Ecological Archives E088-192-A3

Kathryn M. Flinn. 2007. Microsite-limited recruitment controls fern colonization of post-agricultural forests. Ecology 88:3103–3114.

Appendix C. Statistical analysis of the spore-sowing experiment.

TABLE C1. Generalized linear mixed models assessing treatment effects on the presence of gametophytes and sporophytes, and the number of sporophytes.

Factor

df

F

P

Odds of having gametophytes; N = 11520

Humidity

1, 805

612

<0.0001

Substrate

2, 448

157

<0.0001

Species

3, 409

7.71

<0.0001

History

1, 26

0.25

  0.6209

Year

3, 11497

312

<0.0001

Month

6, 11497

83.8

<0.0001

Year × Month

6, 11497

8.23

<0.0001

 

For plots with gametophytes, odds of having sporophytes; N = 4485

Humidity

1, 237

20.8

<0.0001

Year

2, 4471

266

<0.0001

Month

6, 4471

67.9

<0.0001

Year × Month

4, 4471

5.10

  0.0004

 

For plots with gametophytes at elevated humidity, odds of having sporophytes; N = 3913

Substrate

2, 218

92.6

<0.0001

Species

3, 176

2.50

  0.0614

History

1, 22

0.00

  0.9981

Year

2, 3894

231

<0.0001

Month

6, 3894

59.1

<0.0001

Year × Month

4, 3894

6.01

<0.0001

 

Odds of having sporophytes in May 2005; N = 720

Humidity

1, 712

124

<0.0001

Substrate

2, 712

48.3

<0.0001

Species

3, 712

1.48

  0.2184

History

1, 22

0.40

  0.5340

 

For plots with sporophytes, number of sporophytes;
N = 1506

Humidity

1, 237

65.4

<0.0001

Substrate

2, 185

66.3

<0.0001

Species

3, 170

5.60

  0.0011

History

1, 27

0.51

  0.4814

Year

2, 1486

619

<0.0001

Month

6, 1486

368

<0.0001

Year × Month

4, 1486

276

<0.0001

   Notes: A single model tested the effects of all treatments on the presence of gametophytes. For plots with gametophytes, I then tested treatment effects on the presence of sporophytes. However, only 5% of plots at ambient humidity formed sporophytes, making it impossible to analyze the effects of humidity and the other treatments simultaneously. Thus I separately analyzed the effect of humidity on the presence of sporophytes in all plots with gametophytes, and the effects of the other treatments only in plots with gametophytes at elevated humidity. The random factors accounting for sites, blocks and repeated measures are not shown. No interactions were significant at at α = 0.05.



[Back to E088-192]