerca
gregaria). Philos Trans R Soc Lond Biol 239:459–510.
Wilson DM. 1962. Bifunctional muscles in the thorax of grasshoppers. J Exp Biol 39:669–677.
Wilson DM, Weis-Fogh T. 1962. Patterned activity of coordinated motor units, studied in flying locusts. J Exp Biol
Wolf H. 1990. On the function of a locust flight steering muscle
and its inhibitory innervation. J Exp Biol 150:55–80.
Wootten NW, Sawyer KF. 1954. The pick-up of spray droplets by flying locusts. Bull Entom Res 45:177–197.
Wortmann M, Zarnack W. 1993. Wing movements and lift regulation in the flight of desert locusts. J Exp Biol 182:57–69.
Zarnack W. 1983. Untersuchungen zum Flug von Wanderheu
schrecken. Die Bewegungen, räumlichen Lagebeziehungen
sowie Formen und Profile von Vorder- und Hinterflügeln.
In: Nachtigall W, editor. Biona report, vol 1. Stuttgart:
Fischer. p 79–102.
Zarnack W. 1988. The effect of forewing depressor activity on wing movement during locust flight. Biol Cybern
Zarnack W, Möhl B. 1977. Activity of the direct downstroke
flight muscles of Locusta migratoria (L.) during steering
behaviour in flight: I. patterns of time shift. J Comp Physiol
Zarnack W, Send W. 1994. Application of unsteady aerodynamics to the locust flight. In: Elsner N, Breer H, editors.
Sensory transduction. Proc. 22nd öttinger Neurobiology Conference, vol II. Stuttgart: Thieme. p 274.
Zarnack W, Wortmann M. 1989. On the so-called constantlift reaction of migratory locusts. J Exp Biol 147:111–124.
Без категории
Размер файла
274 Кб
Пожаловаться на содержимое документа