(C.1)
Appendix C. Calculation of the probability of plant species being free of epifoliar fungi, under the assumption that fungi are host non-selective.
In order to calculate the probability that plant species of particular abundance would be fungus-free (or have any particular number of fungal collections), we used the binomial distribution as presented in Sokal and Rohlf (1995: pp.71–74), where:
|
|
(C.1) |
where:
k = number of collections of a plant species
Y = number of those collections with epifoliar fungi,
p = proportion of all plant samples that had epifoliar fungi, and
q = 1-p = proportion of all samples with no fungi
For Cape Tribulation p = 109/843 = 0.129 and for San Lorenzo p = 71/564 = 0.126.
For each observed frequency of a plant species at each site, we calculated the probability of observing zero fungi – p(NoFungi) – on a plant species of that frequency.
For each frequency, we then tallied the number of observed species with or without epifoliar fungi. We then used p(NoFungi) in formula (1) to calculate the probability that, under the assumption of host non-selectivity, the observed proportion of plant species of a particular frequency would be free of epifoliar fungi (Fig. C1).
|
| FIG. C1. For both Cape Tribulation, Australia and San Lorenzo, Panama, we observed many more fungus-free species than would be expected by chance under the assumption that fungi are not host selective, as expected if there was local host selectivity. In particular, less common species were less likely to harbor epifoliar fungi than expected by chance. |
Probabilities of being fungus free (p(NoFungi)) for a particular species of given frequency, and of the probability of finding the observed number of fungus-free species of that frequency, as used in Fig. C1.
For Cape Tribulation, Australia
Species frequency |
p(NoFungi) |
Obs. spp. without fungi |
Obs. spp with fungi |
Total no. spp. |
p(that many w/o fungi) |
1 |
0.8707 |
57 |
10 |
67 |
0.121042 |
2 |
0.758118 |
34 |
3 |
37 |
0.00896 |
3 |
0.660094 |
13 |
5 |
18 |
0.175609 |
4 |
0.574743 |
3 |
1 |
4 |
0.322948 |
5 |
0.50429 |
4 |
0 |
4 |
0.064673 |
6 |
0.435723 |
3 |
2 |
5 |
0.263401 |
7 |
0.379384 |
3 |
3 |
6 |
0.261058 |
8 |
0.33033 |
3 |
2 |
5 |
0.161646 |
9 |
0.287618 |
1 |
1 |
2 |
0.409788 |
10 |
0.8707 |
1 |
0 |
1 |
0.8707 |
11 |
0.218049 |
1 |
0 |
1 |
0.218049 |
13 |
0.165307 |
1 |
1 |
2 |
0.275961 |
14 |
0.143932 |
0 |
1 |
1 |
0.856068 |
15 |
0.125322 |
1 |
1 |
2 |
0.219233 |
16 |
0.109118 |
0 |
1 |
1 |
0.890882 |
19 |
0.072028 |
0 |
1 |
1 |
0.927972 |
21 |
0.054606 |
0 |
1 |
1 |
0.945394 |
23 |
0.041398 |
0 |
1 |
1 |
0.958602 |
26 |
0.027326 |
0 |
1 |
1 |
0.972674 |
28 |
0.020717 |
1 |
1 |
2 |
0.040575 |
30 |
0.015706 |
0 |
1 |
1 |
0.984294 |
37 |
0.005958 |
0 |
1 |
1 |
0.994042 |
48 |
0.001299 |
0 |
1 |
1 |
0.998701 |
57 |
0.000374 |
0 |
1 |
1 |
0.999626 |
58 |
0.000325 |
1 |
0 |
1 |
0.000325 |
For San Lorenzo, Panama
Species frequency |
p(NoFungi) |
Obs. spp. without fungi |
Obs. spp. with fungi |
Total no. spp. |
p(that many w/o fungi) |
1 |
0.874113 |
85 |
5 |
90 |
0.015001 |
2 |
0.764074 |
25 |
3 |
28 |
0.05153 |
3 |
0.667888 |
9 |
10 |
19 |
0.039879 |
4 |
0.58381 |
2 |
5 |
7 |
0.089376 |
5 |
0.510316 |
4 |
0 |
4 |
0.06782 |
6 |
0.446074 |
0 |
1 |
1 |
0.553926 |
7 |
0.389919 |
1 |
1 |
2 |
0.475764 |
8 |
0.340834 |
0 |
1 |
1 |
0.659166 |
9 |
0.297927 |
2 |
0 |
2 |
0.088761 |
10 |
0.260422 |
1 |
3 |
4 |
0.421395 |
11 |
0.227639 |
1 |
0 |
1 |
0.227639 |
15 |
0.132898 |
0 |
1 |
1 |
0.867102 |
18 |
0.088761 |
1 |
1 |
2 |
0.161764 |
23 |
0.045296 |
1 |
0 |
1 |
0.045296 |
25 |
0.03461 |
0 |
1 |
1 |
0.96539 |
52 |
0.000915 |
0 |
1 |
1 |
0.999085 |
65 |
0.000159 |
0 |
1 |
1 |
0.999841 |
LITERATURE CITED
Sokal, R. R., and F. J. Rohlf. 1995. Biometry. Third Edition. W. H. Freeman, New York, New York, USA.