Beth
Lavoie
University Scientist
University of Minnesota
St. Paul MN
Overview of Lavoie's
Research | Research
Projects
Abstract
I
applied high and low nitrogen fertilizer treatments to common milkweed
(Asclepias syriaca) grown in a glasshouse
to study the effects of varying nitrogen supply on common milkweed
and monarchs. High nitrogen
fertilization resulted in increased leaf nitrogen content and plant
height and tended to increase pest levels (thrips and whiteflies). In
one cohort the high nitrogen plants also experienced a higher percentage
of necrotic leaf tissue and prevalence of bacterial wilt than the low
nitrogen plants. Reduced nitrogen
for milkweed plants increased monarch fresh and dry weight relative
consumption rates for third through fifth instars. Monarch
performance, measured with development time and relative growth rate,
increased for third and decreased for fifth instar larvae fed low nitrogen
leaves. Low nitrogen second,
third, and fourth instars and adults weighed more than their high nitrogen
counterparts. These results
suggest that milkweed pathogens, other herbivores, and fertilization
may interact to affect monarch fitness in the wild by altering monarch
size, development time, and costs of feeding.
Introduction
Nitrogen is important for caterpillar growth (Scriber
1984, Slansky and Scriber 1985), and insects are usually considered
to be nitrogen limited (McNeil and Southwood 1978, Mattson 1980, White
1993). While animal tissue generally
consists of 7-14% nitrogen by dry weight (dw), plants consist of 0.03-7.0%
nitrogen (dw) (Mattson 1980). Herbivores must therefore consume large quantities
of their host plants in order to accumulate enough nitrogen for growth
and development.
The monarch butterfly obtains
the nitrogen it requires for growth, development, and reproduction
from plants in the family Asclepiadaceae (milkweeds). Milkweeds
defend themselves from herbivores with a secretory canal system made
of specialized plant cells (laticifers) that run throughout each plant
and contain latex (Dussourd 1993). Latex
provides both mechanical and chemical protection to milkweeds. It
is a sticky plant secretion released upon injury that contains cardenolides,
heart poisons with potent effects on vertebrates (Dussourd 1993). Zalucki
and Brower (1992) found that the latex glued some first instars to
the plant while the cardenolides in the latex and the leaf tissue often
immobilized them. Monarch larvae combat milkweed defenses through
feeding behaviors such as petiole cutting and leaf trenching to slow
the flow of latex (Zalucki and Brower 1992, Malcolm et al. 1999), and
through metabolic processes such as sequestering the cardenolides in
their exoskeletons (Malcolm et al. 1999).
In spite of the general finding
that insects are nitrogen limited, studies in which host plant nutrients
were manipulated through nitrogen fertilization have had mixed results. In some studies insect performance correlated
positively with plant tissue nitrogen content (Scriber and Slansky
1981, Scriber 1984, review in Slansky 1993, Kerslake et al. 1998),
but in other cases performance was independent of plant tissue nitrogen
content (review in Slansky 1993).
Compensatory
feeding may explain why low nitrogen food negatively affects the performance
of some insects and not others. Compensatory
feeding occurs when larvae consume more food in response to decreases
in dietary nutrients (Slansky 1993). Here
I use the term "compensatory feeding" when larvae consume more food
in response to decreases in dietary nutrients such that they do not
show any decreases in performance. I
use "increased feeding" or "increased consumption" when larvae consume
more food in response to dietary nutrients but still show decreased
performance. The independence of larval performance characteristics
like growth, size, survivorship, and development time, and plant nitrogen
content may indicate that larvae compensate for a decrease in nutrients
by increasing consumption.
Several
studies of the effects of nitrogen and protein content in artificial
diets on insect performance and consumption have shown that compensatory
feeding occurs. Researchers measure relative consumption rates
(RCR) to assess how much an animal is eating; this rate is the mass
of food consumed relative to the mass of the animal. In most cases, insects fed low nitrogen or
protein diets in the laboratory compensated by increasing consumption
(Slansky 1993, Kingsolver and Woods 1998). In
other cases, some of the larvae that showed decreased dry weight relative
consumption rates (RCRdw = mass dry food consumed per unit body mass
per day) in response to decreases in plant nitrogen content showed
increases in fresh weight relative consumption rates (RCRfw
= mass fresh food consumed per unit body mass per day).
To understand
whether larvae truly increase consumption and compensate for decreases
in dietary nut rients, both RCRfw and RCRdw should be measured. Because
RCRdw does not account for the water in the food consumed, larvae may
actually be consuming and processing much more mass than the RCRdw
value shows (Slansky 1993). Since
plants vary considerably in water content, it is important to include
water weight in consumption rates for accurate comparisons (Slansky
1993). While a few studies have
measured both RCRdw and RCRfw using live prey (Traugott and Stamp 1997)
or RCRdw using leaves from live plants (Murugan and Ancy 1992, Stockhoff
1992, Raps and Vidal 1998, and Obermaier and Helmut 1999), none have
measured larval RCRdw and RCRfw using leaves from live plants over
several instars.
Both decreased
insect performance due to an inability to compensate for low nitrogen
food and the increased feeding due to compensation may affect monarch
fitness. However, no one has examined the effects of
milkweed leaf nitrogen content on monarch consumption rates or feeding. Monarchs may be able to compensate when eating
low nitrogen food or they may be limited by the presence of milkweed
cardenolides. Both scenarios
have been reported in other Lepidoptera; performance was unaffected
for some caterpillars and decreased for others consuming reduced-nutrient
diets containing allelochemicals (Slansky 1993). An
inability to compensate for low quality food decreases insect performance
by decreasing body size, decreasing mass, slowing growth rates, or
increasing development time (Slansky 1993). Body
size, the level of nutrient storage within a particular body size,
and maintaining a particular development period are all associated
with insect fitness as measured through criteria such as fecundity,
dispersal ability, mating success, life span and susceptibility to
natural enemies (review in Slansky 1993). Even
if monarchs are able to compensate for low nitrogen food, an increase
in feeding could increase the costs associated with feeding, such as
energy and nutrient costs of consuming and processing food, costs of
processing allelochemicals, and exposure to natural enemies.
In natural
habitats, monarchs probably experience varying levels of nitrogen in
milkweed leaves. Fertilizing plants with nitrogen tends to increase
total leaf nitrogen content (review in Mattson 1980) and leaf nitrogen
content tends to decrease over the growing season (Mattson 1980, Slansky
1993). Young, new shoots contain
the most nitrogen, followed by leaves from plants with flowers and
seedpods, and then by leaves from old, senescing plants. Plant protein yield and leaf nitrogen content
usually increase when plants are damaged by pathogens or herbivores,
because damage stimulates compensatory plant growth and the saliva
of some herbivores contains chemicals that promote plant growth (Mattson
1980, Hamilton et al. 1998).
I investigated
the effects of varying nitrogen supply on glasshouse-grown common milkweed
leaf nitrogen content and condition, and measured monarch performance
(development time, relative growth rate, larval and adult masses, mass
gain and average forewing length) and consumption rates (RCRfw and
RCRdw) under different nitrogen regimes. These measurements allowed me to determine
if monarch larvae compensate for low nitrogen food.
Methods
Milkweed
Glasshouse Growing Conditions
and Nitrogen Fertilization: I grew two cohorts of common
milkweed in a glasshouse at the University of Minnesota in St.
Paul during the summer of 2001. Low germination
rates, outbreaks of glasshouse pests, and plant mortality necessitated
the second cohort. The temperature
in the glasshouse ranged from 14°C to 43°C,
and natural daylight was supplemented with artificial heat-generating
lights during normal daylight hours to increase the temperature in
the glasshouse when it fell below 24°C. The first cohort included 400 cold-wet stratified
seeds collected in October 2000 from the Clifton E. French Regional
Park in Plymouth, Minnesota and another 200 cold-wet stratified seeds
bought from Butterfly Encounters in Danville, California. I planted these seeds on 4 June in plugs (4.76
cm x 6.03 cm x 6.03 cm) filled with Sun Gro Horticulture's Sunshine
SB300 Universal Mix (45-55% composted pine bark, Canadian sphagnum
peat moss, vermiculite, perlite, dolomitic limestone, gypsum, starter
nutrient charge, and wetting agent), kept them moist, and transplanted
160 seedlings three weeks later into square (13.34 cm x 13.34 cm
x 13.65 cm) pots filled with Sun Gro Horticulture's Sunshine LB2
Mix (70-80% Canadian sphagnum peat moss, vermiculite, dolomitic limestone,
gypsum, and wetting agent; no nutrients added). On
6 July I planted the second cohort of 1200 seeds obtained from Prairie
Moon Nursery (Winona, Minnesota)
in Sun Gro Horticulture's LP5 Germination Mix (70-80% fine Canadian
sphagnum peat moss, perlite, dolomitic limestone, gypsum, starter
nutrient charge, and wetting agent). I kept them moist and transplanted 335 seedlings
19 days later into square pots filled with Sunshine LB2 Mix. I watered both plant cohorts as needed, keeping
the peat moss moist and applied the following biocontrol agents ordered
from The Green Spot Ltd. in Nottingham, New
Hampshire as needed to control glasshouse
pests: Eretmocerus eremicus for sweet potato
whitefly, and Orius insidiosus,
Neoseiulus cucumeris, and Hypoaspis
miles for thrips. I did
not apply biocontrol agents for the fungus gnats present.
I randomly
assigned plants from each cohort to low and high nitrogen fertilizer
treatments. Both plant treatments received a fertilizer
solution containing equal amounts of MgSO4, potash (0-0-60),
granular super triphosphate (0-46-0), and soluble trace element mix
(13% S, 1.35% B, 2.3% Cu, 7.5% Fe, 8% Mn, 0.04% Mo and 4.5% Zn) to
provide background levels of these nutrients. High
and low nitrogen treatment plants received 13.29 g N m-2 wk-1 and
0 g N m-2 wk-1, respectively, through the addition
of NH4NO3 to the fertilizer solution. Both
treatments received approximately equal doses of nitrogen from the
starter soil mixes prior to transplanting. The
fertilizer solution, added to flats holding the potted milkweed, remained
in the flats for 24 hours and then was emptied. The
plants were fertilized weekly after they were transplanted (July 6
cohort 5 times and June 4 cohort 10 times, last fertilization on 9/6). The June 4 low nitrogen plants received 1.55
g N m-2 wk-1 for two weeks (7/30 and 8/6) since
the plants' leaves were very yellow and dropping.
Sampling and Data Collection: To
document the effect of the fertilizer treatments on the plants and
to describe the condition of the plants and leaves on which the larvae
fed, I collected data from the glasshouse-grown milkweed on 1 September
(June 4 cohort plants) and 5 September (July 6 cohort plants). These
dates corresponded to times when the larvae were eating leaves from
each cohort of plants, respectively. I
randomly chose twenty plants from each treatment group of the 4 June
cohort and ten plants from each treatment group of the 6 July cohort
for non-invasive data collection. For nitrogen analysis, I randomly chose three
plants in each treatment and cohort, and harvested the two topmost
mature leaves.
For each plant
I collected several types of data non-invasively. When
I first approached the plant, I counted the number of adult whiteflies
(0, 1, 2-10, 11-100, or 101-1000) on the entire plant without touching
the plant. I measured plant height from the soil surface
to the tip of the new leaves at the terminal bud to the nearest centimeter. I
recorded the level of thrips damage to the plant as none, light, medium,
or heavy by comparing the plant's leaves to standard leaves from each
category and selecting the category into which the most leaves fell. I
used leaf color and tissue condition as indicators of plant health. The leaf area with yellow coloring divided
by total leaf area on the plant provided a visual estimate of the percentage
of yellow leaf tissue on the plant (0%, 1-5%, 6-40%, 41-80%, or 81-100%). I used the same procedure to determine the
percentage of black and necrotic (brown and dried up) leaf tissue on
the plant. I recorded whether
or not the plant had any bacterial wilt (rapidly wilting leaves and
then death under moist soil conditions), soil fungus (visible fungus
on soil), and black spots on the leaves.
I determined leaf
nitrogen content (% dw) of milkweed leaf samples harvested from the
glasshouse. As I collected leaf
samples, I stored them in paper envelopes and immediately dried them
at 65°C for
48 hours in a drying oven. Then
I milled the leaves using a Wiley mill with a 20 mesh screen and determined
the nitrogen content using a Perkin Elmer Series II CHNS/O Analyzer
2400.
Monarchs
Source of Monarchs and Feeding: The
parents of the monarch larvae used in this experiment were second and
third generation lab stock originating from about 40 wild monarchs
caught in Texas during
the spring of 2001 and Minnesota during
the fall of 2000. On 23 August,
I selected one undiseased, relatively pest-free plant from each nitrogen
treatment, placed the plants in a 0.6 m3 mesh collapsible
field cage in the glasshouse, and randomly selected and placed twenty
mated, lab stock female monarchs in the cage. The
females oviposited on the plants for one hour, during which I observed
at least five different females laying eggs on both plants. I
removed the plants from the cage and left them 0.5 m apart in the glasshouse
so that hatched larvae could not move from one plant to the other. I
switched the positions of the two plants every other day. After
the eggs hatched, the larvae fed on their respective treatment plants
until they became late second instars, when they became easier to weigh
and transfer. I randomly selected 36 late second instars
from each treatment plant, moved them to the laboratory where they
experienced natural daylight near an east-facing window and controlled
temperatures of 21°C to
26°C,
and placed them into separate clear plastic deli containers (11.5 cm
diameter, 5.5 cm deep) with holes in the lids.
I fed the larvae leaves from
common milkweed plants growing under the appropriate fertilizer treatment. I
randomly selected enough unwilted treatment plants to feed the larvae, removed
all of the leaves, cut the leaves in half and randomly selected enough
leaves for each larva (see Table 1). I
then removed and weighed each larva's leaves from the previous day
using a Mettler semi-micro analytical balance, weighed its newly selected
leaves, and cleaned the larva's container. To
minimize water loss from leaves, I moistened the new piece of filter
paper each day with an amount of water previously determined to maintain
consistent wet mass in leaves under the experimental conditions (see
Table 1). I then transferred the larva back into its
container. Finally, I placed
the weighed, eaten leaves in labeled paper envelopes, dried them as
described earlier, reweighed them, and determined their total leaf
nitrogen content as described earlier. I
did not feed larvae any leaves from plants that had leaves removed
for nitrogen analysis.
Table 1. Mass of leaves fed to larvae
each day and amount of water used to moisten larval cage filter paper
in order to minimize milkweed leaf water loss and gain in 24 hours.
Instar
|
Approximate mass of milkweed leaves fed to larva
(g)
|
Amount of water added to minimize mass change in
24 hours (ml)
|
Average mass change in 24 hours (g)
|
Percent change in leaf mass in 24 hours
|
|
2nd and first day of 3rd
|
.400 - .750
|
2.0
|
-0.0144
|
1.9-3.6%
|
|
Remaining days of 3rd, 4th,
first day of 5th
|
.800 - 1.500
|
2.0
|
-0.0035
|
0.2-0.4%
|
|
Remaining days of 5th
|
3.500- 5.500
|
3.0
|
-0.0206
|
0.4-0.6%
|
I fed both groups of larvae
June cohort plants from 27 August to 2 September, corresponding approximately
to the first to third instars. From 3-6 Septembe, I fed both groups of late
third to early fifth instars July cohort plants because the June high
nitrogen treatment plants were wilting and dying. From 7-10 September I fed June low nitrogen
plants to the low nitrogen larvae and July high nitrogen plants to
the high nitrogen larvae because the July low nitrogen treatment plants
were gone.
Larval molts were important
to the methods of this experiment. I weighed larvae only when they were molting
because they expel all of their feces before molting and do not eat
while molting, thus maintaining more reliable masses (Waldbauer 1968,
Kogan 1986). The head capsule
is the first part of the old cuticle shed by the larva; this first
lowers and eventually detaches as it molts. 
Once
the head capsule completely separates, the larva finishes molting within
a few hours, crawling out of its old cuticle and then undergoing sclerotization
of the new cuticle (Oberhauser and Kuda 1997). I
always weighed larvae at the beginning of ecdysis, as the head capsule
began to lower and before it completely separated. The morning that
each fifth instar formed a prepupal "J", I weighed the larva by placing
the tared cage lid with the attached larva over a tared cup on the
balance. I determined that larval head capsules remain
lowered for an average of 16 ± 0.5 hours (mean ± SE, range 6 to 24 hours,
N = 70 first through fourth instars), with only three larval head capsules
remaining lowered for less than 10 hours (Figure 1).
Since the head capsules of
these three larvae began to lower at 3:30
p.m. and 8:00 p.m., I could then check for molting larvae
at 6:00 a.m., 12 noon, 6:00
p.m., and 8:00 p.m.daily
during the experiment. If a
larva was molting in the afternoon or evening, I weighed it, weighed
its leaf, and replaced the leaf so that all larvae were fed fresh leaves
in the morning only.
Data Collection: I
measured development times for the egg, first and second instars combined,
third instar, fourth instar, fifth instar, total larva, and pupa to
the nearest half day. Egg development times began at oviposition
and ended when first instar larvae hatched. First to second instar
development ran from hatching to the separation of the head capsule
of the second instar. For the
other instars, I measured development times from molt to molt, at the
same time I weighed molting larvae as described earlier. Larval development time ran from hatching to
the formation of the prepupal "J". Development
time for the pupal stage began at the formation of the prepupal "J" and
ended at eclosion. Using daily
maximum and minimum temperatures in the laboratory, I also calculated
development times in degree days (Ddd) using the formula:
Ddd = S (Tavg - B)
where the sum is calculated over all the days in the development period,
Tavg is the daily mean temperature (°C),
and B is the temperature below which an organism cannot develop (11.5°C for
monarchs) (Zalucki 1982).
To calculate
larval mass gains, I weighed second through fifth instars at the end
of each instar on a Mettler semi-micro analytical balance to the nearest
0.0001 g (M2, M3, M4, and M5). To
calculate mass gains for third, fourth, and fifth instar larvae (Mgain3,
Mgain4, Mgain5), I subtracted the final mass
of a larva from its final mass during the previous instar. I
also calculated the mass gained by each larva in all three stages summed
as an overall measure of mass gain (Mgain3-5).
Because
the larvae remained in the experiment throughout their development,
I could not calculate dry weight relative growth rates; instead I calculated
fresh weight relative growth rates (RGRfw) for each instar using the
following formula (Waldbauer 1968): RGRfwn (g g-1 d-1)=
Mgain-n/[((Mn - Mn-1)/2)*Dn]
where RGRfwn is the fresh weight relative growth rate for
the nth instar, Mgain-n is the mass gain for the nth instar,
Mn-1 is the mass of the nth instar at the end of the previous
instar, Mn is the mass of the nth instar, and Dn is
the development time for the nth instar. I
also calculated an overall measure of RGRfw (RGRfwavg) by
averaging RGRfw3, RGRfw4 and RGRfw5.
I calculated
both fresh weight and dry weight relative consumption rates (RCRfw
and RCRdw) for the larvae (Slansky 1993). I
used the following formula (Waldbauer 1968) to calculate RCRfw: RCRfwn = Ffwn/[((Mn -
Mn-1)/2)*Dn] where RCRfwn is the fresh weight relative consumption rate
for the nth instar, Ffwn is the fresh weight of food consumed
by the nth instar, Mn-1 is the final mass of the nth instar
at the end of the previous instar and therefore at the beginning of
the nth instar, Mn is the final mass of the nth instar,
and Dn is the development time for the nth instar. Ffwn = S (fresh
weight of leaf before - fresh weight of leaf after larva eats it),
summed over the duration of the instar, and the beginning and ending
points of the summation are the fresh weight of the leaf when the larva
is molting, whether or not the larva has eaten any leaf material.
I used
the same procedures and formulas to calculate the RCRdw, replacing
all fresh leaf weights with dry leaf weights. While
I could dry partially eaten leaves and weigh them directly, I had
to estimate the dry weights of the leaves before the larvae ate them. I used separate conversion factors for the
high and low nitrogen treatments to convert fresh weights to dry
weights (high nitrogen fw:dw = 5.4564 ± 0.0863 and low nitrogen
fw:dw = 6.1731 ± 0.1474). I
randomly selected 20 high and 20 low nitrogen plants over four days
of the experiment to determine these factors. I
weighed the leaves immediately after collecting them, dried the leaves
at 65°C for
48 hours and then reweighed them. I
also calculated average RCRs (RCRdwavg and RCRfwavg)
by averaging the fourth and fifth instar RCRs. I
did not include third instars because they consumed little; as a
result the fresh weight they consumed had a higher percentage of
water loss than fresh weights for other instars. This
increased the error for RCRdw3 and RCRfw3.
Adult Measurements: After
the larvae pupated, I moved them into clear plastic deli containers
(11.5 cm diameter, 16.0 cm deep) with screen placed around the edges
so the adult butterflies would have enough room to dry out after eclosion
and could climb back to the top if they fell. The pupae remained in an east-facing windowsill
until they eclosed. On the day
each butterfly eclosed, I recorded the date and sex and placed the
dry butterfly in a glassine envelope overnight. The
following morning I weighed each butterfly on a Mettler semi-micro
analytical balance to the nearest 0.001 g and measured the right and
left forewings to the nearest 0.1 mm..
Mortality: I recorded
the stage during which death occurred and the cause of death if known
or suspected. I assumed larvae or pupae died from infection
with nuclear polyhedrosis virus if they turned black, filled with a
foul-smelling black liquid and died (Urquhart 1987).
Statistical
Analyses
Plant height for both cohorts
and treatments had independent observations, was normally distributed
(Kolmogorov-Smirnov with a Lilliefors significance correction and Shapiro-Wilk
tests), and had homogeneous variances (Levene test); I therefore analyzed
it with ANOVA. Leaf nitrogen
content did not meet the assumptions for ANOVA, so I tested for differences
between the treatments within cohorts using the Mack-Skillings procedure. Since the number of adult whiteflies on the
plants, the amount of thrips damage, and the percentage of yellow,
black and necrotic leaf tissue were categorical variables, I analyzed
them using the non-parametric Mann-Whitney U test for ordered categorical
variables to compare the variables between treatment groups within
each cohort of plants. To test the independence of treatment and prevalence
of wilt, black spot and visible soil fungus I used Fisher's exact test.
Over
50% of the development times, masses, mass gains, forewing lengths,
RGRs
and RCRs failed (P < 0.05) tests of normality (Kolmogorov-Smirnov
with a Lilliefors significance correction and Shapiro-Wilk tests),
and many of these same variables, paired by treatment, failed a test
of homogeneity of variance (Levene's test) with the larger variance
often more than four times the smaller. No single variable, such as RCRfw, consistently
passed the tests of normality and homogeneity of variance across all
instars. Because of these failures
and because many variables also showed a negative skew for one treatment
group distribution and a positive skew for the other, I analyzed all
of these variables using the Mann-Whitney U test. The
analysis did not include two larvae and one pupa that died from a viral
infection because mortality and treatment group were independent (c2 =
3.13, df = 1, P = 0.077). I tested for the independence of treatment
group and mortality and sex, respectively, using Fisher's exact test.
Results
Glasshouse Milkweed Condition
and Leaf Nitrogen Content: Fertilizer
treatment significantly affected leaf nitrogen content for both plant
cohorts (Figure 2).
Leaf
nitrogen content in the high nitrogen treatment was significantly
higher than the leaf nitrogen content for the low nitrogen plants
(2.3 and 2.8 times higher in the June and July cohorts, respectively). Fertilizer
treatment and plant age significantly affected plant height (Figure
3 and Table 2). The high nitrogen
plants were 1.09 times taller and 2.25 times taller than the low
nitrogen plants for the June and July cohorts, respectively. The
June plants were 1.78 times taller and 3.67 times taller than the
July plants for the high nitrogen and low nitrogen treatments, respectively.
Glasshouse
pests tended to affect plants from the two treatment groups differently. For both cohorts, fertilizer treatment did
not significantly affect the level of thrips damage (Figure 4), although
the most frequent level of thrips damage was "light" for low nitrogen
plants and "medium" for high nitrogen plants in both cohorts. High nitrogen plants tended to have more whiteflies
on them than low nitrogen plants in both cohorts (Figure 5). These differences were not significant for
the June cohort and were significant for the July cohort.
Table
2. Two-way ANOVA for the effect
of high and low nitrogen fertilizer treatments and cohort (age)
on A. syriaca height. Cohorts
were planted on 6/4/01 and 7/6/01.
|
Factor
|
Sum of squares
|
df
|
F
|
P
|
|
Cohort
|
9257.633
|
1
|
188.471
|
<0.001
|
|
Fertilizer treatment
|
1235.208
|
1
|
25.147
|
<0.001
|
|
Cohort* fertilizer treatment
|
392.408
|
1
|
7.989
|
0.007
|
|
Error
|
2750.700
|
56
|
|
|
Table 3. Frequency of disease for A. syriaca plants fertilized with high and low nitrogen treatments. Independence of
treatment group and the prevalence of wilt, black spot, and fungus,
respectively, were analyzed using Fisher's exact test.
|
|
6/4/01 cohort (N=20 each treatment)
|
7/6/01 cohort (N=10 each treatment)
|
|
|
High nitrogen
|
Low nitrogen
|
P (1-tailed) Fisher's exact
test
|
High nitrogen
|
Low nitrogen
|
P (1-tailed)
Fisher's exact test
|
|
Wilt
|
11
|
1
|
|
0
|
0
|
|
|
No wilt
|
9
|
19
|
0.001
|
10
|
10
|
0.656
|
|
|
|
Black spot
|
19
|
18
|
|
3
|
3
|
|
|
No black spot
|
1
|
2
|
0.500
|
7
|
7
|
0.656
|
|
|
|
Fungus
|
3
|
1
|
|
0
|
0
|
|
|
No fungus
|
17
|
19
|
0.303
|
10
|
10
|
0.656
|
Some indicators of plant health differed
for plants in the two treatment groups. High nitrogen plants had a significantly higher
percentage of necrotic tissue (Figure 6) and wilted leaves (Table
3) than low nitrogen plants for the June cohort. In
the July cohort, the percentage of yellow leaf tissue was significantly
higher in the low nitrogen than in the high nitrogen plants (Figure
7). The amounts of black tissue,
black spot and fungus visible on the soil were not significantly
different for the high and low nitrogen plants in either cohort (Figures
6 and 7 and Table 3). Most
June plants from both treatment groups had 1-5% black tissue, 1-5%
yellow tissue, black spot, and no fungus. No
July plants in either treatment group had black tissue, necrotic
tissue, wilt or fungus, and the same proportion from each group had
black spot (Figure 7 and Table 3).
Monarch Performance and Consumption Rates: During the
second to fifth instars and as adults, larvae fed high nitrogen treatment
leaves were smaller than or the same size as larvae fed low nitrogen
treatment leaves throughout the larval stage (Table 4). Second,
third, and fourth instar larvae fed high nitrogen leaves had significantly
smaller final masses than those raised on low nitrogen leaves (23%,
24%, and 10% smaller, respectively). Although
these differences did carry through to adult mass, they did not result
in a significant difference in an overall measure of larval mass gain
from the third to fifth instar. Third
instar larvae raised on high nitrogen plants gained significantly less
mass than those raised on low nitrogen plants. Fifth
instar final masses, fourth and fifth instar mass gains, and adult
forewing lengths were not significantly different for larvae fed high
and low nitrogen leaves.
Table 4. Size (mass, mass gain, and average forewing
length) of D. plexippus larvae
and adults fed with leaves from A.
syriaca plants in high and low nitrogen fertilizer treatments. All
values are given as group means followed by standard errors. Size
variables for high and low nitrogen treatment groups were compared
using the Mann-Whitney U test.
|
|
High Nitrogen
|
Low Nitrogen
|
Mann Whitney U Test
|
|
Performance variable
|
Mean ± SE
|
N
|
Mean ± SE
|
N
|
Z
|
P
|
|
Second instar final mass (g)
|
0.0129 ± 0.0005
|
36
|
0.0168 ± 0.0003
|
33
|
-5.14
|
<0.001
|
|
Third instar final mass (g)
|
0.0644 ± 0.0018
|
36
|
0.0850 ± 0.0022
|
33
|
-5.50
|
<0.001
|
|
Third instar mass gain (g)
|
0.0515 ± 0.0015
|
36
|
0.0681 ±0.0019
|
33
|
-5.46
|
<0.001
|
|
Fourth instar final mass (g)
|
0.3352 ± 0.0062
|
36
|
0.3731 ± 0.0087
|
32
|
-3.15
|
0.002
|
|
Fourth instar mass gain (g)
|
0.2708 ± 0.0058
|
36
|
0.2875 ± 0.0080
|
32
|
-1.54
|
0.125
|
|
Fifth instar final mass (g)
|
1.3949 ± 0.0184
|
36
|
1.4054 ± 0.0176
|
33
|
-0.14
|
0.885
|
|
Fifth instar mass gain (g)
|
1.0597 ± 0.0146
|
36
|
1.0363 ± 0.0141
|
32
|
-1.46
|
0.145
|
|
Mass gain from third to fifth instar (g)
|
1.3820 ± 0.0183
|
36
|
1.3886 ± 0.0174
|
33
|
-0.01
|
0.995
|
|
Adult mass (g)
|
0.535 |
