might have dropped into the sample while it was being dried, there being always a certain amount of copper and brass dust in most laboratories. For the work a room was therefore set aside in which copper utensils, brass bunsens, and all apparatus containing copper were rigidly excluded, and any brass fittings were coated with varnish. Further specimens were obtained and analysed, great care being taken to guard against adventitious entry of copper into the assay. The results of analysis of these specimens are given in Tables 3 and 4. They show that there is always an excess of copper (compared with the amounts found by the authorities quoted below) in the livers of sheep suffering from bush sickness, although when these livers are extremely fatty the copper is diminished; but if calculated on the dried fat-free liver the percentage is always excessive. Analyses of the livers of healthy sheep killed for consumption have shown, however, that a comparatively large amount of copper in the liver is quite consistent with health. Further, on dosing sheep and cattle with copper-acetate for over a year their livers were found to take up much larger amounts of copper than any liver hitherto examined without producing any of the final symptoms of bush sickness. Feeding experiments conducted by Mr. H. A. Reid [ unclear: ] , F.R.C.V.S., at Wallaceville Laboratory, in which copper-acetate in small doses was given with the food to sheep, showed that the liver could absorb large amounts of copper and remain healthy. The greater part of the copper was no doubt eliminated in the faeces. Some of the sheep died of a braxy-like disease, but others remained healthy, and the experiments were discontinued after they had been going on for seventeen months. Experiments with calves and rabbits extending over a similar period yielded negative results, which supports the experiments of du Moulin (Journ. Pharm., 5, 13, p. 189; abst. in J.C.S., 1883, p. 483), who gave doses of from ½ gram to 1 gram of copper-subacetate every day for six weeks to dogs and rabbits without producing poisonous effects. The bulk of the evidence at present is against the hypothesis that copper is a causative agent in producing bush sickness, but the results obtained are so full of interest that it is deemed advisable to publish them at this stage.

Copper is certainly a normal liver constituent in sheep and cattle. Wynter Blyth (“Poisons,” 1895, p. 613) states that a sheep's liver contains 1 part of copper in 20,000 (0°005 per cent.), and quotes Dupré's statement that in the kidneys and livers of ruminants copper may always be found. Professor Malcolm informs me that the liver of the ox normally contains 0°00225 to 0°0051 per cent. copper. Professor Gilruth refers to articles by Lehman (Arch. f. Hygiene) in which the author gives 0°0048 as the percentage of copper in dried ox-liver. In sheep normally he found 0°0018 per cent. in the dried liver, but in copper districts only half that quantity in the liver but five times that quantity in the heart. Analysis of hearts from bush-disease areas do not show any such excess of copper.

Ellenberger and Hofmeister (Beid. Centr., 1883, pp. 606–9; abst. in J.C.S., 1884, p. 474) experimented with sheep, giving doses of ½ gram to 3 grams. Among the negative results they obtained were no alteration of the muscular structure, no acceleration of the motion of the heart, no uniform alteration in the microscopic appearance of the blood-corpuscles, no alteration of the respiration nor of the secretion of the urine. Amongst the positive results were the presence of albumen, blood, and bile in the urine, flaccidity of the muscles, weakness, and loss of appetite. They note that the excretion of copper from the system is chiefly by the

bile, partly by the urine but in lesser degree; that the liver retains the copper with great tenacity and the pancreas with almost equal strength, and that the kidneys do not retain it as much as the other two organs. The nervous and muscular systems do not contain enough to interfere with their action. They recommend numerous small doses in order to obtain chronic effects. My experiments show that the pancreas does not contain much copper. Max Klemptner (Chem. Centr., 1894, ii, 620; abst. in J.C.S., 1895, p. 321) records the following symptoms in poisoning by sodium cupric tartrate: Atrophy both in acute and chronic cases; diarrhoea some time before death, and persistent vomiting, in chronic cases; enfeebled pulse and laboured respiration. When subcutaneously injected into the pleura the salt produces weakness and stiffness of the hinder extremities; the same is noticed when the salt or copper haemoglobin is exhibited. In one case blood and albumen were found in the urine. Copper was not found in the blood serum, but in the corpuscles.

The question whether chronic copper poisoning can occur in animals is evidently doubted by many authorities. A. Koldeway (Chem. Centr., 1896, ii, 1041; abst. in J.C.S., 1898, p. 39) states that no noteworthy evil results follow small doses of copper (or even large doses) in people in good health, or in animals that vomit readily; long continuance in the use of copper, however, produces slight degenerative changes in the liver and kidneys, which can only be detected on microscopic examination. He doubts the existence of chronic copper poisoning amongst workers in that metal, any illness occurring probably being due to other metallic impurities of the copper. L. Lewin (Deusch Med. Wochenschr., 1900, 26, 689; abst. in J.S.C.I., 19, 1900, 1183) could not discover any phenomena indicating chronic poisoning in copper-workers. The editor of Taylor's “Medical Jurisprudence” (1905, p. 476) is evidently doubtful as to whether chronic copper poisoning can occur in human subjects. This doubt should be even greater as to whether the poisoning can occur in herbivora.

Dieulafait (Compt. Rend., 89, 453: abst. in J.C.S., 1879, p. 1020; 1880, p. 489) states that copper occurs in all plants that live on primary rocks, or on soils derived from those rocks. One hundred grams of rock will always give a reaction for copper. Dupré (Analyst, ii, 1; abst. in J.C.S., 1877, p. 511) states that copper is found in all vegetables as well as in animals, but rarely amounts to more than 1 gram per 100,000 (0°001 per cent.). Dieulafait states that 1 gram of ash from all plants growing on primordial rocks give the copper-reaction, but plants grown on pure limestone contain but traces of copper, requiring at least 100 grams of ash for its detection. J. B. Harrison, in a recent report (1906) to the Science and Agricultural Department of British Guiana, shows its occurrence in hundredths of 1 per cent. in many igneous rocks of that colony, and Hille-brand (“Analysis of Silicate and Carbonate Rocks,” Washington) considers that it can be found almost invariably if looked for in the rock-analysis carried out by the United States Survey Laboratory, but it is seldom reported unless extra precautions have been taken to prevent its entry into the analysis.

The few determinations which have been made of copper in the soil and grass-ash of the affected country go to show that it exists in amounts of the order of thousandths of 1 per cent. rather than of hundredths.

I have not been able to consult any of the original papers, owing to the difficulty in obtaining scientific literature in New Zealand. One is unable to conjecture how many specimens were analysed to supply the above data, and whether the figures might be taken as a standard.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Note—The specimens 0 185 to 0 1906 are from animals from the experimental plots. They died from “bush sickness,” having been on the plots since January, 1911, at Te Pu, and February, 1911, at Mamakau, respectively. The fat is calculated on the wet sample.