Journal of the Royal Society of Western Australia, 87(2), June 2004 
Table 2 
Plant community mean values for soil, site and climate parameters. Differences between means of groups 1, 2, 3, 6,and 7 tested using 
Kruskal-Wallis non-parametric analysis of variance. Mean species richness and number of quadrats given for comparison. (* indicates 
P<0.05, ** indicates P<0.01, *** indicates PcO.OOl.) Climate parameter codes: Tann, mean annual temperature; Tar, annual temperature 
range; Rann, mean annual rainfall; Rev, rainfall coefficient of variation. 
1 
2 
3 
Community type 
4 5 
6 
7 
EC" 
6.3 
2.7 
2.8 
58.0 
2.0 
6.0 
9.4 
pH~ 
5.1 
4.7 
4.6 
4.4 
5.5 
6.1 
6.8 
hr 
0.125 
0.054 
0.057 
0.090 
0.034 
0.070 
0.082 
p 
230.8 
145.1 
98.5 
180.0 
88.0 
131.5 
165.7 
K"’ 
155.0 
149.6 
83.8 
160.0 
91.0 
220.5 
368.6 
Position 
2.5 
4.7 
3.1 
4.0 
5.0 
5.2 
5.1 
Slope 
2.3 
1.5 
1.4 
2.0 
2.0 
1.7 
1.4 
Aspect 
5.5 
2.1 
3.3 
4.0 
4.0 
3.1 
2.1 
% rock" 
85.0 
22.0 
53.6 
90.0 
10.0 
59.5 
30.7 
% litter* 
10.0 
31.5 
27.7 
10.0 
10.0 
26.0 
42.9 
% bare" 
0.0 
33.0 
19.1 
0.0 
10.0 
13.0 
25.7 
Altitude" 
560 
484 
485 
480 
450 
475 
471 
Latitude 
-29.9700 
-29.9673 
-29.9250 
-29.9851 
-29.9961 
-29.9551 
-29.9693 
Tann" 
18.8 
19.1 
19.2 
19.1 
19.2 
19.2 
19.2 
Tar 
30.2 
30.1 
30.2 
30.1 
30.1 
30.2 
30.1 
Rann* 
254.5 
247.1 
246.6 
247.0 
244.0 
246.2 
246.3 
Rev 
33.0 
33.0 
33.1 
32.9 
33.2 
33.0 
33.0 
Richness 
10.5 
12.7 
14.5 
7.0 
5.0 
16.9 
15.4 
No quadrats 
4 
10 
11 
1 
1 
20 
7 
As community types 4 and 5 were only represented 
by single quadrats, they were excluded from the 
statistical analysis of environmental differences. 
Significant differences were found between mean values 
for three of the four soil parameters, three of the six site 
parameters and three of the six climate parameters for 
community types 1, 2, 3, 6 and 7 (Table 2). Community 
type 1 showed a high percentage of exposed rock and 
occurred at higher altitudes with a higher estimated 
rainfall than other community types. The soils of this 
community were high in total N and had a higher pH 
and EC than community types 2 and 3 which occurred 
on deeper more acid sands. Community types 2 and 3 
were differentiated primarily on available K, percentage 
surface rock and percentage bare ground. The single 
quadrat that sampled a breakaway (type 4) showed very 
high values of EC probably related to erosion of the 
freshly exposed days beneath the duricrust. The soils of 
community types 6 and 7 were close to neutral, showed 
high levels of available K and intermediate levels of total 
N and occurred at lower altitude than types 1 to 3. The 
chenopod rich community type 7 had the highest pH, 
and highest levels of available K recorded as well as the 
best developed litter layer (Table 2). 
The edaphic differences between community types are 
also reflected in the ordination (stress 0.18 for the three 
dimensional solution) on which the highest correlated 
environmental vectors have been plotted (Fig 3). 
Community types 6 and 7 typical of the lower slopes and 
outwash flats occur in the lower left quadrant. The 
skeletal higher nutrient soils of the top of the range (type 
1) occur in the middle right of the figure while the 
nutrient poor sandy soils (of types 2 and 3) in the upper 
right of the figure. The outlying nature of community 
type 5 can be seen while community type 4 is very 
different from the others in the third dimension (not 
shown). In addition to edaphic factors, altitude and 
rainfall, topographic position and percentage litter cover 
were significantly correlated with the quadrats in 
ordination space (Fig 3). 
Discussion 
The total of 238 taxa that we recorded on the Mt 
Manning Range compares with 293 taxa recorded by 
Keighery et al (1995) for the sandplains of Mt Manning 
Range Nature Reserve, in addition to the range. The total 
for the range is low compared with the list for the Helena 
and Aurora Range (324 taxa), and this probably reflects 
in part the sampling of the Mt Manning Range later in 
the season when many of the annuals and geophytes 
were not as apparent (Gibson et al 1997). However the 
flora list for the range is also lower than those for the 
Bremer Range (267 taxa) and the Parker Range (254 taxa), 
which were also sampled in a very poor year for annuals 
and geophytes (Gibson & Lyons 1998a,b). This low total 
species richress is also consistent with a decline in 
species richness along an increasing aridity gradient 
(Beard 1972). The Mt Manning Range lies just south of 
the boundary between the South Western Interzone and 
the Murchison botanical region. The Interzone is the 
change over zone between the species-rich South West 
and the more arid Eremaean zone. 
Beard (1972) mapped the southern extreme of the Mt 
Manning Range as part of the Jackson vegetation system 
but later stated (Beard 1976) that the Mt Manning Range 
was probably similar to the vegetation of the Die Hardy 
system covering the banded ironstone formations of the 
40 
