Search

The objective of this 3-year study was to investigate the relationships between soil salinity and ecophysiological responses of C3 and C4 native plant species around Lake Altham and Lake Coyrecup (both are salt lakes in Western Australia), and to evaluate their potential for use in the remediation of salt-affected soils. Three shrubs (Atriplex vesicaria, Tecticornia lepidosperma and T. indica) that grew in highly saline soil of average Na concentration greater than 300 mM had higher leaf δ13C and δ15N ratios. These species also had higher Na-to-K and Na-to-Ca ratios in both their leaves and stem xylem sap, indicating that these species accumulate high amounts of sodium in their tissues. In contrast, tree species Eucalyptus loxophleba, Casuarina obesa and Acacia acuminata grew in soil of average Na concentration of less than 100 mM and had lower values of δ15N, δ13C, Na content, and Na-to-K ratio in their leaves. These species also had lower xylem Na-to-K and Na-to-Ca ratios. Seasonal effects were observed in leaf total N content, leaf Na, xylem sap Na-to-K ratio and xylem sap Na-to-Ca ratio. Strong and significant positive correlations (r > 0.75; P

The native budworm, Helicoverpa punctigera (Wallengren), is an important economic insect pest of cotton and other crops. It is widely distributed in Australia and has been recorded on a range of host plants including native, non-crop hosts in inland regions. To date, there are few records of its occurrence on plants with the C4 photosynthetic pathway. Here, we assessed the oviposition preferences of H. punctigera for naturally occurring C3 and C4 plants under glasshouse conditions, to establish their potential as hosts. We conducted bioassays on two C4 plants, saltbushes Atriplex nummularia Lindl. and Atriplex vesicaria Heward ex Benth.; and two C3 plants, legumes Medicago polymorpha L. (burr medic) and Cullen cinereum (Lindl.) J.W.Grimes (annual verbine). The two C4 plants attract egg laying in the field; however, C3 plants are the preferred hosts. Ovipositing females showed a preference for the C3 over the C4 plants but oviposition occurred on both. Of the C4 plants, females preferred to oviposit on A. nummularia (77%) over A. vesicaria (24%) in both a multi- and two-choice test. In addition, ovipositing females preferred the upper leaf surface of A. nummularia (68%) and the under leaf surface of A. vesicaria (64%) as their oviposition site. Our findings suggest that under field conditions, when C4 plants such as saltbushes grow near C3 plants, as occurs in inland regions, the C4 plants could receive eggs along with the C3 plants, enabling the initial stages of larval recruitment to occur in C3 plants.

The restoration of severely degraded vegetation communities is often said to require mechanical intervention. However, the degree of intervention required, and its capacity to successfully restore areas of bare (scalded) soil and high weed cover into functioning chenopod shrubland, is unknown. Ten years on from mechanical intervention and direct seeding using a Contour Seeder and Camel Pitter, the abundance and cover of species was compared across disturbed and undisturbed microtopographic zones using one-way repeated-measures ANOVA and pairwise t-tests. Along Contour Seeder rip lines, recruitment of perennial species was greatest in the furrow (e.g. direct seeded: F2,78 = 27.15, P

This paper reports the results of the first intensive grazing trials to study palatability variation between the sex phenotypes of bladder saltbush (Atriplex vesicaria Hew. ex Benth.). The results show that Merino sheep preferentially graze female shrubs over male and bisexual shrubs in South Australia, which supports earlier paddock-scale observations made in New South Wales. Male shrubs with higher densities of flower spikes were consumed less than those with lower densities. However, when male flowers were small and/or sparse, the removal of male flowers by clipping did not significantly increase the consumption of male shrubs. Female shrubs with very low or very high densities of fruit tended to be consumed significantly more than those with moderate fruit densities. Other physical aspects such as plant growth habit and foliage density did not have a consistent influence on grazing behaviour. Chemical differences between the sex phenotypes are implicated. The presence of palatability variation between the sex phenotypes suggests that there is potential to refine the existing use of bladder saltbush as an indicator species in southern Australia.

This paper reports the results of three cafeteria trials used to study palatability variation between the sex phenotypes of bladder saltbush (Atriplex vesicaria Heward ex Benth.). The results of the first trial show that Merino sheep preferentially grazed female samples compared to male ones, which supported earlier paddock-scale grazing trials and observations. In the second trial, the removal of male flower spikes led to increased consumption of male samples, suggesting that male flower spikes contain a grazing deterrent. The third trial showed that sheep were able to detect male material with or without spikes even when it was completely hidden within female plant material. In combination with observations made during the trials, these results suggest that there is a grazing deterrent present in male plants and that sheep use the male flower spike primarily as a visual cue when making grazing decisions.

The TGB Osborn Vegetation Reserve at Koonamore, South Australia, is a 390-ha exclosure in semi-arid chenopod shrubland. The area was heavily overgrazed in 1925 when it was fenced to exclude sheep. Permanent quadrats and photopoints have been maintained to the present. Feral rabbits were sometimes numerous until the mid-1970s but have since been controlled. The records represent 50 years without sheep grazing, followed by 26 years without either sheep or rabbits. Dramatic seedling establishment events have occurred since 1978 for the following species: Acacia aneura Benth., Myoporum platycarpum R.Br., Senna artemesioides subsp. coriacea Randell, S. artemesioides subsp. petiolaris Randell, Acacia burkittii Benth., Dodonaea attenuata A.Cunn., Eremophila longifolia (R.Br.) F.Muell., E. sturtii R.Br. and Maireana pyramidata (Benth.) Paul G.Wilson. However, the chenopod shrubs Atriplex vesicaria Benth. and A. stipitata Benth. increased earlier and did not respond in the same way to episodic rainfall events or rabbit control. Numbers of Alectryon oleifolius (Desf.) S.T.Reynolds and Casuarina pauper F.Muell. ex L.A.S.Johnson have remained almost unchanged, whereas Maireana sedifolia (F.Muell.) Paul G.Wilson and M. astrotricha (L.Johnson) Paul G.Wilson have shown a very gradual increase over time. The data show evidence for both episodic and gradual change among different species.

Dietary preferences of the greater stick-nest rat (Leporillus conditor) and the European rabbit (Oryctolagus cuniculus) were compared using cafeteria trials and direct observations. Despite overlap, these species exhibited differences in dietary preference. L. conditor showed a strong preference for chenopod shrub species and other plant species with a high water content including Gunniopsis quadrifida and Calandrinia remota. L. conditor also preferred female bladder saltbush (Atriplex vesicaria) foliage to male. O. cuniculus ate a greater selection of plant species in most trials with their most preferred species including mulga (Acacia aneura), Salsola kali and Calandrinia remota. Both O. cuniculus and L. conditor preferred seedlings of perennial species to adult cuttings. L. conditor has recently been re-introduced to a 14 km� O. cuniculus, cat (Felis catus) and red fox (Vulpes vulpes) proof exclosure in the arid zone of South Australia. The re-introduction of this herbivore and the removal of exotic herbivores may change the vegetation structure and composition within the exclosure by restricting growth of some succulent chenopod species but allowing the regeneration of tall shrub species such as Acacia aneura which are usually limited by O. cuniculus grazing.

The initial effect of two pulses of heavy cattle grazing on chenopod shrubland plants and invertebrates in northern South Australia was assessed in a two year trial. Both plant and invertebrate communities were structured by subhabitats which were defined by edaphic and water-distribution features. The cover of grasses and the dominant shrubs, Atriplex vesicaria and Maireana astrotricha, declined significantly following grazing but only total vegetation cover was significantly lower than controls one year after grazing. Plant species richness was not affected by the grazing pulses. Abundance of ants increased, but the responses of other invertebrates were inconsistent following grazing. Rainfall had a marked effect upon both plant cover and invertebrate abundance. Key words: chenopod shrublands, pulsed cattle grazing, Atriplex vesicaria, invertebrates, bioindicators

This thesis deals with some aspects of the population ecology of arid zone perennial chenopod shrubs. It examined the effects of grazing on the population dynamics of Atriplex vesicaria, Maireana astrotricha and Maireana pyramidata, the influence of weather on recruitment and survival of Atriplex vesicaria and the effects of various microtopographical features on seedling survival of Atriplex vesicaria.

Distance walked per day, time and frequency of drinking, volume of water consumed and grazing distribution of Merino sheep flocks on a saltbush (Atriplex vesicaria) and a grassland (Danthonia caespitosalStipa variabilis) community were assessed during summer. Both sites were located within the same region and observed simultaneously for three consecutive days in each of the four months December, January, February and March. The sheep on the saltbush site (1 3 10 ha) grazed as a single flock or as several large subflocks ( >300 sheep) and spent much of their time grazing on a relatively small proportion of the paddock. Drinking frequency was once daily in early summer and twice daily in late summer. The volume of water consumed varied from 2.3 l/day in December to 9.3 l/day in March. Average daily distance walked reflected the change in frequency of drinking and ranged from 8 km/day in December (once daily drinking) to 14 km/day in March (twice daily drinking). Subgrouping occurred on the grassland site (910 ha) where as few as two or three sheep would graze at distances of 10-200 m from other subgroups. Drinking frequency was once daily and volume of water consumed varied from 0.6 l/day in early summer (December) to 3.1 l/day in late summer (March). Average daily distance walked was 4.7 km/day with a peak of 6.2 km/day in late summer when forage on offer had fallen to low levels.

This study reviews the distribution of healthy bladder saltbush on the Riverine Plain of south-eastern Australia towards the end of 1986, as well as the characteristics of regeneration following the saltbush dieback of 1977-83. The area of bladder saltbush increased by 53 per cent, from 0.53 to 0.81 million ha, in the period 1983-86. A further 0.32 million ha, which previously supported saltbush, failed to regenerate. Variations in the pattern and progress of regeneration were closely related to variations in soil type and the effects of pre- and post-dieback grazing. Although the cause of dieback remains unknown, it would appear that much of the decline in the distribution of bladder saltbush could be attributed to post-dieback mismanagement of affected stands.

The population structure of a stand of saltbush (Atriplex vesicaria Hew. ex Benth.) was studied within one paddock by sampling the shruLq3opulationon a grid radiating from the only watering point. The pattern of both contemporary and cumulative grazing influence was found to be focussed on the watering point but it was not a simple symmetrical one. The existence and distribution of preferred grazing areas could not be explained by the data collected. There was a preferential grazing pressure of the female shrubs. The cumulative effects of this were reflected in a five-fold change in the sex ratio (males/females) of the shrub population.

Bladder saltbush (Ahiplex vesicaria Heward ex Benth) is a valuable perennial chenopod shrub of the northem Riverine Plain of south-eastem Australia. From 1977 to 1983, dieback affected stands of this shrub over a wide area, reducing the drought reserves on many pastoral holdings. By early 1983 reports of new outbreaks of dieback had ceased, and by late 1983 regeneration, both vegetatively and from seedlings, was evident in most areas. This study reviews what is known of this dieback event. The extent of healthy bladder saltbush stands, pre-dieback and towards the end of the dieback period was mapped. It was found that the area of leafy bush had declined by 53 per cent, from 1.1 to 0.5 million hectares, by the end of 1983. There was considerable variation in timing, pattern and progress of dieback. The cause was not established.

A comparison was made between the productivity of sheep and cattle grazing on a semiarid saltbush (Atriplex vesicaria Heward) community (median annual rainfall 303 mm) over a 4 year period. Hereford cattle were grazed at 11.7 and 17.5 ha/cow and Merino sheep were grazed at 1.7 and 2.5 ha/ewe. The cows and ewes were joined annually so that they calved and lambed in the May- June period. Calves were weaned at 7-8 months, and the lambs were weaned at c. 4 months. Both calves and lambs were then retained on separate plots for a further 12 months. The saltbush community consisted of a low shrub layer containing 250-750 kg A. vesicaria forage per ha and a herbage layer containing 25-650 kg of forage per ha. Despite the higher digestibility and nitrogen content of the sheep diets, the sheep and cattle had similar productivity. The production of body weight to weaning was a mean of 10.6 and 15.1 kg/ha/year (low and high stocking rates) for the cattle and a mean of 9.4 and 13.2 kg/ha/year for the sheep. After correction for metabolic mass (kg0.9/ha), the body weight production of sheep was raised to 12.4 and 17.3 kg/ha/year. At 12 months after weaning, the equivalent values were 12.9 and 19.4 kg/ha for cattle, and 13.4 and 20.1 kg/ha (corrected) for sheep. Thus differences are small and arise from the stocking rate chosen rather than from biological differences in productivity. There was no relative change between sheep and cattle in body weight growth between good and moderately poor seasons. The livestock numbers for equivalence of grazing pressure were calculated as 9.2 sheep to 1 cattle. The cattle drank an average of 46,300 litres of water each in a year, while sheep drank only 3300 litres per year. On this basis the water supply for cattle grazing would need to be 50% greater than for sheep grazing on the same saltbush land. In addition to body growth, the sheep produced 1.4-2.0 kg clean wool per ha, which provides a substantial economic advantage for sheep in these lands.

The application of multiple pattern analysis (multiscale ordination) to a phase-differentiated sample of perennial saltbush (Atriplex vesicaria) rangeland, along with assessments of the vegetative and reproductive states of individual plants, served to indicate the importance of sheep in determining some scales of distributional pattern in the shrub population. By comparison with conventional pattern analysis, it is shown that multiple pattern analysis can be used to provide an economic description of 'total' pattern in temporally related phases of a monospecific plant population. It is demonstrated that the pattern-determining interaction between sheep and saltbush individuals is dependent on plant size, sex and location with respect of other shrub individuals, in particular due to the trampling of seedling and pioneer phases and the selective grazing of sexually mature shrubs by sheep.

The majority of pattern analyses of total density data collected from geographically separated populations of Atriplex vesicaria, both in the western Riverine Plain of New South Wales and in areas of north-western New South Wales, provided evidence of bi- or trimodal contagion. More detailed analyses of pattern in age-related phases of these populations exhibited most intense patterning in 'seedling' phases, with a progressive reduction of pattern intensity in subsequent ('pioneer', 'building' and 'mature') phases of the entire population. It has been found that the pattern of young individuals may be correlated with patterned variation in microtopography, as well as with water and nutrient availability, and that the indirect effects of grazing animals may be responsible for the intermediate scales of the mosaic pattern in populations of sexually mature individuals. The performance of A. vesicaria plants was assessed experimentally in relation to a range of natural soils, with associated watering and defoliation treatments, and in controlled nutrient solution culture. The results suggest that performance (as estimated by dry weight yield) is affected significantly by soil type, water stress and defoliation. In an experimental context the prime effect of soil variation on yield appeared to be mediated by variation in soil cation balance, a patterned balance that may have been historically induced in the field situation.

Factors contributing to changes in the condition of saltbush communities (Atriplex vesicaria) in the Riverine Plain and Barrier Range regions of New South Wales were measured under a variety of seasonal situations. Groups of eight plots, representing a range of past grazing intensity, were established at each location and measurements were made of species biomass, canopy cover and frequency, forage availability and soil fertility. In both regions, high grazing intensity was characterized by the elimination of the saltbush (A. vesicaria) and an increase in other species, such as Astrebla (mitchell grass), Sclerolaena spp. (copperburrs), annual Atriplex spp. and a wide range of ephemeral species. These changes were best characterized by canopy cover measurements which gave a quantitative assessment of the vegetation changes (in contrast to frequency) and gave values that were less seasonally variable than biomass. Up to 83 species were recorded, but these were aggregated into approximately 9 groups to simplify the collection and analysis of data, without loss of information. On the Barrier Range site, the A. vesicaria was replaced by palatable species, so that there was little loss of overall forage production. Similarly, there was little evidence of loss of soil fertility, as measured by various measures of soil nitrogen, available phosphorus and organic carbon. There was no erosion on these sites, although this may occur on other saltbush communities that grow on texture contrast soils. On the basis of these results, four separate indices of condition are proposed for saltbush communities. These are based on the measurement in the growing season of the canopy cover of major perennial and biennial species or species groups. They are (i) the relative cover of A. vesicaria (the major observable change in these communities), (ii) the relative total cover of desirable forage species (an index of forage productivity), (iii) percentage similarity (an index of community change for conservation purposes), and (iv) the relative area of eroded soil surface.

Plant population data for the pass white-top (Danthonia caespitosa) and for bladder saltbush (Atriplex vesicaria) were collected for four years from a pasture continuously grazed by either cattle or sheep (at equivalent stocking rates) at a site near Booligal, New South Wales. During the four years of the experiment the cattle and sheep had different Impacts on the grass and scrub components of the pasture Sheep tended to deplete the numbers of pass plants far more than did cattle, whereas saltbush shrub numbers here little affected by either type of stock. The rainfall pattern experienced throughout the experiment had a far greater influence on the survival of both grass and shrub than did grazing.

Merino sheep grazing a semi-arid natural grassland were given water containing 0.0, 1.2, 1.6 or 2.0 per cent (w/v) soluble salts and similar sheep grazing on saltbush (Atriplex vesicaria) were given water containing 0.0, 0.8, 1.2 or 1.6 per cent soluble salts. The saline water was made up from sodium chloride and magnesium sulphate to give an ionic concentration similar to that occurring in natural underground waters. On grassland, 1.2 per cent saline water was without effect on the sheep, 1.6 per cent water reduced weight and wool growth in the first year and 2.0 per cent water reduced both wool growth and body weight by a significant amount. The effects were confined to the summer period of November to March. On saltbush, the results for four years were essentially similar to those for sheep on the grassland, although the effects of 1.2 per cent and 1.6 per cent saline waters were more pronounced. However, in the fifth year when drought conditions prevailed, the effects of both these treatments were more severe Water intakes were up to 1 lday-1 higher on the saline water treatments than on the fresh water, except when water intakes of the sheep on saltbush were high (7-11 lday-1) and then water salinity reduced water intakes. It was concluded that water salinities up to 1.2 per cent on grassland and 0.8 per cent on saltbush will not affect sheep productivity. Higher concentration (up to 1.6 per cent on grassland and 1.2 per cent on saltbush) can be used for adult sheep when forage conditions are good.

During late summer, Border Leicester wethers running with Merinos on a pure Atriplex nummularia stand turned over 10.6 to 17.5 litres (mean 13.7 litres) of water daily, and diluted the sodium and potassium ingested to less than 1000 m-osmoles/litre in the urine. Leicesters used 46% more water than Merinos as litres/24 hr, and 71% more as ml/kg0.82/24 hr. The two breeds produced similar urine concentrations. The sheep showed a wide variation in response, and their individual water turnovers while feeding on Atriplex vesicaria and A. nummularia ranged from 120 to 833 ml/kg0.82/24 hr. This suggests differences in functional types of adjustment to saltbush associations. On a Danthonia community Leicesters again turned over more water than Merinos but only about half as much water was required as on saltbush. Merinos grazing on A. vesicaria (with some grasses available) over 2 years maintained good condition with half the water intake of Merinos on A. nummularia, although the two species of saltbush had similar sodium and potassium contents. Total body water tended to increase with water turnover rate on saltbush. Plasma sodium concentration was inversely related to water turnover, extracellular volume, and total body water on saltbush pastures. The content of extracellular sodium was, however, greatest in the sheep with the least sodium in the food taken.

Reflection and transmission spectra were measured for the leaves of a number of species of South Australian arid zone vegetation, over the wavelength range 350- 1350 nm. Spectra were also obtained for several non-arid species for comparison. Coefficients of reflection, transmission, and absorption of solar radiation were calculated from these spectra, and also obtained from direct measurements made with a specially constructed reflecting hemisphere. The coefficients of reflection for the arid zone species covered a wide range of values, and the group as a whole could not be clearly separated from the comparison group on the basis of reflection or absorption coeficients. Of the species in the group, Atriplex vesicaria had the highest coefficient of reflection, and also the lowest coefficient of absorption.

Atriplex vesicaria, a xerophytic perennial pasture species of inland Australia, was found to have an extraordinary high tolerance to saline water'cultures; seed, lings were successfully established in solutions up to 1M NaCl.

The factors controlling the relative distribution of Atriplex vesicaria and Kochia sedifolia were obscure, and explanations in the literature unsatisfactory. This paper describes the results of an experimental approach to an understanding of these factors. A survey of the soils on which A. vesicaria and K. sedifolia are found in the arid zone of South Australia revealed that the distribution of the two species is usually correlated with the depth to which the soil is wetted by the normal rainfall. K. sedifolia, a deep-rooted species, is frequently found on soils which can be wetted to a depth of 2 ft or more. A. vesicaria, a shallow-rooted species, is found on soils in which a heavy clay subsoil or hardpan impedes penetration of water beyond 12 in. From experimental evidence it appears that: (1) either species can grow successfully in pots of the surface soil associated with the other species; (2) each community utilizes water at the same rate; (3) there is no difference between the two species in the rate of defoliation in response to drought conditions, when held at night either in a dry atmosphere or in one with a relative humidity exceeding 85 %; (4) A. vesicaria can reduce the percentage moisture in the soil to a significantly lower level than can K. sedifolia when subjected to drought. These observations are discussed in relation to field conditions, and hypotheses are proposed to explain why these two species are usually found in distinct communities. More detailed work is necessary to test these hypotheses. I.

These two species are shown to have a closely similar leaf anatomy, and this is fully described with relevant notes on all the minor points of difference. In discussion, the origin and development of the lamina are compared with other plant groups. The autecological significance of the vesicular hairs is also briefly considered from an anatomical point of view. Brief comments are made on the water-storage properties of the hypodermal layer. The widespread occurrence of the 'Kranz' type of assimilatory tissue (concentric to the xylem of the leaf veins) in the Australian species of Atriplex is noted. The need for further investigations on the thick-walled phloem elements and bundle sheath cells is stressed. A taxonomic significance is suggested for the peculiar ventral development of the midrib in both species.

Soil moisture, soil salinity, and phenological measurements were made over two years on adjacent pastures with and without bladder saltbush (Atriplex vesicaria) in the Riverina district of New South Wales. Soil moisture percentages beneath pastures with and without saltbush were similar in the first year. In the second year moisture percentages were slightly higher under pastures without saltbush. The soil profile was at matric tensions higher than 15 atmospheres for 48 per cent of the first year and 64 per cent of the second. Phenological observations related soil moisture levels to vegetative and reproductive response of the pastures. The maximum depth of soil wetting was 30 inches in the exceptionally wet autumn-winter of and 10 inches in the slightly drier than average winter of 1964. Salt was leached during the first winter of sampling, the chloride level at four inches falling from 0.107 per cent in the previous summer to 0.008 per cent. The level was partly restored during the following summer (0.055 per cent) and this was maintained during the drier second winter.

The diet selected by Merino sheep grazing a bladder saltbush (Atriplex vesicaria Hew. ex Benth.)- cottonbush (Kochia aphylla R.Br.) community was determined by analysing extrusa from wethers fitted with oesophageal fistulas. Visual assessments were combined with hand clipped samples to determine the amount of forage available. The results showed that the dominant constituent of the pasture, bladder saltbush, was sparingly grazed, whereas many of the associated species, which often constituted only a minor portion of the total available forage, were heavily grazed.

A study was made of Merino sheep grazing a bladder saltbush (Atriplex vesicaria Hew. ex Benth.)–cotton-bush (Kochia aphylla R.Br.) community set stocked at three stocking rates (2.5, 1 .2, and 0.6 sheep/hectare) over a 3-year period. In spring the sheep preferred the perennial grass Danthonia caespitosa Gaudich. and the annuals Medicago polymorpha L. and Vulpia myuros (L.) K. C. Gmel. to the more abundant A. vesicaria. From summer to early winter the diet consisted mainly of A. vesicaria with the proportion of dead matter and Medicago burr increasing as the availability of A. vesicaria decreased. Seasonal fluctuations in wool growth were evident at all three stocking rates. Maximum wool growth rates occurred in spring when the nitrogen content (2.3%) and in vitro digestibility (65%) of the forage eaten was high. During autumn and early winter values for nitrogen (1.4%) and digestibility (50%) were lower and wool growth was at a minimum. At 0.6 sheep/ha animal productivity was maintained and there was little decline in the A. vesicaria population during the 3 years. Grazing eliminated all the A. vesicaria from the pasture stocked at 2.5 sheep/ha, and 98% at 1.2 sheep/ha. At these higher stocking rates the pasture proved incapable of maintaining the sheep for the 3-year period. The inability of A. vesicaria to recover after heavy or complete defoliation by grazing demonstrates the necessity of adopting a low stocking rate if the bush is to be retained.

The productivity of Merino sheep grazing on the semi-arid natural shrub (Atriplex vesicaria) and grassland (Danthonia caespitosa-Stipa variabilis, D. caespitosa, S. variabilis) pastures of the Riverine Plain was measured over three years. At 0.5 sheep to an acre, the sheep on A. vesicaria (bladder saltbush) community at first grew more wool than those on the grasslands, but wool growth rate declined in autumn to the same degree as on the grassland. In the second year most of the A. vesicaria bushes did not recover from grazing and wool growth-rate fell to half that on the grasslands. The sheep on A. vesicaria had to be removed in the second year, but on the grasslands there was no evidence that sheep could not be maintained indefinitely at that stocking rate. At 1.0 sheep to an acre, both the A. vesicaria and the D. caespitosa-S. variabilis grassland were incapable of maintaining the sheep for the duration of the experiment. Wool growth of the sheep on the grasslands showed an annual rhythm, with high growth-rate in spring (14 g per day) followed by a gradual decline in rate to 6-8 g per day in late summer and autumn. However, the extent of this decline was dependent on summer rainfall, as in one summer when rainfall was high, wool growth-rate was maintained at near the spring rate. The mean annual rate of wool growth was found to be correlated (r = + 0.92, P

To determine the response of the perennial shrub Atriplex vesicaria Hew. ex Benth. (bladder saltbush) to grazing, bushes were grazed by Merino sheep to two levels of defoliation, severe and complete. Ungrazed bushes were used as controls. Separate groups of plots were defoliated in October and February and two levels of soil moisture were imposed. Results showed that the more heavily the bushes were grazed the less likely they were to regenerate. Completely defoliated bushes did not regrow, irrespective of soil moisture status or season. It is hypothesized that the inability of A, vesicaria to persist after complete or severe grazing is due to the removal, by grazing, of potential growth sites which apparently occur only along the young stems.

The diet selected by sheep grazing a pasture community dominated by bladder saltbush (Atriplex vesicaria Hew, ex Benth.) and pig-face (Disphyma australe (Sol.) N.E. Br.) was determined by botanically analysing the extrusa from Merino wethers fitted with oesophageal fistulas. Quality of the pasture and of the forage consumed was assessed from the nitrogen analysis of samples of material hand-plucked in the field, and of extrusa from the fistulated sheep. Results showed that in spring almost half of the forage consumed consisted of minor species which represented only 4 per cent of the pasture. In summer, when very little other material was available, Atriplex vesicaria made up 90 per cent of the diet. Disphyma australe contributed little to the diet at any time. At no time did the level of crude protein in the diet fall below that necessary for maintenance of sheep although the sodium content of the plants available in summer was high.

The distribution pattern of Atriplex vesicaria (perennial saltbush) was examined in a number of locations extending from Hay in western New South Wales to Koonamore Station in South Australia. The populations sampled were from a range of locations exhibiting a variety of topographical, edaphic, and effective rainfall conditions. The scales and intensities of pattern varied considerably between sites, but in all sites there was evidence of contagion on at least two scales (both larger than could be attributed to the morphology of the individual plants) and in no instance was there evidence of uniform (underdispersed) distribution. The two major scales of pattern were attributable to the development of gregariousness and either the patchiness of nodulated limestone (or gypsum) in the soils or the development of gilgai relief. It was demonstrated for one site that drought stress had brought about an apparent decrease in the stocking density of A. vesicaria; this reduction of density was associated with a decline in pattern intensity, although the scales of pattern present in the original vegetation were maintained in the residual population.