Appendix D. ANOVA tables showing effects of density treatment on tadpole performance (density (Table D1), tadpole body length (Table D2), hind leg length (Table D3), metamorph body length (Table D4), and time for metamorphosis (Table D5)) in high and low density sections.
Appendix D1. Effect of density treatment on tadpole density index.
Site |
Effect of
treatment
|
Year
|
Period
|
Treatment
|
Treatment ×
year
|
||||||||
df
|
F
|
P
|
df
|
F
|
P
|
df
|
F
|
P
|
df
|
F
|
P
|
||
AD17 |
0.82
|
5, 125
|
2.56
|
0.03
|
1, 125
|
0.064
|
0.80
|
1, 125
|
1.12
|
0.27
|
5, 125
|
3.49
|
0.005
|
AD17-98 |
1.14
|
4, 109
|
2.95
|
0.023
|
1, 109
|
0.049
|
0.83
|
1, 109
|
0.45
|
0.50
|
4, 125
|
0.22
|
0.93
|
HP25 |
2.79
|
6, 106
|
7.71
|
<0.001
|
1, 106
|
4.48
|
0.037
|
1, 106
|
20.92
|
<0.001
|
6, 100
|
0.63
|
0.71
|
HP27 |
1.59
|
6, 176
|
11.1
|
<0.001
|
1, 176
|
0.390
|
0.53
|
1, 176
|
15.9
|
<0.001
|
6, 176
|
3.84
|
0.001
|
HP27-91 |
1.74
|
5, 144
|
10.5
|
<0.001
|
1, 144
|
0.072
|
0.79
|
1, 144
|
27.5
|
<0.001
|
5, 139
|
22.50
|
0.034
|
Notes: The test is based on log10 tadpole densities. However, the effect of treatment is recalculated from this value to actually give the average density in the intended high density section divided by that in the low density section. The test is a two-way ANOVA. Data for AD17 and HP27 are analyzed either for all available years or excluding a year (1998 and 1991, respectively) that was identified as an outlier with respect to the effect of the density manipulation. Non-significant interactions (in italics) are not included in tests of main effects. P values are shown as given by the non-corrected test, but because three ponds are tested I only consider values <0.05/3 significant.
Appendix D2. Effect of density treatment on tadpole body length.
Site |
Effect of
treatment
|
Three-way
|
Two-way
|
||||||||||
|
Treatment × period
|
Year
|
Treatment
|
Year × treatment
|
|||||||||
|
df
|
F
|
P
|
df
|
F
|
P
|
df
|
F
|
P
|
df
|
F
|
P
|
|
AD17 |
0.70
|
1, 1717
|
14.23
|
<0.001
|
5,776
|
92.4
|
<0.001
|
<0.001
|
44.9
|
<0.001
|
5, 776
|
7.09
|
<0.001
|
AD17-98 |
0.86
|
1, 1477
|
50.9
|
<0.001
|
4, 668
|
115.7
|
<0.001
|
1, 668
|
59.6
|
<0.001
|
4, 668
|
6.50
|
<0.001
|
HP25 |
0.10
|
1,1150
|
9.00
|
0.003
|
5, 750
|
61.8
|
<0.001
|
1, 750
|
26.9
|
<0.001
|
5, 745
|
2.86
|
0.014
|
HP27 |
0.10
|
1, 593
|
0.44
|
0.51
|
4, 301
|
11.45
|
<0.001
|
1, 301
|
1.33
|
0.25
|
4, 297
|
0.59
|
0.67
|
HP27-91 |
0.00
|
1, 380
|
0.12
|
0.73
|
3, 195
|
14.1
|
<0.001
|
1, 195
|
0.005
|
0.95
|
3, 192
|
0.22
|
0.88
|
Notes: The estimated effect of the treatment is computed for the full model, with the interaction and measures the average difference in mm between tadpoles in low and high density sections respectively. The top set of tests is a three-way ANOVA analyzing the effects of year, treatment (high vs. low section density) and period (May, June). Only the effect of the interaction treatment × period is given in the table. The lower part is an account for a two-way ANOVA analyzing the effects of year and treatment. This test is only based on June data. Non-significant interactions (in italics) are not included in tests of main effects.
Appendix D3. Effect of density treatment on tadpole relative hind leg length.
Site |
Effect of
treatment
|
Year
|
Treatment
|
Year ×
treatment
|
||||||
df
|
F
|
P
|
df
|
F
|
P
|
df
|
F
|
P
|
||
AD17 |
0.024
|
5, 771
|
172.6
|
<0.001
|
1, 771
|
19.5
|
<0.001
|
5, 771
|
3.60
|
0.003
|
AD17-98 |
0.026
|
4, 663
|
190.1
|
<0.001
|
1, 663
|
18.0
|
<0.001
|
4, 663
|
3.96
|
0.003
|
HP25 |
0.004
|
5, 748
|
57.7
|
<0.001
|
1, 748
|
4.26
|
0.039
|
5, 743
|
1.25
|
0.28
|
HP27 |
-0.010
|
4, 297
|
4.59
|
0.001
|
1, 297
|
1.16
|
0.29
|
4, 297
|
2.01
|
0.64
|
HP27-91 |
-0.014
|
3, 195
|
1.68
|
0.17
|
1, 195
|
4.36
|
0.038
|
3, 192
|
0.64
|
0.59
|
Notes: The effect of treatment measures the average difference in the development index between low and high density sections. The test results are for a two-way ANOVA analyzing effects of year and treatment (high vs. low section density) on tadpole hind legs. Only June data are included. Non-significant interactions (in italics) are not included in tests of main effects.
Appendix D4. Effect of density treatment on metamorph body length.
Site |
Effect of
treatment
|
Year
|
Treatment
|
Year ×
treatment
|
||||||
df
|
F
|
P
|
df
|
F
|
P
|
df
|
F
|
P
|
||
AD17 |
0.008
|
4, 548
|
28.6
|
<0.001
|
1, 548
|
0.007
|
0.93
|
4, 548
|
24.9
|
<0.001
|
AD17-98 |
0.32
|
3, 351
|
44.6
|
<0.001
|
1, 351
|
10.0
|
0.002
|
3, 351
|
8.87
|
<0.001
|
HP25 |
0.62
|
3, 307
|
119.0
|
<0.001
|
1, 307
|
10.4
|
0.001
|
3, 307
|
9.66
|
<0.001
|
HP27 |
0.76
|
4, 513
|
201.5
|
<0.001
|
1, 513
|
40.7
|
<0.001
|
4, 513
|
6.83
|
<0.001
|
HP27-91 |
0.58
|
3, 435
|
250.9
|
<0.001
|
1, 435
|
49.6
|
<0.001
|
3, 435
|
6.99
|
<0.001
|
Notes: Effect of treatment is the difference measured in millimeters between metamorphs from low density and high density sections. The test is a two-way ANOVA.
Appendix D5. Effect of density treatment on time for metamorphosis.
Site |
Effect of
treatment
|
Year
|
Treatment
|
Year ×
treatment
|
||||||
df
|
F
|
P
|
df
|
F
|
P
|
df
|
F
|
P
|
||
AD17 |
-1.7
|
4, 459
|
60.5
|
<0.001
|
1, 459
|
2.18
|
0.14
|
4,455
|
1.70
|
0.15
|
AD17-98 |
-0.67
|
3, 283
|
60.8
|
<0.001
|
1, 283
|
0.11
|
0.74
|
3, 280
|
0.65
|
0.60
|
HP25 |
1.6
|
3, 291
|
168.3
|
<0.001
|
1, 291
|
8.00
|
0.005
|
3, 288
|
1.35
|
0.26
|
HP27 |
2.6
|
4, 388
|
19.8
|
<0.001
|
1, 388
|
23.9
|
<0.001
|
4,384
|
1.54
|
0.19
|
HP27-91 |
3.1
|
3, 329
|
1.41
|
0.24
|
1, 329
|
24.6
|
<0.001
|
3, 326
|
1.21
|
0.31
|
Notes: Effect of treatment is the average number of days by which metamorphosis was accelerated in low density sections. The test is a two-way ANOVA. Non-significant interactions (in italics) are not included in tests of main effects.