Nasonia
vitripennis, the jewel wasp
ILLUSTRATING SCIENCE AND BIOLOGY CONCEPTS WITH AN INSECT
�
GOAL: Provide information and exercises to allow teachers to use the
jewel wasp to illustrate biology concepts while meeting the Illinois Learning Goals. This page
grew from a presentation at New Ideas in Science conference for middle and high
school teachers
OUTLINE:
JEWEL WASP BACKGROUND
USING THE JEWEL WASP TO TEST HYPOTHESES
WHO CARES ABOUT A TINY WASP
THAT LIVES IN FLIES?
OBTAINING, MAINTAINING AND
WORKING WITH THE JEWEL WASP SYSTEM
DATA SETS: SEX RATIO, LOCOMOTOR ACTIVITY
WEB SITES, BOOKS, JOURNAL ARTICLES
I. Why use the jewel wasp? (see also Dr. Jack Werren's web
site)
A. Easy to work with living jewel wasps
1. does NOT sting or bite
2. commercially available
2. amazingly easy to maintain and
handle
3. 2 week generation time
4. suited to
independent/class research projects
B. Iinformation available
1. on a wide diversity of
topics: evolution, molecular, genetics, ecology, etc.
2. abstracts about available on the web
3. both basic and applied
research has been done
II. Wasps
A. Large, social wasps including paper wasps, hornets, yellow jackets are best known by the public.
B. But many wasps are small, live alone
C. Parasitic wasps, which the jewel wasp is, parasitize other
invertebrates, usually other insects.
There are more than 100,000 species of parasitic
wasps.
Some species are used in control of pest
insects, e.g., the jewel wasp
is sold commercially to control pest flies.
Usually small, often hidden in another
insect much of their life (photo of Muscidifurax,
of Spalangia cameroni next to a dime, exploring a host, drilling into a host to
lay an egg)
III. Taxonomy, classification of the jewel wasp
A. Domain Eucarya = species with eukaryotic cells
(DNA in membrane-bound nucleus)
Kingdom Animalia
(along with humans)
Phylum Arthropoda
(along with spiders, centipedes, shrimp)
Class Insecta
(6 legs, 2 pairs wings in adult usually)
Order Hymenoptera (= wasps, bees, and ants)
Family Pteromalidae (family names often end with "idae")
Genus Nasonia (must be
capitalized and underlined or in italics)
Species Nasonia vitripennis
(unique binomial (2 part) name, not just vitripennis)
B. Insects
1. = 70-75% of all animal species, more than 1 million
species
2. generally less studied per
species than vertebrates (fish, amphibians, reptiles, birds, mammals)
|
TAXON |
KNOWN SPECIES |
PUBLICATIONS PER Year |
papers per species per year |
|
Sponges |
190 |
|
0.02 |
|
Echinoderms |
710 |
6,000 |
0.12 |
|
Nematodes |
1,900 |
1,000,000 |
0.002 |
|
Annelids |
840 |
15,000 |
0.06 |
|
Molluscs |
1,000 |
100,000 |
0.04 |
|
Insects |
17,000 |
1,000,000 |
0.02 |
|
Vertebrates |
|
|
|
|
Fish |
7,000 |
19,000 |
0.37 |
|
Amphibians |
1,300 |
2,800 |
0.47 |
|
Reptiles |
2,400 |
6,000 |
0.41 |
|
Birds |
9,000 |
9,000 |
1.00 |
|
Mammals |
8,100 |
4,500 |
1.80 |
From May, 1988: Nature 241:1441-1449.
3. why do research on
insects instead of closer relatives
a. ethical
and practical advantages
b. economic
impact:
pollinators
damage
to health, food, belongings
control agents of other pests
C. Less than half of all existing species or organisms have yet been
discovered, even in
e.g., N. vitripennis was the only known species of Nasonia
until the mid 1980's.
Then 2 new species were discovered,
in bird nests.
1. Like most insects, they have no common names, only
their scientific names.
2. Nasonia longicornis
and Nasonia giraulti:
3. Whereas N.
vitripennis males have short, nonflying wings,
Nasonia giraulti
males have wings that are 2.4 times that size.
This difference is known to
result from a few genes, which cause bigger
cells, not more cells
IV. Biology of the Jewel Wasp
A. jewel wasp � 2 mm long
1. female = winged, can fly,
usually larger than males
2. males = brachypterous
= short wings, no flight
3. a parasitic wasp (= parasitoid
wasp)
B. host = fly pupa =
soft white pupa inside hard dark "shell"
host = Sarcophaga bullata, Calliphora vomitoria, etc.
dies when parasitized
C. Natural habitat = carrion, nests, dumpsters
1. dead animal is colonized [succession]
via smell, attracts female Nv, carrion beetles, blow flies, also bacteria,
fungi, mites
2. blow fly maggots move to drier
parts to pupate
a. fly egg -> fly larva = maggot
-> fly pupa -> winged adult fly = complete metamorphosis
b. Wings, legs, antennae, hairs pop
out during pupation, which occurs inside a "shell."
(shell = puparium = hardened
"skin" of last larval stage)
c. The adult escapes from the
"shell" by using a pulsating bulging soft-spot on it's
head.
(really cool to watch). The bulging cracks open the
"shell".
D. Jewel Wasp Life Cycle
(diagram shows female laying eggs next to black dot = where
she injected venom into the host,
then wasp larvae in host shell, then wasp pupae
in host shell, then adult wasps emerging from host shell,
note some bit of fly remains)
1. female Nv
parasitizes a fly pupa
a. She walks over it, feeling
and tasting it with her antennae, then later with the tip of her abdomen.
b. Then she unsheaths her ovipositor (egg layer), and uses it to drill
through the dark fly pupa "shell."
c. She stings into the soft
white fly pupa, injecting venom. She does NOT sting people.
d. She then squeezes eggs
(about 15) out through her ovipositor and onto the white fly pupa inside the
shell.
Or she
may stab repeatedly into the fly pupa, and create a feeding tube that brings
host fluids out onto
the
"shell," where she laps them up.
2. Inside the host "shell"
a. white egg -> wormlike
larva -> white pupa
-> dark winged adult
b. sometimes some larvae go
into a dormancy called diapause
(1) diapause = a period of delayed development and reduced
metabolism that is broken by cold-dark
(2) To break
diapause of the larvae, 3 or more months in the
refrigerator, followed by exposure to 15-25oC.
(3) Nv survive the winter by
(a) diapause
(b) fly shell -> can go 2-3X lower temp
(c) get glycerol (antifreeze) from fly host
c. Adult wasps chew a hole out of
the fly "shell."
3. males develop quicker, come out
first, wait for the females to mate.
E. Haplodiploidy: Wasps, ants, and bees are haplodiploid.
1. female stores sperm in a spermatheca
(photo shows bands=sperm in oval
structure=spermatheca and bend in the duct leading
away)
2. A duct connects the spermatheca
to the oviduct (egg passageway).
If female bends duct to hold back
sperm -> unfertilized egg -> son, 1N
(1N = 1 set of chromosomes)
If female unbends duct, release
sperm -> fertilized egg -> daughter, 2N
(2N = chromosomes matched in pairs based
on size, shape;
for each pair, one chromosome comes from the mother, the other from the
father.)
F. Other ways organisms might control offspring
sex, e.g., humans
XY sex determination -> 50% X sperm -> 50%
daughters
But can still potentially change sex ratio by:
1. Differential fertilization by X versus Y sperm X versus Y
respond differently to pH, stress hormone levels.
The timing of insemination relative
to the female's reproductive cycle (period cycle) affects her chance of a
female,
perhaps via pH changes.
2. Natural or unnatural abortion/infanticide of 1 sex.
USING THE JEWEL WASP TO TEST HYPOTHESES
I. Experimental Design: Have students "block" treatments:
do treatment one and treatment two always in pairs, matching
within each such pair as many variables as possible
-- age of wasps, day and time tested (and hence
temperature, humidity, noise, wind, etc.).
This teaches the concept of control and is easier than
trying to control every single individual wasp in an experiment.
II. Hypotheses about Locomotor Activity Levels
A. How?
Record proportion of time spent locomoting
in a 10 minute period in a terrarium or a petri dish.
B. What to compare?
1. female exposed to good host versus female exposed to dead
host
a. 1 host for about 3 hour or maybe
10 hosts for a day
b. dead hosts: freeze-killed, then
stored at room temp
c. students generate hypotheses:
(1) Predict female
that had poor host will be more active than female that had good host --
(a) this makes her more likely to find good hosts.
(b) she needs to rest to recuperate from drilling
effort
(2) Predict female that
had good host will be more active than female that had poor host --
because she has energy from feeding on the host
2. virgin versus mated females
a. Obtain virgins by separating
males from females in wasps' pupal stage.
b. students
generate hypotheses:
Predict
mated will be active as mechanism to find hosts
Predict
virgin inactive as mechanism to get mated since males can't fly.
c. Data from King, Grimm, and Reno
article (to get, click here and then click on October on left and scroll down)
3. If female's activity differs on novel versus familiar
background
-> a simple way to see what appears
different to them
a. Put each of 40 females in petri
dish with background under dish:
20 females on blue, 20
on yellow for 1 day (or pick how long by convenience, e.g,
for 1 hour, 3 hour...)
b. Then move female to new dish on same color
background or different color background
Record proportion of 10
min active.
c. Result: females on new color more active.
d. Try different
colors or patterns.
If difference,
conclude females can distinguish pattern/color.
No difference,
conclude females don't distinguish.
II. Adaptive Hypotheses (models) about Offspring Sex Ratio = % offspring that are female
A. Testing the local mate competition (LMC) model
1. assumes:
mating @ natal site, then female dispersal to new sites (e.g., carrion) to
oviposit
2. predicts female-bias sex
ratio in a population
Nv
normally produces 80-90% female offspring,
3. Sex ratio is a genetic trait and hence subject to
selection:
Artifical
selection against female bias -> 50-55% females in 14 generations
(Parker ED Jr., Orzack
SH. 1985. Genetics 110:93-105.)
4. predicts
each mother will produce decreased % daughters as more mothers
r = proportion of female
offspring, m = # of ovipositing mothers
r = (n-1)(2n-1)/(n(4n-1)). Plot r versus m to visualize.
a. Lone female will produce
only enough sons to inseminate local females.
Otherwise, competition among brothers -> waste
resources which could have been used for additional daughters.
Additional daughters provide mates for her sons.
b. When multiple mothers are
ovipositing,
(1)
additional sons increase a mother's chance that one of her sons rather than
another female's son will
inseminate locally available females.
(2) the advantage of additional daughters decreases:
they provide mates not just for her own sons, but also for the sons of other
mothers.
c. Reality: data fits model
reasonably well
Greater sex ratio with more mothers in 12 of 14 parasitic wasp species
examined,
the 12
species coming from 4 families.
B. Conclusion
1. The model is supported. BUT we need to make
sure other model(s) do not make the same prediction.
2. Crowding Model also predicts increasing
proportion of sons with increasing number of ovipositing mothers:
More mothers -> more
crowding in host -> offspring smaller.
May be better to oviposit sons when offspring will be small:
males are smaller than
females so perhaps being large is less important for a female's ability to mate
than for a female's egg
production.
3. Differential mortality is another possibility:
perhaps more daughters than sons die from crowding
4. So which model is right?
Answer: look for predictions
that differ between the 2 models, but also realize both models could be right.
C. Interspecific Test of the LMC model
1. N. vitripennis males have short wings, can't
fly
N. giraulti
males are fully-winged, can fly
2. predict N. giraulti does more mating away from the natal site
so predict N. giraulti less female-biased sex ratio than N.
vitripennis
3. reality: the opposite
Did the experiment fail?
Answer: no, failure to support hypothesis not = failed experiment
Do we throw out the model?
Answer: not from just 1 instance of lack of support but yes if repeatedly not
supported
4. Solution to puzzle of not supporting our
prediction:
Our assumption that flight
reduced local mating was wrong:
despite their longer wings,
female N. giraulti turn out to do more local
mating, inside the host, than N. vitripennis.
III. Adaption in general
A. models in general = use of math equations and graphs to predict what
trait will become most prevalent by natural
selection
adaptation = genetic trait that causes its
carrier to produce more offspring than other genetic traits do
e.g., adult Nv
play dead when disturbed which probably results in less predation, so more
offspring
B. Exercise to model natural selection, e.g., on playing dead.
Give students 2 colors of same type candy
2 colors represent 2 genotypes; death by student eating
candy
(candy evolution handout:
http://www.bios.niu.edu/bking/candy.html also shows 3 other major mechanisms of
evolution:
mutation, chance= random genetic drift, and gene
flow = migration)
C. Does adaptive, "smart" behavior
require thought?
1. Do wasps have to mathematically figure out what behavior
is most adaptive in order to behave adaptively?
Answer: no, easiest to see with a
plant e.g.,
e.g., venus
fly trap: contact -> venus fly trap close
->digests Nitrogen from insect -> better survival,
important to carnivorous plants because in N poor soil
2. Generally: intelligent behavior from 1 or more of
the following:
a. Conscious reasoning = logic
no nerves
for in plants; not yet found in insects
b. Associative learning:
if behavior ->
punishment -> stop behavior
if behavior ->
reward, continue behavior
No nerves for in plants;
has been found in insects, including Nv
c. "unthinking
intelligence" = natural selection: "survival of the fittest"
d. simple rules of thumb can -> smart
behavior in a predictable environment
e.g., bumble bees
start at bottom of inflorescence where most nectar.
rule not = start with best flower; rule = start at
bottom and go up
e.g., play dead by
seizure when disturbed versus by pretending
WHO CARES ABOUT A TINY WASP THATLIVES IN FLIES?
I. Nv is
Used in Applied Research (applicable to human problems)
A. biocontrol = control
pests with predators, parasites
1. Nv sometimes parasitizes pest flies (e.g.,
house fly)
2. Nv sold on internet
easy
to rear but other wasp species probably better for biocontrol
3. adv:
nontoxic to humans, wildlife, fairly specific hosts
B. Biocontrol can mess up
"ecological balance" just as chemical control can
e.g., Cane Toad (hilarious
video) in
C. Biocontrol may reduce
resistance problems
1. Flies and other pests
become resistant to insecticides
2. Use of biocontrol agents may select for resistant flies too (it
does in lab)
3. BUT wasps can evolve
ability to overcome fly resistance versus chemicals can't,
though we can switch chemicals
4. not
always economical compared to chemical control
II. Nv is Used in Basic Research (basic knowledge)
B. Why understanding locomotor activity is of
interest:
1. To understand evolution,
behavior
Be more or less active may be
a "simple rule of thumb" for finding good hosts, ensuring mating.
2. May be useful for biocontrol:
Release active wasps if need
them to disperse to locate hosts.
Release
inactive wasps if need them to stay at release site.
C. Why understanding sex ratios is of interest:
1. To understand evolution,
behavior
2. In biocontrol,
female-biased sex ratios may be useful.
One female can mate with many
females, and it is the females which kill the pest hosts.
OBTAINING, MAINTAINING AND WORKING WITH THE JEWEL WASP SYSTEM
I. Wasp N. vitripennis
A. Obtain from Carolina
Biological Supply Co. (They send instructions.) or Ward’s:� http://www.wardsci.com/category.asp_Q_c_E_200247_A_Nasonia
Can
crack off top about 1/6th of host pupa length at pointy (head) end to check on
wasp development.
B. Maintaining adults:
1. Keep in petri
dishes or in test tubes plugged with cotton.
Minimize static with glass versus plastic container and humidity.
2. Feed adults a wasp-size
drop of honey.
Wet honey with water drop every 2-3 days or keep in humid
container.
3. Adults live about 2 weeks
C. Pick up adults with soft forceps/paintbrush or with test
tubes
Wasps walk up tube. Slap tube
to tap them out onto a table. Then place test tube over top.
D. Generating more wasps
1. Put a mated female with 2 Sarcophaga bullata or
5 Calliphora vomitoria
for a day.
(or just leave 1 or 2 females with hosts until they die.)
2. Any female that has
been in a tube with wasps that emerged from host(s) is probably mated.
Or you can assure mating by putting together in a test tube,
a male and female that were isolated as pupae and hence are virgin.
3. Development time: Egg
to adult takes about 2 weeks at 27oC.
males emerge out of host before females by a day or 2.
E. Refrigerate to slow down, warm to speed up
development.
II. Fly Hosts:
A. Obtain
Sarcophaga
bullata = blow fly, come as pupae or larvae from Carolina
Biological Supply Co.
OR
Calliphora
vomitoria = 2nd fly host species, come as larvae,
from Grubco, Inc, OH 1-800-222-3562, $5.50 + delivery for
500
OR
check your local bait
shop.
B. Maintaining hosts:
1. If
you order larvae, open package immediately. Dump into steep sided plastic
container, like a shoebox,
with sheet of newspaper on bottom.
Can put under light -> they crawl under newpaper
to pupate.
Will be parasitizable in 2-3 days.
("Shell" becomes easy to crack with white pupa inside).
2. Store pupae in cup
covered with paper towel+ rubber band, in refrigerator for months.
3. Can crack off a cap
that is 1/6 of host pupa length at pointy (head) end to check on.
Should find a white, puffed out fly pupa. As gets deflated looking or dark, time to get new hosts.
4. Wasps can also
develop on just-thawed frozen pupae, but fewer wasps will develop per host.
III. Disposal: freeze-kill extras wasps and flies.
LOCOMOTOR ACTIVITY: Mean � SE (range = minimum - maximum) seconds
active and number of hops, short flights and long flights. Hops were less than
about 2 cm, short flights about 2 - 6 cm, and long flights about 6 cm.
Data from King,
Bethia H., Grimm, Katie and Reno, Hilary. 2000. Effects of mating on female locomotor activity in the parasitoid wasp Nasonia
vitripennis (Hymenoptera: Pteromalidae).(to get,
click here and then click on October on left and scroll down)
Short Long
Treatment n Time Active
Hops Flights
Flights
Virgin 22 238 �
28
0 0.091
0.045
(51 - 533) (0 - 0) (0 -
1) (0 - 1)
Mated 22 406 �
25
2.23
1.46 3.36
(128 - 572) (0 -29) (0 -
8) (0 - 14)
t = 4.50
df = 42
P < 0.001
OFFSPRING SEX RATIOS OF 2 NASONIA SPECIES: Mean � s.d.. sex
ratio (proportion sons) and number of emerging sons and daughters per female
with different numbers of females ovipositing on four hosts. Sample size is the
number of vials. The H values below the data are the results of analyses by Kruskal-Wallis one-way ANOVA.
Data from King, B. H. and S. W. Skinner. 1991. Sex
ratio in a new species of Nasonia with fully-winged males. Evolution
45:225-228. Plot
of data versus predicted LMC curve.
Number of
Sons
Daughters Sex
Ratio Sample
Mothers
Size
per Vial
Nasonia vitripennis
1
10.0 � 5.5 54.6 �
15.4 0.16 � 0.06 10
2
24.6 � 11.3 34.1 � 15.8
0.43 � 0.16 9
4
30.3 � 7.7 22.6 �
6.7 0.57 � 0.11
8
8
31.3 � 9.6 8.4 �
3.7 0.78 �
0.10 8
16
18.7 � 3.9 4.6 �
1.2 0.80 �
0.07 8
H = 22.97 H =
34.92 H = 34.27
P < 0.001 P <
0.001 P < 0.001
Nasonia giraulti
1
2.9 � 0.4 49.4 �
15.5 0.06 � 0.02
7
2
3.1 � 2.5 25.7 �
11.9 0.10 � 0.06
8
4
3.2 � 2.1 17.8 �
10.0 0.15 � 0.08 11
8
4.5 � 2.7 18.6 �
8.6 0.19 � 0.07
9
16
4.6 � 1.3 16.4 �
2.2 0.22 �
0.04 10
H = 9.08 H =
20.23 H = 22.67
P = 0.06 P <
0.001 P < 0.001
LOCOMOTOR ACTIVITY OF Nasonia vitripennis: Raw Data
variable/column 1: 1 = virgin, 2 = mated female
variable 2: # of hops
variable 3: # of short flights
variable 4: # of long flights
variable 5: duration of testing, minutes
variable 6: duration of testing, seconds
variable 7: time active, minutes
variable 8: time active, seconds
1 0 0 0 10 00.64 05 30.76
2 1 0 0 10 00.75 08 17.97
1 0 0 0 10 00.41 01 36.50
2 0 0 0 10 00.37 06 15.94
1 0 0 0 10 00.54 03 04.71
2 0 0 4 10 00.69 07 40.26
1 0 0 0 10 00.47 04 50.67
2 1 2 1 10 00.43 09 20.95
1 0 0 0 10 00.00 01 54.00
2 29 1 4 10 00.53 05 43.30
1 0 1 0 10 00.75 05 23.50
2 0 0 0 10 00.62 07 31.06
1 0 0 0 10 00.34 05 20.39
2 0 0 1 10 01.53 05 40.40
1 0 0 0 10 01.47 03 01.37
2 0 0 4 10 01.40 07 07.02
1 0 0 0 10 00.41 08 52.95
2 0 0 0 10 01.13 03 50.96
1 0 0 0 10 00.00 02 59.33
2 0 0 0 10 00.00 09 17.32
1 0 0 0 10 00.30 02 27.60
2 0 1 0 10 00.22 08 30.98
1 0 1 1 10 01.52 01 53.01
2 0 4 6 10 00.19 07 20.94
1 0 0 0 10 00.00 05 39.17
2 0 1 8 10 00.85 07 51.94
1 0 0 0 10 02.02 05 23.00
2 0 0 0 10 00.52 09 32.93
1 0 0 0 10 01.31 02 56.74
2 0 0 0 10 00.47 02 08.20
1 0 0 0 10 00.33 05 52.48
2 0 1 7 10 00.57 05 15.96
1 0 0 0 10 00.37 01 08.55
2 0 0 0 10 00.66 09 03.98
1 0 0 0 10 00.00 08 13.00
2 1 6 9 10 00.65 06 58.35
1 0 0 0 10 00.00 00 50.52
2 1 8 14 10 00.00 05 57.33
1 0 0 0 10 00.00 03 25.28
2 8 3 1 10 20.19 05 08.74
1 0 0 0 10 00.53 01 48.74
2 0 4 9 10 00.78 05 40.35
1 0 0 0 10 00.53 05 07.95
2 8 1 6 10 00.00 05 01.30
DATA ANALYSES
Statistics: 1) Ignore and have students just look at means (averages). or 2) Have students rely on the statistical values provided, following the statistical rule: If P < 0.05, conclude the treatments differ from each other. If P > 0.05, conclude the treatments do not differ from each other statistically. or 3) have students do statistics, by hand or by computer.
When asking a question such as, "Does the sex ratio depend on the number of mothers?," we attempt to answer by looking at a sample (e.g., a sample of wasps). If all of the single mothers produce a greater proportion of daughters than all of the paired mothers and if we have looked at lots of wasps, we feel confident that proportion of daughters depends on the number of mothers. However, outcomes often are not that clear cut. Also, because we are looking at a sample and not every single wasp that exists, there is the possibility of sampling error. Sampling error occurs when the sample you are looking at is not representative of the population from which it came. We need criteria by which to decide whether observed differences reflect real differences or are due to sampling error. Statistics provide that criteria.
It is conventional to use 0.05 as the significance level in statistical
tests. If a statistical test indicates that there is a difference among
treatments, and we have used 0.05 (or less) as the significance level, then
there is a 5% or less chance that the difference we observed is just due to
sampling error.
COMPARING TWO MEANS BY A T-TEST (e.g., time active for mated versus virgin mothers):
A t-test tests whether there is a difference between two means (averages). It assumes that the values in each of the two groups being compared are normally distributed. (I.e., if you plotted the range of observed values on the x-axis and the frequency of those values on the y-axis, then for each group the line would be bell-shaped.) The t-test also assumes that the two groups have the same variance (exhibit the same amount of variation). Luckily, the t-test is robust to these assumptions (i.e., it will give reliable results even if the assumptions are off somewhat). If you do your analyses on a computer, you can do statistical tests to check the assumptions, and switch to a nonparametric test (a Mann-Whitney U test) if the assumptions are not met.
DOING A T-TEST:
1. t = (mean of treatment 1 - mean of treatment 2) � square root of the pooled variance
2. n1 = # of values in treatment 1
n2 = # of values in treatment 2
mean for treatment1 = sum of all the
values in that treatment � n1
pooled variance = (n1 - 1)(variance1) + (n2 -
1)(variance2)
------------------------------------------
(n1 + n2 - 2)
variance1 = S (each value in treatment 1 - mean1)2
--------------------------------------
(n1 - 1)
df = degrees of freedom = n1 + n2 - 2
3. In a table of Critical values for the t-test at alpha = 0.05, find your df and the t that is in the table
for that df.
If your calculated t is bigger than the tabular t, conclude
that the two groups were different.
If the calculated t is less than the tabular t, conclude
that the two groups were not statistically different.
(For the Nasonia data set presented here, table t =
2.0)
COMPARING OBSERVED VERSUS EXPECTED FREQUENCIES BY CHI-SQUARE ANALYSIS: (e.g., number of sons and daughters observed versus expected by LMC model:
One way to have students analyze the sex ratio data is by chi-square, which compares observed frequencies to expected frequencies.
1. For each number of mothers in a vial, to generate the observed
frequencies,
multiply the number of sons and the number of daughters each
by the sample size.
2. For each number of mothers in a vial, to generate the expected frequencies
of sons: multiply the total number of observed offspring
(sons +
daughters) by the sex ratio calculated from LMC theory, m =
number of
mothers, proportion sons = (m-1)(2m-1)/(m(4m-1)).
3. To generate the expected frequencies of daughters, subtract the expected
frequency of sons from the total number of observed
offspring.
DOING A CHISQUARE TEST:
1. chi-square = S { (observed - expected)2/expected}
That is for each combination of sex and number of mothers in
a vial, you would compute
(observed - expected)2/expected. Then you
would sum those values up. Sum 10 values in this case.
2. Calculate what is called the degrees of freedom for the test: df = number of categories compared
- 1
df = 10 -1 in this case.
3. In a table of Critical values for the chi-square at alpha = 0.05, find your df and its corresponding table
chi-square.
Compare the table chi-square to the chi-square that you
computed.
a. If your calculated chi-square is bigger than the table
chi-square,
conclude that the observed
frequencies were different from the expected.
To see the logic of this, look at
the chi-square equation in #1 above:
note that the bigger the
difference between observed and expected, the bigger the Chi-Square,
and the more likely that your
observed values differ from the values that you would expect if there
was no difference among
treatments.
b. If your calculated chi-square is less than the table
chi-square,
conclude that the observed
frequencies were not statistically different from the expected.
c. You should be able to figure out which of these two
outcomes allows you to conclude that the wasps responded
differently than the LMC model
predicted.)
RESOURCES ����������
WEB SITES: search on "parasitoid
wasp" or "jewel wasp" or "sex ratio" Use the quotes.
���� ���������������������Teaching lab on inheritance
of eye color in jewel wasps: http://www.biologycorner.com/worksheets/nasonia.html
������������������������� Exploring
the Lotka-Volterra Competition Model using Two Species of Parasitoid Wasps:
http://tiee.ecoed.net/vol/v2/experiments/wasps/synopsis.html
������������������������� Competition Within and
Between Species of Parasitoid Wasps: http://www.radford.edu/~jkell/melittobia.pdf
������������������������� High school class
working with jewel wasps:� http://www.nuttallsnest.com/the_nasonia_project.htm
������������������������� Mating
isolation due to a bacterial infection in jewel wasps http://www.aai.org/committees/education/Curriculum/MariKnutson.pdf
BOOKS Brown L, Downhower JF. 1988. Analyses in Behavioral Ecology.
Behavioral labs with a wide range of species, not just insects; designed for
college age, but might provide
ideas for middle or high school science projects and has short explanations of
statistics to analyze the data
they collect.
Charnov
EL. 1982. The Theory of Sex Allocation.
Wrensch DL, Ebbert, M, eds. 1993. Evolution and Diversity of Sex Ratio in Insects and Mites.
Chapman and Hall,
Godfray HCJ. 1994. Parasitoids.
Waage JK, Greathead D, eds. 1986. Insect
Parasitoids. Academic Press,
JOURNAL ARTICLES
Review Articles:
Werren JH. 1987. Labile sex ratios in wasps and bees. Bioscience 37:498-506.
King BH. 1987. Offspring sex ratios in parasitoid
wasps. Quarterly Review of Biology 62:367-396.
Reviews, for parasitic wasps, all the different things that affect what sex
offspring a mother produces.
King BH. 1993. Sex ratio manipulation by parasitoid wasps.
pp. 418-441. In: Wrensch DL, Ebbert,
M (eds). Evolution
and Diversity of Sex Ratio in Insects and Mites. Chapman and Hall,
Summarizes sex ratio manipulation by parasitoid wasps in
relation to 2 sets of adaptation models.
Primary Literature = original descriptions of
experiment and data
King BH. 1993. Flight activity in the parasitoid wasp Nasonia vitripennis
(Hymenoptera: Pteromalidae). Journal of Insect Behavior 6:313-321.
Measured how long females from different treatments flew when tethered.
King BH, Skinner SW. 1991. Proximal
mechanisms of the sex ratio and clutch size responses of the parasitoid wasp Nasonia
vitripennis to parasitized hosts. Animal Behaviour
42:23-32.
females produce a greater proportion of sons not only
when they are with other females versus alone but also when they encounter an
already parasitized host versus an unparasitized
host. This paper examines how they tell whether or not a host has already been
parasitized.
King BH. 1992. Sex ratios of the wasp Nasonia
vitripennis from self- versus conspecifically-parasitized
hosts: local mate competition versus host quality models. Journal of
Evolutionary Biology 5:445-455.
Examines whether a female can distinguish between a host
that she herself parasitized versus one that another female parasitized. The
theory and experiments are complicated.
King BH. 1993. Sequence of offspring sex production in
the parasitoid wasp Nasonia vitripennis in response to unparasitized versus parasitized hosts. Animal Behaviour 45:1236-1238.
Moderately complex. Shows that
females tend to start laying sons first in parasitized hosts and daughters
first in unparasitized hosts.
Dr. Bethia H. King, http://www.bios.niu.edu/bking, Department of Biological Sciences, Northern Illinois University, DeKalb, IL 60115, bking@niu.edu
� Bethia King 2000, 2005, 2007
