An experimental investigation concerning the effects of “Tuba” (Derris elliptica
J. Argyll Campbell published this experimental pharmacological investigation in 1916 in the Journal of the Straits Branch of the Royal Asiatic Society, presenting the first systematic laboratory study of the toxic actions of Derris elliptica (Malay “tuba”) fish-poison on living animal tissues. Working in the Straits Settlements, Campbell sought to move beyond the descriptive ethnographic accounts of tuba’s effects and establish its precise physiological mechanism of action across a range of vertebrate species.
Summary
Campbell’s central problem was the absence of any experimental research on how tuba poison actually kills, despite extensive ethnographic and chemical literature describing its effects. Previous investigators—Greshoff, Wray, Ridley, and Gimlette—had isolated resinous compounds (“derrid,” “tubaine”) and recorded clinical observations, but no one had performed controlled animal experiments to determine the site and mechanism of toxicity. Campbell prepared the poison in the traditional Malay manner (pounding root, soaking in water or Ringer’s saline, filtering, and boiling) and tested it on fish, tadpoles, mosquito larvae, toads, and two species of macaque monkey, both in living animals and on isolated tissues.
The core finding is that tuba poison kills by paralysing the respiratory centre in the medulla oblongata, producing death by asphyxia. Campbell demonstrated this through a series of elegant experiments: the poison dilates blood vessels and weakens both voluntary and involuntary muscle, but has no direct action on heart muscle or on the blood cells themselves. Crucially, he showed that the respiratory effect persists after vagotomy, proving the action is central (medullary) rather than peripheral (vagal). The poison also abolishes the constrictive action of adrenalin on blood vessels, explaining the marked fall in blood pressure observed in anaesthetised monkeys. Campbell further established a clear inverse relationship between phylogenetic complexity and toxicity: the poison is most lethal to higher vertebrates (monkeys) and least lethal to lower forms (mosquito larvae), consistent with a mechanism targeting the medullary respiratory centre.
The article also addresses practical and forensic questions. Campbell notes that fish killed by tuba can be eaten safely by humans because the lethal concentration in the water is far below what would be toxic to a human consumer. He discusses the forensic implications for suspected poisoning cases in Singapore, noting that post-mortem examination reveals only venous congestion and an acrid odour in the lungs, with no specific chemical test available at the time. He proposes a simple biological assay—testing stomach contents on small fish—as a practical forensic method.
Key Findings
- Minimal lethal dose for monkeys (Macacus cynomolgus and M. nemestpinus): extract from 2 gm (30 grains) of root, administered subcutaneously or by stomach tube, producing death in approximately 40 minutes (pp. 124–125).
- Fish toxicity thresholds: solutions as weak as 1 in 100,000 are fatal to Ophiocephalus gachua (~50 gm); a 1 in 4,500 solution kills in 11 minutes, 1 in 12,000 in 28 minutes, and 1 in 50,000 in 83 minutes (pp. 122–123).
- Mosquito larvæ require far stronger solutions: 1 in 10,000 takes at least two days to kill Stegomyia larvae, whereas 1 in 10 kills in one hour—demonstrating the inverse relationship between phylogenetic level and susceptibility (p. 123).
- No direct cardiac or haematological action: the isolated heart beats strongly in Ringer’s solution containing poison at 1 in 8 concentration; no haemolysis occurs and the oxygen-carrying capacity of blood is unaltered (p. 125).
- Vascular effects: the poison dilates blood vessels and markedly lessens or abolishes the constrictive power of adrenalin on monkey limb vessels, explaining the fall in blood pressure (p. 125).
- Colloidal nature of the extract: only one-seventieth of the poison passes through parchment in five days, indicating colloidal rather than true solution; the extract is not antiseptic and is faintly acid (p. 122).
Conclusion
Campbell definitively establishes that Derris elliptica poison kills by central respiratory paralysis in the medulla oblongata, compounded by vasodilation and muscular weakness, rather than by any direct cardiac, haematological, or peripheral neural mechanism. This finding unifies the diverse clinical observations from ethnographic sources—stupefaction in fish, vomiting and jaw fixation in humans, fatal haemorrhage in abortive use—under a single physiological explanation: progressive asphyxia with secondary vascular collapse.
Context
- The study was conducted with institutional support from the Raffles Museum and the Botanical Gardens in Singapore, reflecting the Straits Settlements’ role as a centre for tropical pharmacological research in the early twentieth century.
- The article represents a significant contribution to the pharmacological literature on tropical poisons, bridging ethnographic description and experimental physiology, and it remains a primary source for the history of toxicology in Southeast Asia.