Sheep grazing treatments

Fig. 10.14. Number of mouse-eared chickweed (Cerastium fontanum) seedlings emerging from plots with various combinations of herbivory. Sheep grazing in the summer was managed so that pastures were maintained at either 3- or 9-cm heights. In the winter pastures were either grazed (+) or not grazed (-). In addition, subplots were maintained either with or without slug herbivores (redrawn from Clear Hill and Silvertown, 1997).

the information you are looking for when you increase their density.

Clear Hill and Silvertown (1997) used herbivore addition and removal experiments to examine the interaction of slugs (dominant species grey field slug, Deroceras reticulatum) and sheep (Ovis aries) on the seedling establishment of several grassland species. Sheep herbivory was managed in two ways. First, some pastures were winter-grazed (+winter) while other were not (-winter). In addition, summer grazing was done to maintain grass height of 3 or 9 cm. Thus there are four sheep grazing treatments (Fig. 10.14). Within each sheep grazing treatment, slug density was manipulated by placing a metal ring (0.1 m-2), 5 cm deep such that slugs could not cross. Then slugs were trapped in the plots to manipulate plot density so they were either absent or at density of 20 m-2. Seeds of mouse-eared chickweed (Cerastium fontanum) were planted in each plot and their emergence counted.

The authors found that more seedlings emerged in sites that had been intensively grazed by sheep in the summer (to 3 cm). This probably occurred because intensive summer grazing created microsites suitable for seed germination. There was an interaction in the effect of sheep and slug grazing. The presence of slugs reduced seedling emergence in all treatments, except that with intense summer grazing but no winter grazing. In this treatment, slugs may have switched from eating chickweed seedlings to other food such as litter and new growth on established vegetation which was more available because they had not been removed by winter grazing sheep.

Clear Hill and Silvertown (1997) were able to do this study because they used organisms whose density was easy to manipulate. Slugs can be trapped and moved, and sheep can be enclosed or excluded using fencing. Experiments like this are much more complicated with more mobile organisms such as birds or mice. While they are easily excluded from treatments, it is harder to envision a way to increase their density but still maintain a natural habitat for them.

Seed prédation experiments

A variety of types of seed prédation experiments can be done. For example, a researcher could survey natural populations to assess levels of pre-dispersal seed predation, or alternatively, the researcher could set up experiments explicitly to test a hypothesis. Fenner and Lee (2001) conducted a survey of pre-dispersal seed predators in 13 species native to Britain and weeds in New Zealand. They collected flowering heads from 1000 individuals of each species in three locations in both countries. They dissected the flowering heads and noted whether predatory larvae were present and calculated the percentage infestation rate of each species (Table 10.6). Their results showed that infestation rate was higher in the native country (Britain) and that seed predators were almost absent in the invaded habitat (New Zealand). This is a survey approach and provides general information about the presence of pre-dispersal predators. Note that predators were not identified and their abundance per inflorescence was not counted.

Swanton et al. (1999) used an experimental approach to examine whether farming practices affected the rate of pre-disper-sal seed predation of two pigweed species (Amaranthus retroflexus and Amaranthus powellii) when grown under a maize crop.

They created different environments by varying the maize row width (37.5 and 75 cm) and maize density (75,000 and 100,000 ha-1). They found that pre-dispersal seed predation was higher when maize was planted at low density but that row width had no effect (Table 10.6).

Cromar et al. (1999) looked at post-dispersal seed predation in a similar agricultural situation. They conducted two experiments which examined the effect of: (i) tillage (mouldboard, chisel and no-till) and (ii) crop residues (maize, soybean, wheat) on post-dispersal seed predation of lambsquar-ters (Chenopodium album) and barnyardgrass (Echinochloa crusgalli). In both experiments, seeds were placed in petri dishes which were buried flush with the soil surface and soil residue was placed over the top. These dishes had small mesh cages placed over them to exclude various seed predators. Mesh of 1.5 mm excluded all organisms and was used as a control to calculate losses due to effects such as wind. A mesh size of 7 mm excluded mammals and birds but allowed insects to enter. By comparing the retention rates of seeds under the two mesh sizes, seed loss due to predation by insects was estimated. They found that seed loss was lowest under chisel plough systems, and lowest under wheat and soybean litter (Table 10.7).

Table 10.6. Percentage infestation of inflorescences of 13 species in the aster family at three locations in Britain (native habitat) and New Zealand (adapted from data in Fenner and Lee, 2001).

Britain New Zealand

Table 10.6. Percentage infestation of inflorescences of 13 species in the aster family at three locations in Britain (native habitat) and New Zealand (adapted from data in Fenner and Lee, 2001).

Britain New Zealand

Species

1

2

3

Mean

1

2

3

Mean

Yarrow, Achillea millefolium

0

0

0

0

0

0

0

0

English daisy, Bellis perennis

0

5.0

1.5

2.2

0

0

0

0

Canada thistle, Cirsium arvense

1.5

1.0

0

0.8

0

0

0

0

Bull thistle, Cirsium vulgare

2.0

28.0

4.0

11.3

0

0

0

0

Smooth hawk's-beard, Crepis tectorum

6.0

0

1.5

2.5

0

0

0

0

Mouse-ear hawkweed, Hieracium pilosella

0

0.5

0

0.2

0

0

0

0

Nipplewort, Lapsana communis

0

0

0

0

0

0

0

0

Oxeye daisy, Leucanthemum vulgare

35.5

30.5

12.0

26.0

0

0

0

0

Tansy ragwort, Senecio jacobaea

0

1.5

2.5

1.3

0

0

0

0

Common groundsel, Senecio vulgaris

0

0

0

0

0

0

0.5

0.2

Dandelion, Taraxacum officinale

0

3.0

3.0

2.0

0

0

0

0

Scentless mayweed, Tripleurospermum

23.5

48.0

27.5

33.0

0

0

0

Table 10.7 Percentage seed lost to post-dispersal seed predation in tillage and crop cover experiments by Cromar et al. (1999). Percentage predation rates (se) are based on spring and autumn sampling periods averaged over 3 years. Within experiments, treatments followed by the same letter are not significantly different, according to Tukey's test.

Tillage experiment

Crop cover experiment

No-till

Chisel plough

Mouldboard plough

Maize

Soybean Wheat

Residue biomass

The effect of sampling

A final point we would like to make is that sampling itself can have an effect on what you are measuring. Touching a plant while making observations or measurements can induce a change in its growth, survival, resource allocation or many other variables. For example, Cahill et al. (2000) tested whether visiting a plant weekly and touching it (to simulate the act of taking measurements) would influence its survival or losses due to herbivory. They found that after 8 weeks, visitation affected leaf loss in two of the six species tested: visitation increased leaf loss in hemp dogbane (Apocynum cannabinum) but decreased it in sulphur cinquefoil (Potentilla recta). In addition, there was a decrease (but not significant) in survival in yellow toadflax (Linaria vulgaris). Leaf loss and survival of the other species (horsenettle, Solarium carolinense; Canada thistle, Cirsium arvense; Kentucky bluegrass, Poa pratensis) were not affected.

This phenomenon is sometimes called the observer effect. While we presented a herbivory example, the observer effect applies to all types of experiments where plants are repeatedly visited or measured. It should be considered when deciding what types of manipulations you will be doing during the course of an experiment.

0 0

Post a comment