Lowe, Wheeler & Twigg: Impact of rabbits 
6-10, 3 = 11-20, 4 = 21-30, 5 = 31-50, 6 = 51-100, 7 = >100). 
The height (m) and cover (%) of the dominant 
(overstorey) shrub layer was also visually estimated for 
each 10 m x 10 m plot. 
Rabbit abundance index (RAI) 
Indications of the numbers of rabbits, and changes in 
rabbit abundance, were obtained as a rabbit abundance 
index (RAI) by counting rabbit dung on permanently 
marked quadrats. The dung quadrats were located near 
each open plot within the fenced and unfenced areas of 
the remnant vegetation in both experiments. In addition 
to these, dung quadrats were also positioned around the 
perimeter of each experimental area (see Fig 1). The 
number of dung pellets was counted on these quadrats 
every time that the associated vegetation plots were 
surveyed. There were 20 dung quadrats in each area of 
Experiment 1 and 30 quadrats in each area used in 
Experiment 2. The quadrats comprised 1 m 2 of bare soil 
(sand), and they were brushed clean after each count. 
Benefits and costs of rabbit-proof fencing 
To demonstrate the benefits and costs associated with 
the erection of rabbit-proof fencing to farmers and other 
landholders, areas of crop lost to rabbits were estimated 
by physically measuring the area affected on foot and/or 
by vehicle in areas where rabbits had ready access to 
paddocks. Production losses were calculated from the 
paddock yield per hectare and the current price (2001, 
Australian $) at 'harvest'. The cost of erecting the rabbit- 
proof fences was also determined. 
Results 
Short-term effects of rabbits on remnant vegetation 
The numbers of rabbits in the fenced and unfenced 
areas were similar, and low, at the time the fence was 
constructed (Fig 2). Rabbit numbers in the fenced 
Figure 2. Changes in the rabbit abundance index (mean dung 
pellets per plot) for the fenced and unfenced areas of remnant 
vegetation during the investigation of the short-term effects of 
confined rabbits. Data are mean (± se) of the number of pellets 
per 1 m 2 quadrat (n = 20). 
vegetation, as measured by the RAI, remained low 
throughout the experiment. However, in the unfenced 
vegetation there was a considerable rise in the dung 
counts in November which corresponds with the end of 
the breeding season when independent sub-adult rabbits 
are most abundant. The RAI suggests that rabbit numbers 
outside the fence rose to about five times those within 
the fenced vegetation. From then until the end of the 
experiment in July 2001, rabbit density in the unfenced 
area was always considerably higher than in the fenced 
vegetation with the confined rabbits (Fig 2). 
The overall seasonal pattern in the numbers of 
seedlings and reshoots (i.e. re-sprouting vegetation) 
observed in the exclosures and open plots at Parsons' site 
was similar between the fenced and unfenced areas (Fig 
3). There was a rise in the abundance of this vegetation 
class in July, followed by a decline to extremely low 
levels in summer (January to March), and a sharp rise 
between May and July following the late break of season. 
There were, however, some differences in the detail of 
this response that are worthy of note. The most marked 
decline in the numbers of seedlings and reshoots 
occurred in the open plots in the unfenced vegetation 
between September and November. This corresponds 
with the considerable rise in rabbit density between the 
same sampling times (Fig 2). In January, when the 
numbers of seedlings and reshoots were declining to low 
levels, the open plots on both the fenced and unfenced 
areas showed significantly lower numbers of seedlings 
and reshoots than in the fenced exclosures, where rabbits 
had no access to the remnant vegetation. It is interesting 
to note that on the unfenced area, the rise between May 
and July was similar to the rises in the fenced area, and 
to that which occurred within the exclosures. This was in 
spite of the continuing higher levels of rabbit density as 
indicated by the dung counts (Fig 2). It must be 
remembered, though, that the rabbits in the unfenced 
area also had access to annual species growing in the 
open paddock surrounding the site. 
Changes in the percentage vegetation cover of the 
sedges and grasses vegetation class over time, with and 
without confined rabbits, were similar throughout the 
experiment, with one exception (Fig 4). The increase in 
the amount of sedges and grasses seen between 
September and November (Spring growth) in the 
exclosures and in the open plots on the fenced area, was 
not seen in the open plots on the unfenced remnant 
vegetation where grazing by rabbits was unrestricted. 
Interestingly though, the effect on sedges and grasses did 
not appear to continue through January although the 
number of rabbits on the unfenced area remained 
relatively high during this period (Fig 2). 
There was also a decline in the percentage cover of 
sedges and grasses on all open plots and in most 
exclosures between May and July 2001 (Fig 4). The 
biggest decline was in the open plots with unrestricted 
rabbit access within the fenced area. In the previous year, 
sedges and grasses had been increasing at this time of 
year. The decrease in the sedge and grass cover between 
May 2001 and July 2001 possibly reflect the later on-set of 
autumn rains in 2001 in comparison to 2000. 
Changes in the percentage cover of small shrub 
category had the same seasonal trends as reported above 
for the other vegetation classes. There was no change in 
99 
