Nutritive value of Australian browse plants*

A.D. Wilson and G.N. Harrington

CSIRO, Rangelands Research Centre, Division of Land Resources Management, Deniliquin, N.S.W.

1. Introduction

2. Nutritive value of species

2.1 Chenopodiaceous shrubs

2.2 Woody shrubs

2.3 Browse trees

2.4 Conclusions

3. Browse productivity

3.1 Leaf production

3.2 Reduction of herbage yield

3.3 Animal production

4. Browse management

5. Conclusions

References

1. Introduction

Shrubs and trees are a regular feature of most of the Australian arid and semi-arid rangelands. These rangelands vary from the chenopodiaceous shrub-steppes of southern Australia, where the shrubs may constitute 100% of the diet of sheep on occasions, to arid woodlands where browse trees are sparse and unimportant for forage.

Land use varies from predominantly sheep in the southern winter rainfall areas to cattle in the northern arid and summer rainfall regions, whilst goats only occur as feral animals. The palatability of trees and shrubs also varies widely, from the inedible, but ubiquitous Eucalyptus species, to the readily browsed species of Acacia and Atriplex. Such variation is to be expected from such a wide area (3.5 × 10⁶ km²) with varying climates and landforms. If the reader is interested in obtaining a better overall appreciation of these lands, he is referred to Wilson and Graetz (1979). Although Australia has specialist vertebrate browsers amongst the native fauna in the forests, there are none in the wooded grassland and steppe communities.

In this paper we do not attempt to describe the shrub communities, although community relationships are undoubtedly important in browse ecosystems. Instead, we concentrate on the specific attributes of browse species and of browse in general, as these are the aspects that may be relevant to evaluating Australian browse plants for African conditions. This paper details the known information on the nutritive value and productivity of important species. Further detailed information may be obtained from the references cited.

2. Nutritive value of species

In evaluating forage species, it is important to obtain information on crude protein content, digestibility and forage intake. These values usually determine the amount of energy that can be obtained by animals. In individual situations, high ash or oil content may also affect quality. Oil and other unpalatable substances are common in Australian shrubs and trees and these may reduce the intake by animals. These cannot be evaluated by analysis and are presented as qualitative estimates of palatability. Fortunately, seasonal variation plays a minor part in determining browse quality (Leigh et al 1978) and may be ignored in this paper. All of the published data on these attributes of the important browse plants has been compiled into tables. Where data from several sources is available, a decision was made to present only one value; a value which in our opinion is the most accurate or representative. Many species have been excluded (particularly from the tables compiled by McDonald and Ternouth, 1979) on the grounds of unpalatability, slow growth, content of toxic principles or infrequent occurrence. Data on crude fibre, etc., have been excluded because these do not aid our interpretation of nutritive value.

Information on the season of rainfall and soil characteristics of their natural habitat is also included, as these attributes will be of importance in assessing their suitability in exotic situations. In general, Australian soils are low in phosphorus and no information is available on the response of these species to soils of higher phosphorus content.

2.1 Chenopodiaceous shrubs

The available information has been assembled in Table 1. These shrubs are widespread in the southern rangelands, varying from arid to semi-arid and receiving a predominance of winter rainfall. They are highly regarded as drought forages for sheep, and as all the forage is within sheep height, no hand harvesting is required.

Table 1 Forage quality of chenopodiaceous shrubs

^aSyn. Kochia

^bChenopod shrubs are high in soluble ash content, so that organic matter intakes are 20-25% and digestibilities 8 units lower

^cRelative to other browse plants. High = eaten readily, but after herbaceous plants. Low = not eaten by sheep.

References: Siebert, Newman and Nelson (1968), Wilson (1966).

The protein content is always adequate in these species so that protein deficiencies never occur in livestock grazing on these shrublands.

Digestibility is also adequate for the maintenance of body weight. Atriplex nummularia is particularly high in both protein and digestibility. However, the forage intake of sheep is quite variable and unrelated to digestibility. In particular, the intake of A. nummularia is only about one-third of that of lucerne, although the protein content and digestibility of the two plants are similar. The intake of the less palatable Maireana pyramidata is also low and it is clear that unpalatability factors limit the amount that animals will eat. Both of these plants are eaten sparingly in the field, and in general, this group of browse plants is eaten after the major part of the herbaceous plants have been consumed (see Wilson et al, 1969, Leigh et al 1979). This low palatability relative to herbage ensures that they are reserved for the dry summer and drought periods when they can be of most value. Regardless of individual characteristics, all these species play the same role as maintenance forages in dry times.

One of the most widespread and important species, A. vesicaria, is shortlived and easily killed by defoliation and so is not considered for introduction to other countries. Another factor to consider is their high content of soluble ash, so that intakes of organic matter may be 20–25% lower than the intakes of dry matter. This salt is eliminated in urine, and hence gives rise to high water requirements, which approach 12 litres day for sheep when the diet is wholly Atriplex spp. (Wilson, 1974). This is a considerable disadvantage unless water is plentiful and well distributed within the grazing area.

2.2 Woody shrubs

Nearly all of the Australian woody shrubs are unpalatable and hence are not browse plants, or are slow-growing and difficult to establish from seed. Several of those that could be considered as useful, although more for goats than sheep, are shown in Table 2. In their natural habitat they are only eaten at times of extreme shortage of forage. The analyses show a high digestibility and protein content, but clearly these attributes are of little importance if the plants are unpalatable. The five species mentioned are all killed by repeated defoliation.

Table 2. Forage quality of browse shrubs

	Crude protein (%)	Digestibility (% D.M.) in vitro	Palatabilityb	Habitat		Characteristics
				Season of rainfall	Soil
Cassia nemophila var. coriaceaa	14	55	Low	Even	Loam
C. nemophila var. platypoda	16	56	Low	Even	Loam
Dodonea attenuata	13	–		Even	Loam
D. viscosa	13	–	Low	Even	Loam
Eremophila longifolia	13	60	Moderate	Even	Loam	Fast growth
Myoporum desertii	11	56	Moderate	Even	Clay/Loam	Can be poisonous

^aSyn. Cassia sturtii

^b Relative to other browse plants. See Table 1.

References: Harrington (1980a, 1980b), McDonald and Ternouth (1979), McLeod (1973), Maconochie (1973).

Reports that Cassia nemophila var. coriacea (syn. C. sturtii) is a palatable and desirable browse shrub in Israel (National Academy of Sciences, 1975) is not supported by field observation in Australia. This could indicate a varietal difference, which must be taken into account in any forage introduction programme. However, the reported palatability of C. sturtii is more likely to be an artifact created by its infrequent occurrence in the test situation. Shrubs of low occurrence may be heavily browsed yet not maintain that acceptability when occurrence is high.

2.3 Browse trees

Information on the major browse trees is shown in Table 3. These carry the major part of their foliage above the browsing height of sheep. Hence, in Australia, these are frequently referred to as 'topfeed' plants.

Table 3. forage quality of browse trees

	Crude Protein (%)	Digestibility in vitro	(% DM) in vivo	Intake g/sheep/day	Palatabilityb	Habitat		Characteristics
						Season of rainfall	Soil
Acacia aneura	12	40	45	600	High	Summer with	Sandy/loam	Cut for drought forage
A. aneura with supplementsa	–		50	700-1400	Winter component	Winter component
A. cana	12	41	–	–	High	as above	Stony/clay
A. homalophylla	12	40	–	–	Moderate	Even	Clay/loam
A. kempeana	–	–	–	–	High	Summer	Loam	Browsed by cattle
A. pendula	15	47	43	940	High	Winter	Clay
Atalaya hemiglauca	12	40	36	1140	High	Summer	Loam
Brachychiton rupestre	13	45	–	–	High	Summer	Clay/loam
B. populneum	13	47	52	–	High	Even	Loam
Capparis mitchellii	9	48	–	–	High	Even	Loam	Slow growth when young
Casuarina cristata	9	33	29	840	High	Even	Clay/loam	Reproduces from suckers
Heterodendrum oleifolium	13	45	40	820	High	Even	Clay/loam	Reproduces from suckers
Gefjera parviflora	15	55	55	420	Moderate/low	Even	Clay/loam	Dense crown Palatability variable
Yentilago viminalis	13	42	50	770	High	Summer	Loam

^aSupplemented with phosporus and molasses; the latter providing sulphur and energy.

^bRelative to other browse plants. See Table 1. 

References: Entwistle and Baird (1976), Everist (1967), Harrington (1980', 1980b), Harvey (1952), McDonald and Ternouth (1979), McLeod (1973), McMenimen (1976), Rohan-Jones et al (1972), Wilson (1977).

The most widespread of these is Acacia aneura, which has several forms that vary in leaf morphology. In Queensland the trees are lopped in drought times to provide a maintenance ration for sheep. It has a digestibility of only 45% and an intake of approximately 600 g/day, but research by Entwistle and Baird (1976), McMenimen (1976) and McMenimen and Little (1974) has clearly shown that the leaves are deficient in phosphorus, sulphur and available energy. Pen studies have shown that the forage intake of sheep may be almost doubled by the addition of 2 g of phosphorus and 100 g of molasses (which provides both energy and sulphur) per day, and an additional response may be obtained from urea. Recent unpublished work has also shown that there is an additional response to an intestinally digestible protein supplement, such as cottonseed meal. Digestibility is increased to 50% by these supplements. Under grazing conditions, the addition of molasses alone dramatically increased the lambs reared, from 15 to 73 lambs per 100 ewes joined. The response under grazing situations is dependent on the amount and quality of other forage available as herbaceous plants may provide some of the deficient elements.

The other Acacia species have a similar quality on analysis and it may be assumed that these too are only low-maintenance fodders and would respond in a similar manner to supplements.

For most of the 13 species in Table 3 protein content is adequate to good. However, these values must be tempered somewhat by the knowledge that not all of the protein is available for digestion. Thus in contrast to herbaceous species and chenopod shrubs, where the true digestibility of protein is 97–100%, the protein of tree forages has a true digestibility (or availability) of only 70–80% (Wilson 1977). This arises from the protection of protein from digestion by tannins (Gartner and Hurwood, 1976).

The digestibility of dry-matter is nearly always low, and only equals or exceeds 50% for three species in this list: Myoportum desertii, Geijera parviora and Brachychiton populneum. Also food intakes are generally low, with the paradox that some of the species with the lowest digestibilities have the highest intakes and vice versa. Thus although they are browsed at a low intensity throughout the year, as a major source of food these browse trees are forages of last resort.

2.4 Conclusions

These results show that laboratory analyses of browse forages, including in vitro digestibility, are of limited value in assessing nutritive value. Apart from the problems of variations in the in vitro method, food intake is not necessarily related to digestibility so that the value of in vitro digestibility for assessing quality is much reduced. Even pen feeding results are difficult to evaluate because of the possibility of mineral deficiencies. These may arise only because of the limited range of foods and may be less severe in field situations where mixtures of browse species or browse and herbaceous species are eaten.

The forages with a high digestibility and relatively low intake apparently have unpalatability factors that reduce consumption. Thus palatability is not only important in a relative sense which determines the order in which species are grazed, but also in an absolute sense which may affect forage quality. There are no browse species of both high quality and high palatability and perhaps we should not expect there to be. Such plants would quickly succumb to the concentration of grazing. Browse plants are, by definition, perennial with their growing points exposed. Although species such as Acacia aneura demonstrate abilities to survive after repeated defoliation, they are better adapted to infrequent browsing.

Forage quality is also evaluated in a relative way, in comparison to the other herbaceous forages that grow in particular regions. In Australia, herbaceous quality is usually poorer in environments dominated by summer rainfall and perennial grasses than in environments dominated by winter rainfall and containing numerous forbs. As a consequence, top feeds are seen to be of more value in the former than in the latter. However, in the subtropical regional of Australia, where herbage quality is very low in the dry season, browse is not a significant factor. Erythrophleum chlorostachys, a leguminous tree, is eaten by cattle when available. It is unlikely to be superior to African species of the same genus. Acacia aneura is the main top feed species in latitudes 25 to 29° S, and which receive a dominance of summer rainfall.

3. Browse productivity

3.1 Leaf production

It is difficult to measure the amount of leaf produced by browse species, as it is influenced by age of tree, defoliation and rainfall. However, we may consider the purpose of browse plants is to provide a drought reserve of fodder which will be used in the dry season on infrequent years. In this situation productivity may be seen as the amount of leaf accumulated over several years and retained in a drought year. It is then no longer necessary to engage in measures of true productivity.

The standing crop of foliage on trees and shrubs is most readily estimated by the development of regression equations relating leaf weight to stem diameter or shrub height. This has been successfully done for Acacia aneura by Burrows and Beale (1970), and Beale (1973) and for various shrubs by Harrington (1979).

The amount of leaf retained by Australian browse plants has been measured for only a few species. Information for an A. aneura community is presented by Beale (1973) and is presented in a generalized form in Table 4. Foliage weight per tree varied from 3 to 5 kg, depending on density (640 to 40 trees/ha), while weight per hectare varied from 1900 to 200 kg/ha respectively. Elsewhere it has been reported that an individual large tree may retain 14 kg of leaf (Pressland 1975), but this is in a low density situation.

Table 4. Relationship between browse and herbage yield for an acacia aneura community in Queensland

	Tree Density (No./ha)
	40	160	640
Foliage weight (hg./tree)	5	4	3
Foliage weight (kg/ha)	200	640	1900
Herbage weight (kg/ha)	750	400	100

The standing crop of leaf in a mixed species situation (Eucalyptus populnea semi-arid woodland) was found to be 300 kg/ha, but only 290 kg/ha of this was palatable to sheep (see Table 5) (Harrington, 1979). In chenopodiaceous shrubsteppe communities (Atriplex vesicaria), the standing crop of leaf can vary from 200 to 500 kg ha.

Table 5. Foliage weight of trees and shrubs in an undisturbed eucalyptus populnea semi arid. Woodland (Harrington 1979) and three years after the clearing of all shrubs

Species	Foliage weight (kg/ha)
	Undisturbed	3 years regrowth
Acacia aneura	176	15
Cassia nemophila var platypoda	244	640
Dodonea viscose	18	6
Geijera parviora	616	62
Heterodendrum oleifolium	78	11
Myoporum desertii	36	16
Unpalatable species	2206	158
TOTAL	3174	908

These yields of standing crop cannot be directly compared to herbage yields. The standing crop of herbage may have been accumulated in only a few months and under continued grazing, while that of the trees may have accumulated without browsing and over several years. Hence the herbage weights presented in Table 4 should not be compared directly with the browse weights.

Another way of assessing browse production is to measure leaf fall. For mature stands, foliation can be assumed to be constant, with growth equivalent to leaf shedding. Australian browse plants fall into two groups: those that grow on a seasonal basis, usually in spring and early summer, and those that grow after rainfall (Maconochie,1973). However, regardless of group, foliation and defoliation is synchronous (Maconochie and Lange, 1970). Leaf fall occurs steadily throughout the year, but with a marked peak in early summer (Leigh et al 1978). The amount of leaf fall from the trees in an arid Casuarina woodland is shown in Table 6, and amounted to 180 kg/ha/year. Burrows (pers. comm.) reports that leaf fall from an Acacia aneura community would seldom exceed 500 kg/ha/year.

Table 6. Leaf fall from fodder trees in an arid casuarina cristata woodland

Species	Weight		Crude Protein	Digestibility (in vitro %)
	Kg/tree	kg/ha
Acacia homalophylla	1.3	1	10	33
Casuarina cristata	1.7	113	9	29
Geijera parviora	5.7	21	13	59
Heterodendrum oleifolium	4.0	48	11	41

Source: Leigh, Wilson, Mulham (1978)

Leaf fall is of low nutritive value to sheep, both because it is of lower quality than fresh leaves (Leigh et al, 1978) and because it is of low palatability. Sometimes sheep will eat the leaf-fall of trees that are normally considered inedible, but only under conditions of low forage availability, and their intake is then below maintenance (Beale and McMenimen, 1978, Harrington 1980) Leaf-fall is more readily eaten by goats.

3.2 Reduction of herbage yield

In assessing browse production, we must also assess the reduction in yield of herbage plants, as the density of browse plants is increased. An example of this is the data in Table 4 from Beale (1973), which shows a reduction in herbage yield from about 750 kg/ha to about 100 kg/ha, as the standing crop of tree leaf increases from 200 to 1900 kg/ha. The reduction in herbage yield can be readily assessed by the equation presented by Beale (1973) of herbage yield (g/m²) = 9.48 + 8.84 × 0.098^x (x = basal area of trees in m²/ha). This was derived from an Acacia aneura community, but the relationship for other semi-arid communities is similar. As as consequence, Pressland (1975), working in south-west Queensland, recommends that the major portion of each farm be kept clear of trees, leaving a small portion as a drought reserve.

3.3 Animal production

Several aspects of animal production on shrub communities has already been mentioned. Detailed information on the production to be obtained in comparative situations, with and without browse species, is generally not available. Some information for sheep on chenopod shrub communities, including A. nummularia was presented by Leigh et al (1970). They concluded that shrubs are useful as a drought reserve, but that the net gain is not large. In their trials the extra return was insufficient to warrant the planting of Atriplex spp. into a semi-arid grassland.

In all Australia arid and semi-arid lands, cattle, sheep and goats browse selected species as a minor part of their diet throughout the year and increase their intake of browse as herbage availability declines. (Wilson et al, 1969; Harrington 1980a, 1980b). In view of the low nutritive value of the browse, it is probable that it only achieves significance under these conditions of nutritional stress, and then only for survival purposes. Under such circumstances the value of browse must be considered against the costs and possibilities of stock migration.

In climatic zones where herbage falls to a low nutritive value on a seasonal basis, the use of browse may be justified, but only a few Australian species are sufficiently nutritious to justify specific introductions. A low level of browse throughout the grazing land (for instance a browse to herbage ratio of 1:20) is useful for short-term herbage shortages. If a higher ratio is required for long term use (for instance a ratio of 1:3), then it would be better to set aside reserves. Such a system would minimize drought damage to herbage plants and the costs and benefits could be accurately assessed. Nevertheless, long-term lowered herbage production on large browse reserves is a big price to pay for the infrequent benefit.

The broad generalization is that animal production is greatest with a high ratio of herbage to browse production. Regardless of whether they occur as scattered trees or in a plantation, browse plants represent a survival reserve.

4. Browse management

We have presented the conclusion that open grassland, with scattered browse plants, is a desirable vegetation composition, although it is unlikely that Australian plants could be introduced and managed in such a way in African grasslands. Even in Australian situations, this desirable composition is difficult to attain. In the arid areas, trees may have difficulty in regeneration and the added burden of browsing leads to their elimination. In contrast an overdense shrub and tree population, is common in the semi-arid Acacia aneura and Eucalyptus savanna woodlands.

In these situations it is desirable to clear the tree layer. If this is done by mechanical means it is usual to clear the trees in strips or on a wide face, and leave other areas or strips of dense trees as drought reserves (Beale 1973, Burrows 1979). However, it is desirable to manage tree and shrub density by lower cost procedures, such as the control of grazing, or by fire. Sheep grazing alone has a marked effect on the regeneration and growth of young Acacia aneura and may eliminate the regeneration of Heterodendrum oleifolium. Heavy grazing by sheep can quickly reduce the density of Atriplex vesicaria to undesirably low densities. Hence, the intensity of sheep grazing can be manipulated to some effect, usually on young regenerating populations. Cattle may have different effects because of their ability to browse to a greater height. Their grazing on young A. aneura lead to a more shrubby habit, with a greater proportion of leaf within browsing height. Burning is also a potent force in the manipulation of shrub density, and may be encouraged where browse is overdense or avoided where shrubs should be retained (e.g. Atriplex communities).

5. Conclusions

In Australia browse plants are used and managed only within their natural habitats. They are never planted or nurtured in plantations. Nevertheless, the Australian shrub Atriplex nummularia has been selected for higher palatability and lower salt content within South Africa. It has been highly recommended as a drought reserve fodder crop (De Kock 1967) and planted in some of their arid regions (Hugo, 1968). Such plantations are possibly the only way that an introduced browse plant could be exploited. In such situations, social and economic factors are equally as important as nutritive value, and have so far been ignored. In general, we believe that the productivity of arid communities is too low and the improvement arising from the inclusion of browse plants is too small to warrant consideration of plants that do not regenerate naturally. More progress can be made by learning to manage the native vegetation communities including their browse plants component.

References

Beale, I.F. (1973). 'Tree density effects on yields of herbage and tree components in south west Queensland Mulga (Acacia aneura F. Muell.) scrub'. Trop. Grassi. 7, 135–42.

Beale, I.F.; McMenimen, N.P. (1978). 'Measurement of dietary intake of phosphorus by grazing sheep using 32 p, Proc. 1st Inter. Rangel. Congr. 429–30.

Burrows, W.H. (1979). 'Vegetation management decision in Queensland's semi-arid sheeplands'. In: K.M.W. Howes (ed.). Rangeland Ecosystem Evaluation and Management. Perth, Australian Rangeland Society.

Burrows, W.H.; Beale, I.F. (1970). 'Dimension and production relations of mulga (Acacia aneura F. Muell.) trees in semi-arid Queensland'. Proceedings 11th International Grassland Congress. Surfers Paradise.

De Kock, G.C. (1967). 'Drought resistant fodder crops'. Proceedings Grassland Society South Africa. 2, 147–56.

Entwistle, K.W.; Baird, D.A. (1976). 'Studies on the supplementary feeding of sheep consuming mulga (Acacia aneura). 2. Comparative levels of molasses and urea supplements fed under pen conditions'. Aust. J. Exp. Agric. Anim. Husb. 16, 174–80.

Everist, S.L. (1967). 'Fodder trees in the sheep lands'. Queensl. Agric. J. 93, 540–6.

Gartner, R.J.W.; Hurwood, I.S. (1967). 'Tannin and oxalic acid contents of Acacia aneura (mulga) and their possible effect on sulphur and calcium availability'. Aust. Vet. J. 52, 194–6.

Harrington, G.N. (1979). 'Estimation of above-ground biomass of trees and shrubs in a Eucalyptus populnea F. Muell. woodland by regression of mass on trunk diameter and plant height'. Aust. J. Bot. 27, 135–43.

Harrington, G.N. (1980a). 'Diet of herbivores in a semi-arid Eucalyptus populnea woodland. I. Sheep'. (In preparation).

Harrington, G.N. (1980b). 'Diet of herbivores in a semi-arid Eucalyptus populnea woodland. II. Feral goats'. (In preparation).

Harvey, J.H. (1952). 'The nutritive value of some Queensland fodders'. Queensl. J. Agric. Sci. 9, 169–184

Hugo, W.J. (1968). 'The Small Stock Industry in South Africa. Pretoria, Government Printer.

Leigh, J.H.; Wilson, A.D.; Mulham, W.E. (1978). 'Seasonal variations in the leaf fall and quality of the leaves of four Australian fodder trees'. Aust. Rangel. J. 1(2), 137–41.

Leigh, J.H., Wilson, A.D., Mulham, W.E. (1979). 'A study of sheep grazing on a Belah (Casuarina-cristata)— Rosewood (Heterodendrum oleifolium) woodland in western New South Wales'. Aust. J. Agric. Res. 30, 1223–36.

Leigh, J.H., Wilson, A.D., Williams, O.B. (1970). 'An assessment of the value of three perennial chenopodiaceous shrubs, for wool production of sheep grazing semi-arid pastures'. Proceedings 11th International Grassland Congress Surfers Paradise.

McDonald, W.J.F.; Ternouth, J.H. (1979). 'Laboratory analyses of the nutritional value of western Queensland browse feeds'. Aust. J. Exp. Agric. Anim. Husb. 19, 344–49.

McLeod, M.N. (1973). 'The digestibility of the nitrogen, phosphorus and ash contents of the leaves of some Australian trees and shrubs'. Aust. J. Exp. Agric. Anim. Husb.13, 245–50.

McMenimen, N.P. (1976). 'Studies on the supplementary feeding of sheep consuming mulga (Acacia aneura). 3. The provision of phosphorus, molasses and urea supplements under pen conditions'. Aust. J. Exp. Agric. Anim. Husb. 16, 818–22.

McMenimen, N.P., Little, D.A. (1974). 'Studies on the supplementary feeding of sheep consuming mulga (Acacia aneura), I. The provision of phosphorus and molasses supplements under grazing conditions'. Aust. J. Exp. Agric. Anim. Husb. 14, 316–21.

Maconochie, J.R. (1973). 'Leaf and shoot growth on Acacia kempeana. F. Muell. and selected other arid-zone species'. Trop. Grassl. 7, 49–55.

Maconochie, J.R., Lange, R.T. (1970). 'Canopy dynamics of trees and shrubs with particular reference to arid-zone topfeed species'. Trans. Roy. Soc. S. Aust. 94, 243–8.

Pressland, A.J. (1975). 'Productivity and management of mulga in south-western Queensland in relation to tree structure and density'. Aust. J. Bot. 23, 965–76.

Rohan-Jones, W.G. et al (1972). 'Measurements of energy metabolism in sheep fed kurrajong (Brachychiton populnea), mulga (Acacia aneura) and native pasture (Stipa spp.)'. Proceedings Australian Society Animal Production. 9, 341–5.

Siebert, B.D., Newman, D.M.R.; Nelson, D.J. (1968). 'The chemical composition of some arid zone pasture species'. Trop. Grassl. 2, 31–40.

Wilson, A.D. (1966). 'The value of Atriplex (saltbush) and Kochia (bluebush) species as food for sheep'. Aust. J. Agric. Res. 17, 147–53.

Wilson, A.D. (1974). 'Water consumption and water turnover of sheep grazing semi-arid pasture communites in New South Wales'. Aust. J. Agric. Rest. 25, 339–47.

Wilson, A.D. (1977). 'The digestibility and voluntary intake of the leaves of trees and shrubs by sheep and goats'. Aust. J. Agric. Res. 28, 501–8.

Wilson, A.D., Graetz, E.D. (1979). 'Management of the semi-arid rangelands of Australia'. In: B.H. Walker (ed.). Management of Semi-arid Ecosystems. Amsterdam, Elsevier.

Wilson, A.D., Leigh, J.H.; Mulham, W.E. (1969). 'A study of Merino sheep grazing a bladder saltbush (Atriplex vesicaria)—cotton bush (Kochia aphylla) community on the Riverine Plain'. Aust. J. Agric. Res. 20, 1123–36.