Table 1.
The effect of short-term (7 days) and long-term (14 days) oviposition-site deprivation on egg batch size (EBS) and hatch rate
| EBS | EBS | 0-exl | OvipRt b | HR | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
Source |
DF |
Z/F |
P |
DF |
Z/F |
P |
DF |
χ2 |
P |
DF |
F |
P |
| Treatment group |
6/238 |
12.1 |
0.001 |
6/138 |
6.72 |
0.001 |
6 |
34.5 |
0.001 |
4/127 |
12.9 |
0.001 |
| Experiment |
-- |
0.56 |
0.29 |
|
0.67 |
0.25 |
1 |
2.5 |
0.12 |
1/127 |
0.31 |
0.53 |
| Replicate(Experiment) |
-- |
0.77 |
0.22 |
-- |
ND |
ND |
-- |
ND |
ND |
-- |
ND |
ND |
| Residual |
-- |
10.9 |
0.001 |
|
8.31 |
0.001 |
6 |
6.8 |
0.34 |
127 |
-- |
-- |
| (AIC) / (−2LL)a |
-- |
(2433) |
(2437) |
|
1354 |
1350 |
|
|
|
|
|
|
|
Contrasts |
|
|
|
|
|
|
|
|
|
|
|
|
| OD-7 vs. Gon1 & Gon2 |
1/238 |
12.9 |
0.001 |
1/138 |
0.01 |
0.91 |
|
11.9 |
0.001 |
1/127 |
48.7 |
0.001 |
| Gon1 & Gon2 |
1/238 |
0.23 |
0.63 |
1/138 |
0.20 |
0.65 |
|
0.0 |
0.97 |
1/127 |
0.03 |
0.86 |
| OD-7 vs. BD-7 |
1/238 |
7.0 |
0.009 |
1/138 |
0.60 |
0.44 |
|
9.8 |
0.002 |
1/127 |
31.7 |
0.001 |
| OD-14 vs. Gon1 & Gon3 |
1/238 |
31.9 |
0.001 |
1/138 |
0.04 |
0.83 |
|
16.8 |
0.001 |
-- |
ND |
ND |
| Gon1 & Gon3 |
1/238 |
7.64 |
0.006 |
1/138 |
15.9 |
0.001 |
|
0.0 |
0.95 |
-- |
ND |
ND |
| OD-14 vs. BD-14 |
1/238 |
0.6 |
0.44 |
1/138 |
2.64 |
0.11 |
|
7.1 |
0.008 |
-- |
ND |
ND |
| OD-7 vs. OD-14 | 1/238 | 9.25 | 0.003 | 1/138 | 0.04 | 0.84 | 7.0 | 0.008 | -- | ND | ND |
Mixed model ANOVA (Proc Mixed, SAS) was used to estimate variance among Experiments and Replicates (nested in Experiment) as well as Treatment group (fixed) on EBS. A log-linear categorical model (Proc Catmod, SAS) was used for oviposition rate. The effect of short-term oviposition-site deprivation on hatch rate was estimated using ANOVA (Proc GLM, SAS) using Experiment as random blocks. Hatch rate could not be estimated in females with long-term oviposition-site deprivation and blood deprivation because of very low oviposition rate (See also Figure 2 and the Methods for more details). Bold values were significant after accommodating the number of contrast tests (for each response variable separately) using the Sequential Bonferroni procedure [22].
a Model fit statistics residual −2 Log Likelihood and AIC values are given in parentheses in the DF column.
b Log-linear categorical models (Proc Catmod, SAS) were used to assess the effects of various factors.