Graham N. Harrington and Allan D. Wilson
CSIRO Division of land resources management
Deniliquin, N.S.W. 2710, Australia
2. The place of browse in the production system
4. Techniques for partial evaluation of secondary productivity from browse
4.1 Nutritional deficiencies in the system.
4.3 Absolute palatability and limits to intake.
4.4 Primary productivity of browse relative to the demands of the system.
4.5 Interaction of browse with herbage production.
4.6 Intake and chemical analysis.
4.8 Integration of the value of the browse into the production system.
Secondary production is a term which embraces all products from animals which feed on plants. These products may or may not be useful to humans and frequently only useful products will be considered when the value of a forage or an animal production system is being considered. Secondary production is the ultimate measure of the true nutritive value of the fodder eaten whether it includes all or a limited range of products.
Direct measurement of secondary production is logistically difficult because large numbers of animals are required for statistical purposes; the animals need to be in an environment which is realistically related to the production system for which the research is undertaken; and because production is usually simultaneously expressed in a variety of continuous variables such as liveweight gain, wool growth and milk production and sometimes in binomially distributed variables such as reproduction and mortality. Browse is usually only a part of the total diet so the value of the browse fraction cannot be gauged from measurement of secondary production directly, whilst for some browse fodders the indirect methods developed for assessing the chemical content and digestibility of herbage require adjustment (e.g. Wilson, 1977).
Secondary production from browse must first be assessed in terms of the role of browse within the whole production system on a year-round or longer-term basis. Normally browse is supplementary to a herbage-based system. For example, it may serve to lift the protein intake for all or part of the year or it may be essential that the animals subsist on browse during cold or dry periods so that they survive to utilize and breed on the herbage at other times of the year. Without such a conceptual model of where browse fits into the total production system, the real value of browse in terms of secondary production cannot be investigated effectively.
Any evaluation of browse will be relative to alternative fodders (usually herbage), and normally browse will be found to be available year-round but only intensively utilized when herbage is limiting in quantity or when the protein level in the herbage drops below that required for maintenance of the animal. Despite this supplementary role to a varying herbage source, browse itself varies in quality and quantity at different seasons and young growth is frequently more nutritious and/or palatable than mature growth (Short, Blair and Burkart, 1972). Thus a knowledge of the phenology of the browse plants may be an important part of the conceptual model of the production system.
Browse also frequently competes with herbage for light, moisture and nutrients. The role it plays in the nutrition of the animals must be assessed in terms of its competition with, possibly, more nutritious plants. In many systems, browse provides a stabilising influence on the nutrition of the animals by providing maintenance or subsistence rations when herbage is deficient. However, too many browse plants will depress overall secondary productivity by competing with the herbage on which that productivity depends.
In the evaluation process it must be recognized that eating browse may either reduce or increase the consumption of alternative fodder, by replacement or by stimulating the appetite for other fodders. These factors need to be measured. Also, the value of browse may be expressed through one nutrient only and frequently browse will be important because it increases the nitrogen content of a diet.
Yet another role for browse is that secondary productivity from mixed herds of browsing and grazing herbivores can be greater than monospecific herds without applying additional stress to the ecosystem, e.g. Denney (1972).
In summary, whether the production system of interest is a natural one, involves the exploitation of range pastures or incorporates plantations of browse, the assessment of the influence of the browse on secondary productivity should be made in terms of the place it holds in the long term (including such factors as survival of infrequent droughts) and how it affects all aspects of secondary productivity (growth, wool and milk production, reproduction, year-round carrying capacity, survival etc). The evaluation should not be confined to short-term studies of intake, digestibility and immediate animal growth response and it will be made in terms of the secondary products that are affected. Such an evaluation will attempt to compare or simulate the production with and without the browse in question and implies a nutrient deficiency within the system in the absence of browse.
The ideal evaluation is to test the production system directly with and without the browse. This is rarely possible in a simple way. With plantation browse, such a test is only the final stage of various partial and simulated tests while alternative browse plants are compared for their suitability. For range pasture conditions, some tests have been made on areas from which browse has been removed or added artificially (e.g. Leigh, Wilson and Williams, 1970) but these were trial grazings and not whole system assessments. In natural systems no absolute test is possible. In all types of production systems seasonality in the use of browse is a major problem in the evaluation process.
Thus for assessment of secondary productivity from any type of production system, some type of partial evaluation is normally undertaken and inferences drawn at various stages during the tests, in order to cut down the time and expense involved.
Partial evaluation will include some or all of the following stages but not necessarily in the order presented, and the strategy adopted will be dictated by the investigator's understanding of the production system (what needs to be quantified and what can be merely estimated or inferred):
These have been well reviewed by Corbett (1978) and will only be summarised here.
All systems of forage production and utilization vegetation provide a varying plane of nutrition. The first stage in evaluation is to subjectively estimate whether any aspect of secondary production is seriously limited and if so whether browse offers a realistic change of relieving the nutritional deficiency.
Monitoring liveweight of animals through the year will indicate the variations in the nutritional plane although they may be more crucially expressed by limitations to the reproduction rate, wool growth, milk production etc. Nutritional deficiency will not always affect the factor of secondary production of interest: for example a low plane of nutrition during pregnancy may have no effect on the overall rate of reproduction and a minimal effect on weight of offspring at weaning.
Given a free choice, herbivores exhibit marked preferences amongst the foods available, but will rarely eat one food to the exclusion of all others (e.g. Wilson, 1977). Repeatable results can be obtained in quantitatively measuring the species composition of the diet selected from any one pasture, but on a different pasture, or on the "same" pasture at a different time of year, any one food may form a quite different fraction of the total diet. Thus palatability is relative to the alternative foods available and assessment of palatability and intake in an environment, which does not offer the same foods in the same condition as they will be found within the production system itself, will give an erroneous idea of its potential nutritive value.
Similarly, it cannot be concluded that an uncommon shrub in the open range which is browsed by livestock would necessarily prove to be useful if it was increased artificially. Uncommon shrubs are browsed for a variety of reasons unconnected with their palatability.
To assess the nutritive value of any one species of browse it is necessary to know the fraction of the total food intake which it comprises at the different seasons of the year. This may be assessed by the measurement of browse material removed from that on offer either in a stall or on the range (e.g. Cook, Harris and Stoddart, 1948; Skinner and Telfer, 1974) over a specified period. Adequate sampling under range conditions make such methods arduous and animal-centred observations are usually preferred. The intake and relative palatability of plant mixtures has also been obtained by "eating down" experiments (e.g. Leigh and Mulham, 1966). Animals are enclosed on small areas and the qualitative and quantitative consumption of each plant species is measured daily by measuring the changes in the plant material present.
Direct observation of animals and counting the bites of each plant or class species (Buechner, 1950) is the easiest way of collecting dietary data and is particularly suited to browse studies because browse is readily visible from a distance. Corrections are needed to account for the different size of bite on different classes of plant or samples in time may be substituted for bites (Harrington, 1978). Only approximate quantitative measures are achieved by these methods but where interest centres only on assessing the fraction of one or a few browse species in the whole diet, and where variation in the fraction may be considerable from year to year, the level of accuracy is often sufficient to be useful.
Collecting samples of the diet via an oesphageal or rumen fistula (Van Dyne and Torell, 1964) is usually considered the most accurate means of quantitatively assessing the diet of herbivores. However, certain drawbacks are apparent. Delicate plant parts disintegrate and may not be recovered in the same ratio as they were eaten. This particularly applies to rumen fistulae but oesophageal samples also suffer from this drawback (Marshall & Squires, 1978). Generally, browse is not delicate but young growing shoots might not be recognised during macroscopic separation of the sample. Collecting samples with fistulated animals is time consuming, the sample of the total grazing day is small and on open rangeland it is often impossible to simulate the feeding behaviour of the normal animals. Thus the accuracy of the fistula method is to some degree spurious and may not be worth the additional trouble over direct observation, unless chemical and digestibility analyses of the whole diet are also of interest. Such analyses can be carried out on the fistula samples in toto, but not on the separate species making up that sample, because the more delicate and most digestible portions are not recognisable.
Most browse material can be identified macroscopically from oesophageal fistula extrusa (e.g. Leigh and Mulham, 1966) but this is not the case with gut contents obtained from shot animals or faeces. In these cases microscopic examination of suspensions of the gut contents has been used to identify the epidermis of the plant species eaten and by using a point sampling procedure the relative area of epidermis of each species can be calculated. In order to convert this to dry matter ingested however, stall feeding trials must be undertaken to assess the differential breakdown of the epidermis of the species being compared (Stewart, 1967). The whole procedure is so time consuming that this would only be the method of choice if no other alternative was available. The method is useful for establishing whether a browse species is eaten and the changes in its use with season (i.e. qualitative assessment), provided that there are cells of the browse plant which are distinguishable from the cells of all the other plants present (Ellis et al, 1977).
Browse plants range in palatability from being completely inedible to being so palatable that they do not persist under unrestricted use. Palatability of plant species varies between species of herbivore. Most domestic stock and large herbivores are generalist herbivores, and may not thrive on a monospecific diet. The intake of browse plants is frequently lower than expected because of secondary compounds such as resins, terpenes and volatile oils, which are present. Unlike herbage plants, in which digestibility and intake are strongly correlated, there are often severe limits to the intake of highly digestible browse such as Atriplex nummularia and Geijera parviflora, whilst indigestible browse may have high palatability (Wilson, 1977). Absolute palatability may vary with seasonal changes in the chemical composition of the plant (Cook, 1972).
The measurement of limitation to intake is important when calculating the potential nutritive value of a browse plant to be used as a plantation forage crop and can be done simply using stall-fed animals. If the anticipated production system includes forages other than the browse under test, these forages should be offered together with decreasing amounts of these other forages to establish the level of maximum intake.
Under range feeding the absolute limits to intake of any one food will be less important than relative palatability. However, in some animal production systems, abnormally poor seasons can reduce the food supply, so that browse that is of relatively low palatability is the only forage available. In such a case it is important to know the absolute limits to intake. With the use of browse plantations absolute limits to intake are a vital part of the assessment process.
Measurement of the primary production from browse is the subject of another paper in this symposium by J.C. Bille. However, the estimation of the potential demands for browse dry matter by the animals is important in the husbanding of the browse plants for survival and vigour.
Trees and shrubs growing within the pasture will affect herbage production, both qualitatively and quantitatively. Reduction in tree or shrub biomass will normally increase herbage production although the relationship is quadratic and a measurable influence may not be apparent below a certain level of shrub/tree biomass, e.g. Beale (1973). The herbage species growing close to trees and shrubs are usually different species to those growing in open situations. They have different nutritional characteristics and may also have a different growing period.
These factors are important components in the nutritional system and need measuring or estimating because they will influence the overall effect of the browse on secondary productivity.
The chemical analysis of a plant will indicate the gross amounts of the various nutrients present. These values are only useful when the intake and digestibility of the plant is known so that the metabolizable levels can be calculated. As both the gross nutrient levels and their digestibilities vary independently between different parts of the plant and according to the age of the plant material, the plant samples collected for analysis must be representative of the diet selected by the herbivore.
Herbivores frequently eat young growing shoot tips or young leaves, rather than mature material, and stems which have undergone secondary thickening are usually avoided. The digestibility of browse DM decreases rapidly with age and this particularly affects the metabolizable energy levels (Short, Blair and Burkart, 1972). Slight discrepancies in sampling too young or too mature material may cause quite erroneous calculations of the fraction of the secondary productivity that is derived from browse (Rutherford, 1979).
Removal of browse material from trees and bushes by animals may be assessed by marking sample shoots on which leaves have been counted and the length of shoot measured. Where whole shoots are being eaten the mass of material removed may be calculated directly from the diameter of the bitten-off twig because twig diameter is often highly correlated with twig mass (Basile and Hutchings, 1966). The quality of that material may be assessed by collecting ungrazed twigs of the same diameter. The accuracy may be improved by protecting parts of the bush from browsing and comparing twigs paired for their similar phenology.
The collection of browse material for chemical analysis via the oesophageal fistula is only relevant when the browse is the sole constituent of the diet.
Where a browse species is being assessed for its nutritive value, the gross intake of that species must be measured and expressed as a fraction of the whole food intake. This may be directly assessed by measuring the depletion of the food on offer over time by a known number of animals or by indirect methods on the grazing animals themselves. Indirect methods have been reviewed by Cordova, Wallace and Pieper (1978) and they concluded that all the methods available "lack precision and are often tedious, expensive and time-consuming". Harris (1972) suggests that indirect methods are unsuitable for complex diets that include both herbs and browse because all methods, which involve the measurement of the concentration of an indigestible indicator in the faeces, require stall-fed trials of animals on the same diet, in order that corrections for the digestibility of the feed can be made. The collection by hand of large quantities of plant material for such trials is both time consuming and prone to considerable error even if guidance from oesophageal samples is obtained. It must be accepted that only in the case of simple diets where the browse plant is the major source of food, will gross intakes be measured accurately.
Having established the gross intake and part of the plant that is being eaten, it is necessary to measure the digestible crude protein and metabolizable energy values of the ingested material. In some cases other nutrients such as carotene, phosphorus, or calcium may also be of interest but it is rare for such substances to be limiting animal production in grazing systems, if only because they are not deficient for very long, and even less likely that browse will eliminate the deficiency. Browse may be deficient in these substances however and a long period of dependence on browse may require these deficiencies to be rectified if secondary productivity is not to suffer.
The digestibility of energy and crude protein may be measured (in vivo techniques). However, this is a time-consuming process and there are problems in getting animals to eat one species of browse for long enough to carry out the test (Wilson, 1977). Harvesting browse of the quality normally eaten by stock is physically arduous and prone to error. Drying the fodder may affect its overall nutritional quality and provision of fresh browse is even more difficult. Finally the animals concerned must be amenable to life in stalls. Digestion stall trials are rarely a practicable proposition for browse assessment (Dietz, 1972).
The determination of digestibility by in vitro and chemical techniques (Tilley and Terry 1963; Van Soest 1967) can overcome these logistical problems. They have been developed for herbage plants and although the methods have proved suitable for some browse species (Dietz, 1972) they have proved markedly wrong for others and must therefore be used with caution if in vivo checks are not possible (Wilson, 1977). Lignin: cell wall ratios and lignin-protection of carbohydrate and protein from digestion in browse appears to be different from that in herbage (Short, Blair and Burkart, 1972) and there is a need for in vitro digestibility techniques to be developed specifically for browse. Standard equations for estimating the digestibility of crude protein in herbage (Milford and Minson, 1965) are not applicable to browse. Wilson (1977) found a wide variation in true digestibility of crude protein, in different browse species, which was not related to their crude protein content. It should also be noted that the digestible energy obtainable from shrubs cannot always be corrected to metabolizable energy by a standard correction for losses in urine and methane. High energy oily constituents in some shrubs may be excreted in the urine and give a spuriously high metabolizable energy value unless energy losses in the urine are actually measured (Cook, Stoddart and Harris, 1952).
Figures obtained in the partial analysis can be used to help evaluate the browse within the conceptual model of the production system. Where manipulations. of the browse in the system or of the animals which use the browse seem appropriate a numerical model may be necessary. It is emphasized that the evaluation will concern only the nutrients which are deficient and in terms of the factor of production which is affected.
The role of browse in supporting secondary production is usually supplementary to herbage although specialist browsing animals are sometimes of interest. Given that the methods of assessing the intake and true digestibility of browse are inexact, and accepting that the use made of browse is going to vary from year to year according to variations in alternative and more palatable forages, it is not worthwhile to go to great lengths to make accurate partial measurements of the nutritive value of browse to the production system. Far more important is to establish in general terms the fraction of the diet devoted to browse season by season and to assess, again in general terms, whether secondary productivity is satisfactory at that time. Where there appears to be room for improvement it must then be decided whether there is an alternative plant which will improve the plane of nutrition without reducing it at other seasons of the year or being critically less reliable in times of stress. It is vital to take a holistic view of the whole system when assessing the value of browse to secondary production. Whilst being generally inferior to herbage in terms of primary production and digestible nutrients, browse is often less affected by seasons and climatic extremes in terms of the quality of forage available, it is common for browse plants to produce new shoots or leaves when herbage is dormant and it is perennial in environments that are too harsh for perennial herbs. As such it is a stabilizing factor in many production systems enabling animals to survive adverse periods in order to capitalise on better seasons when higher quality is available. It is worthy of greater investigation than it has hitherto received.
Basile, J.V., Hutchings, S.S. (1966). 'Twig diameter—length–weight relations of bitterbrush'. J. Range Manage. 19: 34–38.
Beale, I.F. (1973). 'Tree density effects on yields of herbage and tree components in south west Queenslands mulga (Acacia aneura F. Muell.) scrub'. Trop. Grassi. 7: 135–42.
Buechner, H.K. (1950). 'Life history, ecology and range use of the Pronghorn Antelope in Trans-Pecos, Texas'. Am. Midl. Nat. 43: 257–354.
Chippendale, G.M) (1962). 'Methods of investigation of fodder preferences under open range conditions'. Advancing Frontiers of Plant Sciences. 1: 23–30.
Cook, C.W. (1972). 'Comparative nutritive values of forbs, grasses and shrubs'. In: Wildland Shrubs—Their Biology and Utilization. Ogden, Utah, USDA Forest Service. Gen, Tech. Rep. INT-1. pp. 303–310.
Cook; C,W., Harris, L.E. and Stoddart, L.A (1952)). 'Measuring the 'nutritive content of a foraging sheep's diet under range conditions'. J. Anim. Sci. 7: 170–80.
Cook, C.W., Stoddart, L.A. and Harris, L.E. (1952)). 'Determining the digestibility and metabolizable energy of winter range plants by sheep'. J. Anim. Sci. 11: 578–90.
Corbett, J.L. (1978)). 'Measuring animal performance'. In: Measurement of Grassland Vegetation and Animal Production Chap. 7, 163-231. Ed. L. t'Mannetje. Hurley, Commonwealth Bureau of Pastures & Field Crops. Bull. 52, 260 p.
Cordova, F.J., Wallace, J.D. and Pieper, R.D. (1978)). 'Forage intake by grazing livestock: a review'. J. Range Manage. 31:430–38.
Deney, R.N. (1972)). 'Relationship of wildlife to livestock in some developed ranches in the Laikipia Plateau, Kenya'. J. Range Manage, 25: 415–25.
Dietz, D.R. (1972). Wildland shrubs—their biology and utilisation. Ogden, Utah, USDA Forest. Service. Gen. Tech. Rep. Int-1. 289–302.
Ellis, B.A., et al (1977)). 'Seasonal changes in diet preferences of free ranging red kangaroos, euros and sheep in western N.S.W.'. Aust. Wildl. Res. 4: 127–04.
Harrington, G.N. (1978)). 'The implications of goat, sheep and cattle diet to the management of an Australian semiarid woodland'. Proc. 1st Ranges. Congr. pp. 447–450.
Harris, L.E. (1972)). 'Physiological problems in animal use of shrubs as forage'. In: Wildland Shrubs—Their Biology and Utilization. Ogden, Utah, USDA Forest Service. Gen. Tech. Rep. Int-1. pp. 319–330.
Leigh, H.H. and Mulham, W.E. (1966)). 'Selection of diet by sheep grazing semiarid pastures on the Riverine Plain. 1. A bladder saltbush (Atriplex vesicaria—cotton bush (Kochia aphylla) community'. Aust. J. Exp. Agric. Anim. Husb. 6: 460–67.
Leigh, J.H., Wilson, A.D. and Williams, O.B. (1970)). 'An assessment of the value of three perennial chenopidiaceous shrubs for wool production of sheep grazing semiarid pastures'. Proc. XIth Int. Grassl. Congr. pp. 55–59.
Milford, R. and Minson, D.J. (1965)). 'The relation between crude protein content and digestible crude protein content of tropical pasture plants'. J. Br. Grassl. Soc. 20: 177–79.
Rutherford, M.C. (1979)). 'Plant-based techniques for determining available browse and browse utilization: a review'. Bot. Rev. 45: 203–28.
Skinner, W.R. and Telfer, E.S. (1974)). 'Spring, summer and fall foods of deer in New Brunswick'. J. Wildl. Manage. 38: 210-14.
Short, H.L.; Blair, R.M. and Burkart, L. (1972)). 'Factors affecting nutritive values'. In Wildland Shrubs — Their Biology and Utilization. Ogden, Utah, USDA Forest Service. Gen. Tech. Rep. INT-1 pp. 311–318.
Stewart, D.R.M. and Terry, R.A. (1963)). 'A two-stage technique for the in vitro digestion of forage crops'. J. Br. Grassl. Soc. 18: 104–111.).
Van Dyne, G.M. and Torell, D.T. (1964)). 'Development and use of the esophageal fistula: a review'. J. Range Manage. 17: 7–19.
Van Soest, P.J. (1967)). 'Development of a comprehensive system of feed analyses and its application to forages'. J. Anim. Sci. 26: 119–28.
Wilson, A.D. (1969)). 'A review of browse in the nutrition of grazing animals'. J. Range Manage. 22: 23–28.
Wilson, A.D. (1977)). 'The digestibility and voluntary intake of the leaves of trees and shrubs by sheep and goats'. Aust. J. Agric. Res. 58: 501–8.
