C.T.White Memorial Lecture for 2002


Ross A. McKenzie1

Animal & Plant Health Service, Department of Primary Industries Queensland and School of Veterinary Science, University of Queensland


Cyril Tenison White (1890-1950) (Fig. 1) died 52 years ago in the year after I was born, so I know him only from his published work and that of others in the field of plant poisonings of livestock. Never-the-less, I honour his memory as a man who recorded many incidents of poisoning by plants in Queensland and who tried to inform livestock producers about the hazards faced by their animals from poisonous plants to prevent further disease2 and death. As Queensland Government Botanist for 33 years from 1917 to his death in 1950 (Blake 1952, Orchard 1999), he recorded in his annual reports to the Queensland Parliament many cases of poisoning of domestic animals linked to plants. As the main basis for this paper, I have used the work of C.T.White and his collaborators published in the Queensland Agricultural Journal, the Australian Veterinary Journal and in Annual Reports of the Government Botanist for Queensland. To trace the plant poisoning-related papers of C.T. White, I began with Hurst (1942) and Everist (1981) and the memorial to C.T. White penned by Stan Blake (Blake 1952) that includes a near-complete bibliography.

The text of this paper is a referenced, slightly modified and expanded version of the lecture delivered on 17 June 2002 as the 51st C.T.White Memorial Lecture.

C.T.White’s toxicological knowledge base and contributions

A little learning is a dang’rous thing;

Drink deep, or taste not the Pierian spring:

There shallow draughts intoxicate the brain,

And drinking largely sobers us again.

From An Essay on Criticism. Alexander Pope 1711

Cyril Tenison White (1890-1950)

Figure 1. Cyril Tenison White (1890-1950). Queensland Government Botanist (1917-1950) [Image used with permission of Queensland Herbarium]

C.T. White benefited from the knowledge developed by his maternal grandfather Frederick Manson Bailey, Colonial Botanist of Queensland, under whom he worked from 1905. Bailey’s publications - Bailey & Gordon (1887) with the then Chief Inspector of Stock for Queensland, Patrick Robinson Gordon, and Bailey (1906) – provide evidence of the limited extent of that knowledge compared with ours. He would have had access to publications from other countries on poisonous plants such as Long (1917, 1924) from Britain and Steyn (1934) from South Africa with which to study the effects of plants introduced to Australia from Europe and southern Africa and to compare those of related Australian taxa. He recognised the lack of information available, writing in his first report of the (Acting) Government Botanist (White 1917): “Every year a host of specimens are received suspected as being the cause of sickness or deaths amongst stock. In only a few instances can a definite reply be given, our lack of definite knowledge on this most important subject being deplorable.” The expertise subsequently developed by C.T.White was nationally recognised and he was acknowledged for “helpful suggestions” and “reading the proofs” of the pamphlet produced by “The [New South Wales] Poison Plants Committee” (Finnemore et al. 1934) describing 18 poisonous plants of northern Australia. He was an inaugural member of the Queensland Poisonous Plants Committee, founded in 1937, on which he served with the veterinarians H.R. (Bert) Seddon (founder and chairman), A.H. Cory, J.A. Rudd and John Legg, and the chemists E.H. Gurney and H.J.G. Hines (McKenzie 1995).

The state of knowledge of toxic hazards in the Queensland flora at C.T. White’s death is reflected in the compilation by Len Webb on the medicinal and poisonous plants of Queensland (Webb 1948) and in the more strictly-assessed review of Queensland poisonous plants by Everist (1962). The growth of this knowledge can be very roughly assessed by comparing the 54 reputed poisonous plant taxa included in Bailey & Gordon (1887) with the 645 included in Webb (1948) and the 486 included in Everist (1962) - a ten-fold increase.

C.T. White contributed to animal welfare and production in Queensland by: identifying new species of toxic plants in conjunction with his veterinary and chemist colleagues; publishing accounts of known toxic plants for livestock producers in the Queensland Agricultural Journal and for politicians and others in the Annual Reports of the Queensland Government Botanist; and answering numerous enquiries from the public on the identity and properties of toxic plants, as evidenced by the numerous “Answers to Correspondence” pages of the Queensland Agricultural Journal.

An early and major contribution was botanical support for surveys of cyanogenic glycoside-containing plants with chemist Frank Smith (Smith & White 1918, 1920a). Plants in which the capacity for cyanide production was detected included Grevillea banksii (Fig.2), but this plant has not been recorded as poisoning domestic animals. Cyanide generated in the gastrointestinal tract when these plants are eaten will kill sheep, cattle and other ruminants very quickly.

Grevillea banksii R.Br

Figure 2. Grevillea banksii R.Br. A flowering shoot. Cyanogenic glycosides were detected in the flowers of this taxon by Smith & White (1920a). [Fresh specimen directly scanned by R.A.McKenzie]

From his publications, C.T.White appeared to be familiar with a number of toxins known from plants. The most prominent of these were cyanide, alkaloids (in a general sense), and saponins. In fact, cyanide so dominated the imagination then that it was suggested as the cause of poisoning by plants which today we know poison through other quite different toxins. The discovery of plant nitrate converted to nitrite as toxic for ruminants was made during C.T White’s working life by Rimington & Quin (1933a,b) in South Africa while researching Tribulus terrestris ‘geeldikop’ toxicity and further established by Bradley et al. (1939a,b) in North America during investigations of poisonous oaten hay and by Williams & Hines (1940), C.T White’s colleagues on the Queensland Poisonous Plants Committee, during investigations of poisoning incidents with cattle consuming Salvia reflexa in this state.

The herbicides available for control of weeds and other plants during C.T.White’s working life were very limited. In many of his publications where advice on weed control is given, the use of arsenic solutions is commonly, if not universally, recommended. In the same period, arsenic, nicotine and copper sulphate were the standard treatments dosed to sheep to control internal parasites. After the Second World War, hormonal herbicides became available and were advocated in their turn, for example 2,4-D against weir vine (Ipomoea sp. Q6 (aff. calobra)) (White 1949).

Since C.T. White: the effort continues

The most powerful drive in the ascent of man is his pleasure in his own skill. He loves to do what he does well and, having done it well, he loves to do it better.

From The Ascent of Man. Jacob Bronowski 1973

The period from 1950 to 1990 appears, from the perspective of the early 21st century, as a golden age of veterinary and chemical investigation of plant poisoning in Australia. The volume and quality of work in that period eclipsed that which underlay it by a large margin. C.T. White was part of the ignition sequence. During the last 20 or so years of C.T.White’s life and under his aegis, Selwyn Everist began and matured the interest in poisonous plants that flowered magnificently in the milestone publication of two editions of Poisonous Plants of Australia (Everist 1974, 1981). Since about 1985-90, the momentum has slowed as many of the people involved have died (Selwyn Everist in1982), retired fully (chemists Clive Culvenor, Mervyn Hegarty, Tom McEwan, Peter Oelrichs; veterinarians Lionel Laws, Bill Hartley, Brian O’Sullivan, Peter Hooper) or partly (chemist John Edgar, veterinarian Alan Seawright) or changed direction due to reduced resources and changing priorities in their employing organisations (Mike Pass, Clive Huxtable, Peter Dorling, myself). These people and others established Australian expertise in this field as second-to-none in the world, linking it with that from other countries in an on-going series of international symposia (six to date) starting in 1977 (see McKenzie 1995).

Since 1950, investigation of intoxications has benefited from the continued application of established scientific methods and the development of new ones

Veterinary pathology techniques produce detailed descriptions of the gross, microscopic and chemical changes produced by toxins in body organs and tissues. This allows poisoning syndromes3 to be effectively differentiated from each other and from other types of disease. It is noteworthy that in several accounts of experimental intoxication, for example, Rudd & White (1933), Legg & White (1939, 1941b), necropsies4 were performed on dead animals, but apparently no specimens were collected for histopathology5. Had they been collected and examined, characteristic lesions6 should have been discovered in some cases. For example, necrosis7 of liver cells in a characteristic pattern should have been seen in Myoporum acuminatum (strychnine bush) poisoning of cattle or sheep.

Analytical techniques for detection of toxins have become very much more sophisticated, for example high performance liquid chromatography (HPLC), mass spectroscopy (MS) and various immunological techniques.

Techniques for determining the chemical structure of complex toxin molecules have been developed. As a result,

Many plants now recognised as toxic were not so recognised in C.T. White’s time

Many plants suspected as toxic in C.T.White’s time have not been confirmed as such by closer investigation

The identity and structure of the toxins responsible for many more plant poisonings are known

The effects of many poisonings on animals are better understood, providing us with more accurate diagnostic appraisal of cases and with the chance to develop rational effective treatments and preventive measures for animals at risk

Available information has greatly expanded. Large increases in information about syndromes and toxins have been published in the scientific literature. Selwyn Everist’s compilation of plant poisoning information for Australia (Everist 1974, 1981) plus numerous other works on poisonous plants in the world from both English-speaking and non-English-speaking sources have been published. My bookshelves carry substantial works on poisonous plants in North America, South America, Britain, Europe, Southern and Eastern Africa and India, none of which were available before 1950.

Access to information has been greatly improved by the digital revolution. Now there is rapid interstate and international consultation of experts on cases through e-mail including the use of digitised images of plants and affected animals and rapid searching of the scientific literature through web-mounted databases (for example PubMed, CABI).

Public demand for knowledge remains high from rural and regional Queensland. C.T.White answered a substantial correspondence including much on the toxic properties of plants. Current Queensland Herbarium staff members continue to answer such questions. I complement this tradition by dealing with over 300 toxicological enquiries annually - many through the DPI Call Centre, another modern innovation linking those with questions with sources of information. As working veterinary pathologists, my DPI Veterinary Laboratory colleagues and I also pursue laboratory investigations with specimens from suspected poisoning cases submitted by field veterinarians.

Community attitudes have shifted, swinging from the outright exploitation of natural resources and the free introduction of exotic plant species to meet perceived needs to the sustainable use of resources, the conservation of biodiversity, the biological control of weeds and the prevention of further importation of potentially weedy species. An increased awareness of animal welfare has resulted in less, but better-controlled and better-directed, animal experimentation

Resources, financial and human, for research on poisonous plants have fluctuated since 1950, trending up until the 1980s, and then down. Queensland Herbarium is now, rightly in my opinion, part of the Environmental Protection Agency and plant-poisoning investigation gets no resources except through the plant identification service. There have been no chemists working on plant toxins in DPI for some 10 years. I am the only QDPI scientist with any close interest in this field. Since Professor Alan Seawright retired in 1993, there has been no veterinary toxicologist employed directly by the University of Queensland Veterinary School. I deliver the toxicology course under a contract between UQ and DPI.

Since C.T.White: the task remains

The woods are lovely, dark, and deep,

But I have promises to keep,

And miles to go before I sleep

From Stopping by Woods on a Snowy Evening. Robert Frost. 1923

Despite the large expansion of knowledge since 1950, long-known toxins such as nitrate-nitrite and cyanogenic glycosides continue to kill domestic animals. However, the scenes of such carnage have shifted with time and engineering technology. Motor transport has reduced (not abolished) the number of animals that die of plant poisoning in droving mobs on stock routes - there was a recent serious mortality from Bryophyllum delagoense (mother-of-millions) on a stock route in New South Wales. The move from hoof to tyres has tended to increase poisoning incidents in and around stockyards where hungry animals are unloaded into contact with such hazards as lush Cenchrus ciliaris (buffel grass) with soluble oxalates (McKenzie et al. 1988), lush Dactyloctenium radulans (button grass) with nitrate (McKenzie et al. 2002) and Portulaca oleracea (pigweed) with both. The manure-rich soil of stockyards boosts the toxin content of these plants to dangerous concentrations in contrast with these taxa in less nutrient-rich environments.

One of the old methods of trying to reduce the number of plant poisoning deaths is still employed - education through publishing illustrated books describing the hazards (compare Bailey & Gordon 1887 with Dowling & McKenzie 1993). It continues to fail to completely penetrate its intended target - the working knowledge base of owners and managers of livestock - and consequently, animals continue to die unnecessarily despite our often quite adequate knowledge of prevention and treatment strategies. This is an area of endeavour yet to be thoroughly addressed and is in need of serious effort. The digital revolution may help, but the old adage “You can lead a horse to water but you can’t make it drink” still applies. Somehow, we have to make the water irresistible - a difficult task in the face of the numerous more pressing demands on those with a need for the knowledge.

In his annual report for 1917-18, C.T. White writes “Visits were paid to Cooloolabin (Blackall Range) and Cooroy (North Coast line) to make an examination of paddocks where stock had died, supposed to have been through eating poisonous weeds. In visiting these “scrub” areas very rarely is it possible to point to any one plant that can definitely be blamed as the cause of the trouble. The best one can do is to point out all plants of definitely known or suspected deleterious properties and recommend their eradication.” (White 1918b). My experience of being consulted on suspected plant poisonings and carrying out paddock inspections is not much better, but knowing what had happened to the animals through veterinary clinical and post mortem examinations, and checking this against our current knowledge of plants, does help to focus investigations almost a century later. Never-the-less, I consider achieving a positive outcome in every second case as the best that can be expected of such attempts. The development of chemical assays to detect significant amounts of known toxins in stomach contents and tissues of poisoned animals is an area also needing serious effort. We can detect nitrate and cyanide, but there are many other toxins for which we have no diagnostic assays at present.

New serious toxicological challenges continue to emerge periodically, requiring effective responses from those charged with protecting human and animal health. Recent examples have been:

Cyanobacteria (blue-green algae) in fresh water supplies are an increasingly-recognised threat to livestock and human health. C.T. White may have heard of the Nodularia spumigena intoxications in Lake Alexandrina in the nineteenth century (Francis 1878), but this group of organisms achieved serious public prominence only in the 1990s with the huge Anabaena circinalis bloom in the Darling River (Jones & Negri 1995). Cylindrospermopsis raciborskii blooms regularly in many water storages throughout Queensland (McGregor & Fabbro 2000), was associated with serious illness in humans on Palm Island (Hawkins et al. 1985) is suspected as the cause of Barcoo spews in humans (Hayman 1992), and has caused several cattle poisoning incidents (McKenzie et al. 2002). The alkaloid toxin it produces damages the liver, kidneys and heart.

The toxigenic fungal pathogen, sorghum ergot (Claviceps africana), was suddenly discovered to be widespread in Queensland grain sorghum crops in 1996, with subsequent serious local impacts on the health of pigs and cattle fed infected grain (Blaney et al. 2000).The alkaloid toxins in the sclerotes (ergots), which the fungus generates as a resting life-cycle stage replacing the seeds of the infected crop, interfere with the function of peripheral blood vessels disrupting the body’s temperature regulating capacity and with the pituitary gland reducing prolactin production and diminishing or stopping milk production.

Poisonous plants in Queensland: then and now

We are like dwarfs on the shoulders of giants, so that we can see more than they, and things at a greater distance, not by virtue of any sharpness of sight on our part, or any physical distinction, but because we are carried high and raised up by their giant size.

From Metalogicon Book III, Chapter iv. John of Salisbury 1159 citing Bernard of Chartres

Relative stability since C.T. White

Toxic plants known to, or suspected by, C.T.White for which we have no significantly new knowledge today include:

Rhodomyrtus macrocarpa (finger cherry). White (1921a) for the first time reported intoxication of cattle after browsing the plant resulting in death and blindness. Eating the fruit (either unripe, ripe or fungus-infected depending on information source) was known to permanently blind humans, particularly children, since the late 19th century, and an apparently-effective education campaign was mounted in northern Queensland schools to prevent it. The toxin responsible for damage to the optic nerve remains unidentified.

Lamium amplexicaule (dead nettle). White (1921a, 1938a) recognised the plant as a cause of staggers in stock. The responsible toxin remains unknown.

Bridelia exaltata (scrub ironbark). Smith & White (1918) & White (1921b) recognised the cyanogenic potential of the plant.

Baccharis halimifolia (tree groundsel). White (1923a, 1936) reported negative feeding trials with cattle and guinea pigs in an attempt to test field suspicions of toxicity.

Sorghum halepense (Johnson grass) & Sorghum verticilliflorum (wild sorghum). White (1937a) reported the cyanogenic glycosides as toxic hazards.

Ricinus communis (castor oil plant). White (1938b) recognised seeds as toxic and that the toxin was a toxalbumin (toxic peptide).

Extended knowledge since C.T. White

Toxic plants known to, or suspected by, C.T.White for which we have extended knowledge today include:

Dactyloctenium radulans (button grass). White (1937b) reported poisoning incidents with hungry sheep given access to button grass in stockyards without establishing the cause of the poisoning. Tests for HCN on the grass were negative. Only recently has the nature of this syndrome been fully explained as due to nitrate-nitrite toxicity (McKenzie et al. 2002). It is not surprising that White did not realise this, as this type of poisoning was not recognised in Australia until late in the 1930s (see above).

Senecio lautus (variable groundsel, native fireweed). White reported cases of liver cirrhosis in cattle in central Queensland in his Annual Report for 1921 (White 1921c). Cases seen in 1992 were confirmed as resulting from pyrrolizidine alkaloids in the plant (Noble et al. 1994). These are persistent toxins causing long-term liver damage and a build-up of scar tissue (= cirrhosis) and occur in a range of plants worldwide, mostly in the genera Senecio, Heliotropium, Echium and Crotalaria.

Myoporum acuminatum (strychnine bush). Legg & White (1941a,b) produced intoxication experimentally, confirming field suspicions of toxicity. Their reports describe only gross necropsy findings and do not recognise the primary lesion in the liver. The toxins are now known to be furanosesquiterpenes. The pattern of liver necrosis that they produce depends on the state of activity of the xenobiotic8 biotransformation enzymes9 in liver cells (Seawright 1989).

Wikstroemia indica (tie bush). Pound & White (1920) produced diarrhoea and dysentery in each of two heifers (weights not given) after feeding about 2 and 4 kg of plant spread over 4 and 6 days respectively. Red deer have been poisoned by this plant, dying of a defect in the clotting mechanism of the blood - a quite different mode of poisoning due to coumarin derivatives in the plant (Dowling 1985). This mode is identical with the effects of the anti-coagulant rodenticides (for example, warfarin and brodifacoum) commonly used domestically.

Trema tomentosa (poison peach, peach leaf poison bush) (Fig. 3). Smith & White (1920b) reviewed the state of published knowledge and opinion on toxicity of T. tomentosa, the consensus of which was against toxicity. They detected faint to fairly strong positive cyanide reactions to picric acid papers on leaves of 4 of 10 specimens from different localities. Today we know that T. tomentosa is toxic, although non-toxic plants are acknowledged, and that it produces rapid death of liver cells leading to liver failure (acute periacinar coagulation necrosis of hepatocytes), a very different effect to that of cyanide. The nature of the toxin has still not been determined. The plant continues to kill cattle, goats and occasionally, horses.

Trema tomentosa (Roxb.) Hara (poison peach)

Figure 3. Trema tomentosa (Roxb.) Hara (poison peach). Parts of branches with unripe fruits. Smith &White (1920b) described its toxicity to cattle, but misattributed this to cyanogenic glycosides. [Fresh specimen directly scanned by R.A.McKenzie]

Ipomoea sp. Q6 (aff. calobra) (weir vine). White (1920a) provided two graziers’ description of the deranged behaviour in sheep, cattle and horses that develops after feeding on the vines for some weeks. The syndrome was recognised as closely resembling that caused by Swainsona spp. (Darling peas). We now know that Swainsona poisoning is due to the alkaloid swainsonine, which also causes locoism in North American livestock grazing on species of Astragalus and Oxytropis. Swainsonine is a glucosidase inhibitor interfering with the handling of the sugar mannose in body cells. Mannose builds up within the cells, particularly those of the brain, interfering with their efficient functioning and causing the deranged behaviour. Weir vine also contains swainsonine and another glucosidase inhibitor, calystegine B2 (Molyneux et al. 1995). Weir vine-poisoned animals have the same microscopic changes in their brains as Darling pea-poisoned ones do - a fine foamy appearance (cytoplasmic vacuolation) of nerve cells (neurones) from the storage of unprocessed sugars.

Melia azedarach var. australasica (white cedar). White (1920b) recognised the toxicity of the drupes (fruit) for pigs. The toxins have now been identified as tetranortriterpenes of the limonoid class and the effects on pigs described in detail (Oelrichs et al. 1983).

Crotalaria pallida (streaked rattlepod). White (1923c) noted cases and feeding trial with goats in the Northern Territory in which sudden death occurred. Laws (1968) produced acute pneumotoxicity in sheep, describing severe fluid out-pourings in the lungs and chest cavities. The plant contains monocrotaline, a known lung-damaging pyrrolizidine alkaloid.

Asclepias curassavica (red cotton). White (1926) noted the toxic potential of the plant, but did not connect this with cardiac glycosides, their known toxic principle.

Acacia georginae (Georgina gidyea). White (1927a,b) reported the toxicity of the plant. Fluoroacetate, the toxin used in the vertebrate pesticide 1080, and found in toxic species of Gastrolobium (mostly in south-western Australia), Dichapetalum (in Africa) and Palicourea (in South America), was found to be the toxic principle by Peter Oelrichs and Tom McEwan (Oelrichs & McEwan 1961, 1962). Genetic manipulation has been used to produce a bacterium capable of destroying fluoroacetate in the rumen of exposed cattle and sheep and thus protecting them from poisoning (Gregg et al. 1998). This organism has not been released for use by industry because of concern that it may establish and persist in pest vertebrates such as goats and rabbits, making them resistant to control by 1080 poisoning campaigns. The recent development of immunocontraception for the control of feral rabbits using viral vectors spread by biting insects may modify this stance.

Lantana camara. White (1929) attempted to sort out confusion produced by researchers who fed lantana to cattle experimentally without having their plant material examined by a botanist. This issue continues to cause difficulties to this day when researchers do not recognise the potential volatility of plant taxon names and fail to secure voucher specimens in herbariums (McKenzie 1993). White (1935c) described the effects of lantana on cattle, giving emphasis to the effects of dehydration, which, while important, are a side issue to the liver dysfunction that underlies the photosensitisation10 in these cases. Lantana poisoning of cattle can now be effectively treated by neutralising the reservoir of toxin in the rumen of affected animals at the same time as taking care of the animal’s state of hydration (Pass 1986, McKenzie 1991).

Cestrum parqui (green cestrum) (Fig. 4). Rudd & White (1933) recognised toxicity, describing haemorrhage into the alimentary tract as a major effect. Today we recognise that the prime site of damage is the liver. The alimentary haemorrhage probably results from increased blood pressure in the vessels flowing from the gut to the liver as a result of swelling in the damaged liver impeding blood flow (McLennan & Kelly 1984, Kudo et al. 1985). We also know that the toxins responsible are diterpenoid (kaurene) glycosides, mitochondrial poisons (Pearce et al. 1992).

Cestrum parqui L’Herit (green cestrum)

Figure 4. Cestrum parqui L’Herit (green cestrum). A flowering branch. Rudd & White (1933) described its toxicity to cattle without detecting the major lesion, liver necrosis. [Fresh specimen directly scanned by R.A.McKenzie]

Xanthium occidentale (noogoora burr). White (1933) alerted farmers to the hazard from recently-germinated burrs carrying cotyledons (seed-leaves). Now we know that these contain diterpenoid (kaurene) glycosides (Cole et al. 1980), the liver-damaging toxins that also occur in green cestrum. We also know that the burrs themselves are toxic if accidentally included in feed grains.

Verbesina encelioides (crownbeard). White (1925) had no evidence for toxicity available, but later reported toxicity and its experimental reproduction (White 1939). The plant is now known to contain galegine (Eichholzer et al. 1982) and to poison ruminants and pigs, causing a massive out-pouring of fluid into the lungs and thoracic cavity.

Pimelea trichostachya (flaxweed, Borgia’s bouquet). Legg & White (1940) demonstrated diarrhoea in sheep fed the plant. It is now known as the main cause of Pimelea poisoning of cattle (St.George disease) which emerged as a major problem in the 1960s in southern inland Queensland, causing chronic diarrhoea, anaemia and chronic heart failure with massive fluid swellings of the brisket and beneath the lower jaw (Kelly & Seawright 1978). Only cattle are affected, in major part because of their unique lung blood vessels that are more sensitive to the toxins (Clark 1973), now recognised as irritant diterpenoids (Freeman et al. 1979). Effective control measures remain beyond our grasp.

Toxicity recognised since C.T. White

Plants known to C.T.White without evidence of toxicity that are now known to be toxic include:

Ageratina adenophora (Crofton weed) and Ageratina riparia (mist flower). White (1934b, 1938c, 1944b) records no poisonous properties of either A. riparia or A. adenophora. Now we recognise that consumption of either plant by horses can produce irreversible lung damage (O’Sullivan et al. 1985). Natural cases caused by A. adenophora are known, but not A. riparia.

Pteridium esculentum (austral bracken). In 1918 (White 1918b), C.T.White says “In Europe & North America the common bracken has been accused of poisoning stock, but the accounts are conflicting. I have never heard of any of the Australian forms causing harm to stock in any way.” By 1935, experimental evidence from Britain and New South Wales confirmed toxicity (White 1935a). Ptaquiloside, a norsesquiterpene glycoside known from bracken (Smith et al.1994) and mulga fern (Cheilanthes sieberi) (Smith et al. 1989) in Australia , causes bracken poisoning in cattle (not other livestock) by suppressing the bone marrow thus causing widespread haemorrhage and reduced resistance to bacterial infection (Clark & Dimmock 1971). It also causes urinary bladder neoplasia in cattle, a condition called bovine enzootic haematuria (McKenzie 1978).

Heliotropium amplexicaule (blue heliotrope). White (1924a) did not recognise this plant as toxic. It contains pyrrolizidine alkaloids and has poisoned cattle in Queensland (Ketterer et al. 1987).

Heliotropium indicum (Indian heliotrope). White (1921d) records the plant only as weedy. It is now recognised as capable of causing pyrrolizidine alkaloidosis (van Weeren et al. 1999).

Cryptostegia grandiflora (rubber vine). White (1923b) had heard of no cases of poisoning, remarking on the low palatability of the plant. Cardiac glycosides in the plant have since killed cattle and horses (McGavin 1969).

Alstonia constricta (bitter bark). White (1924b) did not recognise the plant as toxic, but toxicity was demonstrated soon after (Copeland & Seddon 1931). The plant contains alkaloids with an action similar to that of strychnine (Collins et al. 1990).

Phytolacca dioica (paccalacca, bella sombra). Advocated by White (1924c) as a fodder tree (which it is), it has been recognised that unripe fruits from the female trees can be toxic to cattle and poultry (Storie et al. 1992).

Celtis sinensis (Chinese elm, Chinese hackberry, Portuguese elm). Advocated by White (1924c) as a fodder tree, it has become a serious environmental weed in southern Queensland and its fruits have been suspected as toxic to dogs (R.A. McKenzie, unpublished data 2001).

Gomphrena celosioides (gomphrena weed). White (1934a, 1944a) thought the plant could be quite good fodder and had no toxic properties. It is now known to cause incoordination in horses forced to eat large amounts from run-down pastures (Newton 1952).

Acroptilon repens (Russian or creeping knapweed) & Centaurea solstitialis (St.Barnaby’s thistle). White (1935b, 1946) knew of no toxicity from these plants that are now known to cause the rare brain disease nigropallidal encephalomalacia in horses - a toxicity reported mostly in California, but once in New South Wales (Gard et al. 1973).

Generating new knowledge of poisonous plants and plant poisoning

Some work of noble note may yet be done

From Ulysses. Alfred, Lord Tennyson

How can we know that a plant is poisonous?

There are no features of a plant’s physical form that distinguish the toxic from the non-toxic. Additionally, different animal species vary in their susceptibility to plant toxins, in some cases to the extent that one species may be totally unaffected while a second species dies. For example, horses resist lantana toxicity while cattle succumb.

We label plants as toxic by one or a combination of the suspected taxon

being associated with multiple cases of consistent syndromes under field conditions

yielding positive results from feeding experiments in target animal species

having known toxins isolated or detected in hazardous amounts

Of course, for a plant to be reliably known to the community as toxic requires that evidence in the above categories be published in the scientific and popular literature and be widely read and understood.

C.T.White and his contemporaries favoured the last 2 of the above methods as the most scientifically rigorous of the three. During the last 15 years of the 20th century, the first and last methods have been more favoured, as research scientists acknowledged the community’s growing concern for enhancing the welfare of animals, particularly experimental animals and our knowledge of toxins from plants expands.

The importance of an accumulation of observations

A greater reliance on observation of field cases of toxicity in recent years has been influenced by:

Lack of resources for experimentation

Increased respect for animal welfare

Direct observation of the natural disease in the animals affected in the field involves no inter-species extrapolation difficulties posed by some experimental results. Investigations of poisoning incidents can now be careful and detailed through using veterinary clinical and pathological (necropsy, histopathology) techniques coupled with chemical assays and accurate botanical identification of associated plants. Compared with descriptions of incidents written in the first half of the 20th century, these methods now produce much more detailed definitions of poisoning syndromes. This in turn allows fine comparisons to be made between incidents and a composite picture to be built up from natural poisoning incidents.

Successful recent examples of this method in Queensland include Senecio lautus poisoning of cattle (Noble et al. 1994) and Macadamia integrifolia fruit poisoning of dogs (McKenzie et al. 2001).

The positive and negative aspects of experimentation

All experimentation with animals and poisonous plants must by thoroughly justified in terms of producing as big a net positive benefit as possible for the animal species concerned. Positive benefits to be aimed for include more accurate diagnostic methods for the poisonings and more effective therapies and preventative measures.


Well designed experiments produce rapid and reliable results applicable to field conditions. A successful experiment lasting a week can confirm the toxicity of a plant with a high degree of confidence. To reach the same degree of confidence by field observation alone would require the careful documentation of several natural poisoning incidents. These of course occur sporadically over a much greater length of time and involve more variables (and thus more uncertainty). The controlled conditions of experiments minimise the number of variable factors, thus boosting confidence that the effects observed are actually caused by the toxin or plant used. The doses used are known compared with being estimated (at best) in field cases. Careful design and thorough execution of experiments will maximise the harvest of data from each experimental animal, placing the greatest possible value on each life. For example, close observation will allow the optimum preservation of tissue samples. In field situations, animals may be dead for too long or not immediately available for useful samples to be collected.


It is often difficult to closely match experimental conditions to those under which natural cases of poisoning occur. Thus negative results, where feeding the suspected plant produces no adverse effect, may be inconclusive. Deliberately intoxicating animals compromises animal welfare, but a positive outcome increases the probability that natural cases will be prevented or effectively treated - the suffering of a few may prevent or alleviate suffering for many. Also, choosing endpoints for experiments that minimise or eliminate pain or ensure that any effects can be swiftly reversed, by withdrawing toxin administration, applying therapy or both, may reduce the negative welfare impact of experimentation. Non-natural-target species may need to be chosen as experimental subjects because of resource constraints. For example, sheep (or laboratory rodents) may be used to study a poisoning of cattle because they are cheaper to buy and maintain. This produces difficulties of extrapolation between animal species and reduces confidence in the effective application of the results to natural cases.

Successful recent examples of the use of experimentation to advance animal welfare have included the development of effective therapies for Bryophyllum spp. (McKenzie & Dunster 1987) and Lantana camara toxicities in cattle (Pass 1986, McKenzie 1991).

Living with poisonous plants

Nam et ipsa scientia potestas est (Knowledge itself is power)

From Religious Meditations. Of Heresies. Francis Bacon (1561-1626)

C.T. White would have agreed with the sentiment expressed in the aphorism above. Knowledge powers the prevention of plant poisoning by giving us the chance to avoid the combination of poisonous plants with the circumstances under which intoxications can occur. In this, we have improved our position significantly since C.T. White, but we have not prevented major poisoning incidents from continuing to occur. The efficiency of knowledge transfer from the scientific literature to the minds of those charged with the management of animals is imperfect and needs improvement. Knowledge also provides the power to treat poisoned animals effectively in some cases, but in most plant poisonings, therapy is futile and prevention remains the only option. Again, our position is a significant improvement on that of C.T. White and his contemporaries.

Working in this field has generated in me a profound respect for plants that is seated in a sense of wonder at the enormous kaleidoscopic range of toxins that they contain. The dynamic tension between insect and molluscan herbivores and their food plants is believed to be the stimulus for the evolution of this arsenal of defense chemicals - the eaten evolving more effective chemicals and the eaters in turn evolving with more effective avoidance or detoxication methods. I regard actual poisoning of livestock and humans as quite minor “collateral damage” when we and “our” animals stumble blindly into the midst of skirmishes in the multi-millennium-long struggle between the invertebrate eaters and the eaten.

Much new knowledge has been captured since C.T. White. Many people have contributed to it. While animals continue to die from plant poisoning, the task pursued for so many years by C.T. White and his heirs remains uncompleted. New challenges undoubtedly lie ahead.


BAILEY FM, GORDON PR (1887) Plants Reputed Poisonous and Injurious to Stock.

James C. Beal, Government Printer, William Street, Brisbane.

BAILEY FM (1906) The Weeds and Suspected Poisonous Plants of Queensland with brief

botanical descriptions and accounts of the economic, noxious, or other

properties, to which are added in most instances figures illustrating portions of

the plants brought under notice. H. Pole & Co. Printers, Brisbane.

BAILEY FM (1913) Comprehensive Catalogue of Queensland Plants. 2nd edition. [C.T.

White prepared the 976 line drawings included]

BLAKE ST (1952) Cyril Tenison White 1890-1950. Proceeding of the Royal Society of

Queensland 60:35-48.


MARYAM R (2000) Sorghum ergot (Claviceps africana) associated with

agalactia and feed refusal in pigs and dairy cattle. Australian Veterinary Journal

78: 102-107.

BRADLEY WB, BEATH OA, EPPSON HF (1939a) Oat hay poisoning. Science 89:365.

BRADLEY WB, EPPSON HF, BEATH OA (1939b) Nitrate as the cause of oat hay poisoning.

Journal of the American Veterinary Medical Association 94:541-542.

CLARK I.A. (1973) Research in Veterinary Science 14: 341-.

CLARK IA, DIMMOCK CK (1971) The toxicity of Cheilanthes sieberi to cattle and sheep.

Australian Veterinary Journal 47:149-152.

COLE RJ, STUART BP, LANSDEN JA, COX RH (1980) Isolation and redefinition of the

toxic agent from cocklebur (Xanthium strumarium). Journal of Agricultural and

Food Chemistry 28:1330-1332.


Plants for Medicines. A Chemical and Pharmacological Survey of Plants in the

Australian Region. CSIRO Australia, Melbourne. p.4.

COPELAND HM, SEDDON HR (1931) Poisoning of sheep by quinine bush or bitter bark

(Alstonia constricta). Agricultural Gazette of New South Wales 42:925-926.

DOWLING RM (1985) Tiebush: an old problem for a new industry. Queensland Agricultural

Journal 111(2): no pagination – printed on both sides of the rear cover.

DOWLING RM, McKENZIE RA (1993) Poisonous Plants. A Field Guide. Queensland

Department of Primary Industries, Brisbane.


Galegine and a new dihydroxyalkylacetamide from Verbesina encelioides.

Phytochemistry 21:97-99.

EVERIST SL (1947) Plants poisonous to sheep. Queensland Agricultural Journal 64: 13-

25, 82-95, 139-163.

EVERIST SL (1962) Presidential address: A review of the poisonous plants of Queensland.

Proceeding of the Royal Society of Queensland 74:1-20.

EVERIST SL (1974) Poisonous Plants of Australia. Angus & Robertson, Sydney.

EVERIST SL (1981) Poisonous Plants of Australia. Revised edition. Angus & Robertson,



SEDDON HR, GILRUTH JA, CHEEL E (1934) Some Important Poison Plants

of North Australia. Pamphlet No.49. Council for Scientific and Industrial

Research, Melbourne.

FRANCIS G (1878) Poisonous Australian lake. Nature (London) 18:11-12.

FREEMAN PW, RITCHIE E, TAYLOR WC (1979) The constituents of Australian Pimelea

spp. . The isolation and structure of the toxin of Pimelea simplex and Pimelea

trichostachya Form B responsible for St George disease of cattle. Australian Journal of Chemistry 32:2495-2506.

GARD GP, DE SAREM WG, AHRENS PJ (1973) Nigropallidal encephalomalacia in horses

in New South Wales. Australian Veterinary Journal 49:107-108.


modified ruminal bacteria protect sheep from fluoroacetate poisoning. Applied and

Environmental Microbiology 64:3496-3498.


hepatotoxicity caused by the tropical cyanobacterium (blue-green alga)

Cylindrospermopsis raciborskii (Woloszynska) Seenaya and Subba Raju isolated

from a domestic water supply reservoir. Applied and Environmental Microbiology


HAYMAN J (1992) Beyond the Barcoo – probable human tropical cyanobacterial poisoning in

outback Australia. Medical Journal of Australia 157:794-796.

HURST E (1942) The Poison Plants of New South Wales. New South Wales Poison Plants

Committee, Sydney.

JONES G, NEGRI A (1995) Blooming algal neurotoxins. Chemistry in Australia 62(6):12-


KELLY WR, SEAWRIGHT AA (1978) Pimelea spp. poisoning of cattle. In Keeler RF, van

Kampen KR, James LF (eds.) Effects of Poisonous Plants on Livestock. Academic

Press, New York. pp. 293-300.

KETTERER PJ, GLOVER PE, SMITH LW (1987) Blue heliotrope (Heliotropium

amplexicaule) poisoning in cattle. Australian Veterinary Journal 64:115-116.

KUDO K, KELLY WR, OELRICHS PB (1985) Experimental poisoning of mice and sheep

with Cestrum parqui. In Seawright AA, Hegarty MP, James LF, Keeler RF (eds.)

Plant Toxicology. Queensland Poisonous Plants Committee, Brisbane. pp.540-553.

LAWS L (1968) Toxicity of Crotalaria mucronata for sheep. Australian Veterinary

Journal 44:453-455.

LEGG J, WHITE CT (1939) Hoya australis: a plant poisonous to stock. Australian

Veterinary Journal 15:34-36.

LEGG J, WHITE CT (1940) Broom bush (Pimelea trichostachya): a plant poisonous to

stock. Australian Veterinary Journal 16:175-177.

LEGG J, WHITE CT (1941a) Myoporum acuminatum (strychnine bush). A plant

poisonous to stock. Queensland Agricultural Journal 56:124-125.

LEGG J, WHITE CT (1941b) Myoporum acuminatum: a plant poisonous to stock.

Australian Veterinary Journal 17:104-105.

LONG HC (1917) Plants Poisonous to Live Stock. Cambridge University Press.

LONG HC (1924) Plants Poisonous to Live Stock. 2nd edition, Cambridge University


McEWAN T (1964a) Isolation and identification of the toxic principle of Gastrolobium

grandiflorum. Nature (London) 201:827.

McEWAN T (1964b) Isolation and identification of the toxic principle of Gastrolobium

grandiflorum. Queensland Journal of Agricultural Science 21:1-14.

McGAVIN MD (1969) Rubber-vine (Cryptostegia grandiflora) toxicity for ruminants.

Queensland Journal of Agricultural and Animal Sciences 26:9-19.

McGREGOR GB, FABBRO LD (2000) Dominance of Cylindrospermopsis raciborskii

(Nostocales, Cyanoprokaryota) in Queensland tropical and subtropical reservoirs:

Implications for monitoring and management. Lakes & Reservoirs: Research and

Management 5:195-205.

McKENZIE RA (1978b) Bovine enzootic haematuria in Queensland. Australian Veterinary

Journal 54:61-64

McKENZIE RA (1991) Bentonite as therapy for Lantana camara poisoning of cattle.

Australian Veterinary Journal 68:146-148.

McKENZIE RA (1993) Plant Poisoning? Which Plant?! [Leading Article] Australian

Veterinary Journal 70:201-2.

McKENZIE RA (1995) The Queensland Poisonous Plants Committee: its history and

functions. Australian Veterinary Journal 72:10-17.

McKENZIE RA, DUNSTER PJ (1987) Curing experimental Bryophyllum tubiflorum

poisoning of cattle with activated carbon, electrolyte replacement solution and

antiarrhythmic drugs. Australian Veterinary Journal 64:211-214.


(1988) Acute oxalate poisoning of sheep by buffel grass (Cenchrus ciliaris).

Australian Veterinary Journal 65:26.



BUTTERWORTH EC, BIRTWELL BL (2002) Cylindrospermopsin toxicity in

cattle exposed to blooms of the cyanobacterium Cylindrospermopsis raciborskii:

epidemiology, pathology and toxin detection. Australian Veterinary Journal 80:

in press.


CA, THOMPSON GK, PIDGEON GF, BURREN BR (2002) Nitrate-nitrite

toxicity in cattle and sheep grazing Dactyloctenium radulans (button grass) in

stockyards. Australian Veterinary Journal 80: in press.

McLENNAN MW, KELLY WR (1984) Cestrum parqui (green cestrum) poisoning in cattle.

Australian Veterinary Journal 61:289-291.


of the glucosidase inhibitors swainsonine and calystegine B2 in weir vine

(Ipomoea sp. Q6 {aff. calobra}) and correlation with toxicity. Journal of Natural

Products 58:878-886.

NEWTON LG (1952) Gomphrena celosioides - a plant causing ataxia in horses. Australian

Veterinary Journal 28:151-154.


MORLEY AA (1994) Pyrrolizidine alkaloidosis of cattle associated with Senecio

lautus. Australian Veterinary Journal 71:196-200.

OELRICHS PB, McEWAN T (1961) Isolation of the toxic principle of Acacia georginae.

Nature (London) 190:808-809.

OELRICHS PB, McEWAN T (1962) The toxic principles of Acacia georginae. Queensland

Journal of Agricultural Science 19:1-16.


tetranortriterpenes of the fruit of Melia azedarach. Phytochemistry 22:531-534.

ORCHARD AE (1999) A history of systematic botany in Australia. Flora of Australia 1

(2nd edition):11-103.

O’SULLIVAN, BM, GIBSON JA, McKENZIE RA (1985) Intoxication of horses by

Eupatorium adenophorum and E. riparium in Australia. In Seawright AA, Hegarty

MP, James LF, Keeler RF (eds.) Plant Toxicology. Queensland Poisonous Plants

Committee, Brisbane. pp. 423-426.

PASS MA (1986) Current ideas on the pathophysiology and treatment of lantana poisoning of

ruminants. Australian Veterinary Journal 63:169-171.


SANDERS JKM, WILLIAMS DH (1992) Parquin and carboxyparquin, toxic

kaurene glycosides from the shrub Cestrum parqui. Journal of the Chemical Society,

Perkin Transactions 1:593-600.

POUND CJ, WHITE CT (1920) Results of feeding experiments with a suspected

poisonous plant. Wickstroemia indica C.A.Mey. Queensland Agricultural

Journal 13:172-175.

RIMINGTON C, QUIN JI (1933a) The presence of a lethal factor in certain members of

the plant genus Tribulus. South African Journal of Science 30:472-482.

RIMINGTON C, QUIN JI (1933b) Studies on the photosensitisation of animals in South

Africa. II. The presence of a lethal factor in certain members of the genus

Tribulus. Onderstepoort Journal of Veterinary Science and Animal Industries


RUDD JA, WHITE CT (1933) A plant poisonous to livestock (Cestrum parqui).

Queensland Agricultural Journal 40:143-144.

SEAWRIGHT AA (1989) Animal Health in Australia. Volume 2 (Second Edition).

Chemicals and Plant Poisons. Bureau of Rural Resources, Department of

Primary Industries and Energy, Canberra. pp.77-79.


(1989) Carcinogen in rock fern (Cheilanthes sieberi) from New Zealand and

Australia. Australian Veterinary Journal 66:154-155.


Ptaquiloside in bracken ferns (Pteridium spp.) from eastern Australia and from a

cultivated collection of bracken from world-wide sources. Chapter 9 in Plant-

Associated Toxins: Agricultural, Phytochemical & Ecological Aspects. eds. Colegate

SM, Dorling PR, CAB International, Wallingford UK, pp.45-50.

SMITH F, WHITE CT (1918) An interim census of cyanophoric plants in the Queensland

Flora. Proceedings of the Royal Society of Queensland 30:84-90.

SMITH F, WHITE CT (1920a) On the occurrence of cyanophoric glucosides in the flowers

of some Proteaceae. Proceedings of the Royal Society of Queensland 32:89-91.

SMITH F, WHITE CT (1920b) The peach leaf poison bush, Trema aspera Blume: Its

occasional toxicity. Proceedings of the Royal Society of Queensland 32:132-134.

STEYN DG (1934) The Toxicology of Plants in South Africa together with a consideration

of poisonous foodstuffs and fungi. Central News Agency, Cape Town.

STORIE GJ, McKENZIE RA, FRASER IR (1992) Suspected packalacca (Phytolacca

dioica) poisoning of cattle and chickens. Australian Veterinary Journal 69:21-



POVEDA LJ (1999) Mortality supposedly due to intoxication by pyrrolizidine

alkaloids from Heliotropium indicum in a horse population in Costa Rica: a case

report. Veterinary Quarterly 21:59-62.

WEBB LJ (1948) Guide to the Medicinal and Poisonous Plants of Queensland. Council for

Scientific and Industrial Research Bulletin No. 232, Commonwealth of Australia,


WHITE CT (1917) Report of the Acting Government Botanist. In: Annual Report of the

Department of Agriculture and Stock for the Year 1916-1917, p.104.

WHITE CT (1918a) Report of the Government Botanist. In: Annual Report of the

Department of Agriculture and Stock for the Year 1917-1918, p.74-75.

WHITE CT (1918b) Weeds and poisonous plants of the Atherton Tableland. Queensland

Agricultural Journal 9:147-155.

WHITE CT (1920a) The weir vine (Ipomaea calobra) with notes on its reputed poisonous

properties, based on information supplied by Donald Gunn, Esq. MLA.

Queensland Agricultural Journal 13:269-272.

WHITE CT (1920b) The white cedar (Melia azedarach var. australasica): a plant

poisonous to pigs. Queensland Agricultural Journal 14:146-147.

WHITE CT (1921a) Two plants poisonous to stock. Queensland Agricultural Journal

16:194 & 196.

WHITE CT (1921b) Queensland trees. No.1. Scrub ironbark (Bridelia exaltata).

Queensland Agricultural Journal 15:173-175.

WHITE CT (1921c) Report of the Government Botanist. In: Annual Report of the

Department of Agriculture and Stock for the Year 1920-21, p.78-81.

WHITE CT (1921d) Illustrated notes on the weeds of Queensland. No.22. Indian

heliotrope (Heliotropium indicum L.). Queensland Agricultural Journal 16:103-


WHITE CT (1923a) Weeds of Queensland. No.33. Tree groundsel (Baccharis halimifolia).

Queensland Agricultural Journal 20:49-50.

WHITE CT (1923b) Weeds of Queensland. No.31. The rubber vine (Cryptostegia

grandiflora). Queensland Agricultural Journal 19:286-287.

WHITE CT (1923c) The weeds of Queensland. No.34. A “rattle pod” (Crotalaria striata).

Queensland Agricultural Journal 20:236-237.

WHITE CT (1924a) Weeds of Queensland. No.36. Blue top heliotrope (Heliotropium

anchusaefolium). Queensland Agricultural Journal 21:284-285.

WHITE CT (1924b) Weeds of Queensland. No.35. Bitter bark, native cinchona or quinine

(Alstonia constricta). Queensland Agricultural Journal 21:252-253.

WHITE CT (1924c) Two valuable fodder trees. Queensland Agricultural Journal 22:352-


WHITE CT (1925) Weeds of Queensland. No.40. American dog weed (Verbesina

encelioides). Queensland Agricultural Journal 23:520-521.

WHITE CT (1926) Weeds of Queensland. No.42. Milky cotton bush (Asclepias

curassavica). Queensland Agricultural Journal 26:125-127.

WHITE CT (1927a) Gidyea poisoning in stock. Queensland Agricultural Journal 27:149.

WHITE CT (1927b) Poisoning of stock by Gidyea pods. Queensland Agricultural Journal


WHITE CT (1929) Weeds of Queensland. The correct botanical identity of the lantanas

naturalised in Queensland. Queensland Agricultural Journal 31:294-296.

WHITE CT (1933) The Noogoora burr (Xanthium pungens). A weed poisonous to

livestock. Queensland Agricultural Journal 40:413.

WHITE CT (1934a) Queensland weeds. Gomphrena weed. (Gomphrena decumbens).

Queensland Agricultural Journal 42:83-84.

WHITE CT (1934b) Queensland weeds. Mist flower (Eupatorium riparium). Queensland

Agricultural Journal 41:519-520.

WHITE CT (1935a) The common bracken (Pteridium aquilinum). Queensland

Agricultural Journal 43:150-151.

WHITE CT (1935b) Weeds of Queensland. Creeping knapweed (Centaurea repens).

Melilot or Hexam scent (Melilotus indica). Queensland Agricultural Journal


WHITE CT (1935c) Lantana (Lantana camara) and poison peach (Trema aspera). Their

effects on stock. Queensland Agricultural Journal 43:369-373.

WHITE CT (1936) Groundsel bush or tree groundsel (Baccharis halimifolia). Queensland

Agricultural Journal 45:575-576.

WHITE CT (1937a) Queensland weeds. Johnson grass (Sorghum halepense) and wild

sorghum (Sorghum verticilliflorum). Queensland Agricultural Journal 47:365-


WHITE CT (1937b) Is button grass poisonous to sheep? Queensland Agricultural Journal


WHITE CT (1938a) A cause of shivers or staggers in stock. Queensland Agricultural

Journal 50:533.

WHITE CT (1938b) Two weeds poisonous to stock. Queensland Agricultural Journal


WHITE CT (1938c) Queensland weeds. Shrubby or upright mist flower (Eupatorium

adenophorum). Queensland Agricultural Journal 50:83.

WHITE CT (1939) Wild sunflower – a poisonous plant. Queensland Agricultural Journal


WHITE CT (1944a) Gomphrena weed. Queensland Agricultural Journal 58:179-180.

WHITE CT (1944b) Crofton weed, a serious pest. Queensland Agricultural Journal


WHITE CT (1946) St.Barnaby’s thistle. Queensland Agricultural Journal 62:271-272.

WHITE CT (1949) Weir vine - a declared noxious plant. Queensland Agricultural Journal


WILLIAMS CH, HINES HJG (1940) The toxic properties of Salvia reflexa. Australian

Veterinary Journal 16:14-20.

1 Dr Ross McKenzie B.V.Sc., M.V.Sc., D.V.Sc. holds joint appointments as Principal Veterinary Pathologist with the Department of Primary Industries, Queensland Government, at the Animal Research Institute, Yeerongpilly, and as Senior Lecturer with the University of Queensland School of Veterinary Science, St.Lucia.

2 The term disease is commonly, but mistakenly, taken to mean infection (the invasion of the body by micro-organisms or parasites). However, disease actually means any abnormality of health, and thus includes the effects of genetic mutations, toxins, nutritional deficiencies and metabolic abnormalities as well as infections.

3 A syndrome is a grouping of clinical signs and pathological lesions that constitute an individual disease.

4 A necropsy is a post mortem examination, literally, an inspection of the dead. The alternative single-word term, autopsy, literally means a self inspection.

5 Histopathology is the examination by light microscope of fine slices of preserved tissue stained with dyes and the interpretation of any abnormalities detected as evidence useful in the diagnosis of disease or the elucidation of the disease process in the tissue.

6 A lesion is any abnormality in the structure or appearance of an organ or tissue.

7 Necrosis means cell death.

8 A xenobiotic is a chemical foreign to the body, that is, not produced by normal body processes.

9 Xenobiotic biotransformation is the mammalian body’s defense against toxic chemicals in food. Incoming chemicals are modified, mainly in liver cells, by enzyme-driven chemical changes that improve their water-solubility and thus the body’s capacity to remove them through bile or urine. The process may have positive or negative consequences for the body, with xenobiotics having their toxicity either destroyed (detoxified) or enhanced (potentiated) by these enzyme systems.

10 Photosensitisation is excessive sensitivity of the skin to sunlight, resulting in severe dermatitis of un-pigmented parts of the skin. The syndrome is most commonly caused by liver damage from various causes preventing excretion of the light-sensitive chemical phylloerythrin, or rarely directly by ingestion of fluorescent chemicals from a small number of plants such as Hypericum perforatum (St.John’s wort).