Document jyO6RyxqDVav4o8qr1JZz3N79

FILE NAME: Celanese (CEL) DATE: 1950 DOC#: CEL051 DOCUMENT DESCRIPTION: Laubly Exhibit #4 - Medical Journal L. ___ _________________ _________ THE I PRACTITIONER | Edited by i j SIR HENEAGE O G ILV 1E ( I D .M ., M .C ll., F .JI.C .i. I W ILLIAM A. R. THOMSON. M.D. ROBERT M. STECHER. M.D., r . A . c . r . Vol. 164 January--June 19 5 0 TH E P R A C T I T I O N E R 5 ffifc'NTINCK S T R E E T , L O N D O N . W.i \ ') i o A il lights Rethrvtd OCCUPATIONAL CARCINOGENESIS llv M. W. GO t.m il.ATT, M.D., Pii.n. LeIcntduuresrtriinalIHndyugsietrniealLMabeodriactionrei,esU, nImivperesriitayl oCfhMemainccahleIsntedru;stHrieeas,d Lotfdt.he N or even the most sanguine would claim that the present state of knowledge on the fundamental cause or causes of non-occupational neoplastic disease is so far advanced that reasonable hope may be entertained of a curative or inhibitory chemotherapy or of a satisfactory preventive technique in the near future. It is manifest that a great deal has been learnt in the last thirty years on the reaction of animal tissues and cells to a large number of synthetic chemical compounds and complex mixtures possessing carcinogenic properties. The imposing list of such compounds is rather a cause for de pression than for hope, for the immense variety of structure of these materials seems almost to rule out any unitary hypothesis on how the effects arc brought about. The chemical view of the etiology of cancer in man has naturally received great support from this work. Even the dilemma of the tumours induced in animals by virus does not present an insuperable barrier to the chemical view, for the time must come when the action of a virus will be expressible in terms of chemistry. It is the ancillary but vital character of a virus in the epidemiological field which might colour the picture, hut the ultimate action on the cell must be of a chemical nature. T H E O R I E S OF C AUSAT IO N Whether it is a chemical compound as such or one produced by or in corporated in the structure of a vims, it is necessary for the carcinogenic agent to come into intimate contact with the cell which ultimately under goes the change in character and rate of division which wc call cancer. Various views arc available as to what is the basic nature of such change, for example, that the enzymatic life of the cell is transformed; that the cell undergoes a type of mutation; that the normal processes of control or inhibition of growth arc neutralized by carcinogens; that the cancerous transformation is really an adaptation of normal cells to adverse conditions created by the carcinogen; that the cancer cell arises from a slow breakdown of the activities of normal cells and not from a stimulus to its growth capacity, and no doubt many others. The common character of all such theories is that a point is reached in their exposition when experiment is left aside anti imagination is allowed to enter. The irreversible nature of the carcinogenic process which sometimes sreins remarkable is perhaps not so surprising when the quite undisturbing irreversibility of tissue and organ differentiation from the primordial cell is borne in mind. The chemical nature of the so-called " organizers" which are apparently responsible for specific embryologies! differentiations seems well O C C U ! AT I O N A I. C A H C 1 N O 0 li N KS I S 4S established. There is probably a considerable specificity in the nature of organizers, bringing about definite effects. Such specificity docs not, on the surface at any rate, appear to apply to carcinogens, 'rumours can be induced experimentally in animals which are indistinguishable from tumours arising in man, both "spontaneously" and as a result of occupational absorption of the experimental carcinogen which is quite foreign to the metabolism of man. ex perim en tal evidence The ultimate object in all cancer research is to prevent, inhibit or reverse the carcinogenic process. A great deal of experimental world lias been done to attain this object in the case of experimental cancer. 'Ilte skin cancer induced in mouse skin by the polycyclic hydrocarbons (c.g. 3 ^-benzpyrene or 1,2,5,6-dibehzanthracene) and certain of their derivatives can be in hibited in different degrees by such widely differing substances as mustard gas and some of its derivatives, cantharidinc, carbon dioxide snow, aliphatic aldehydes, and even strong sunlight. Even such unexpected compounds as naphthalene, anthracene and phenanthrene and bromobcnzenc were found to be inhibitory, and the fact that these compounds are excreted in animals in the form of mcrcapturic acid derivatives which arise from the condensa tion of a molecule of the compound and a molecule of the N-acetyl derivative s of cysteine, led to speculation as to the importance in carcinogenesis of the deviation of sulphur from the normal metabolism of the skin cells. Ilut even such anti-carcinogenic action in what must, compared with " spontaneous" cancer, be regarded as a relatively simple situation, is in no sense specific. Thus, certain skin-irritant compounds can act as co-carcinogcns on mouse skin, but in appropriate dosage these can also act as anti-carcinogens, c.g. croton oil and croton resin; and the same applies to anti-carcinogens. In all experimentation of this kind the inducer of the tumour anti how it entered tire body is known, the tissue affected is known, the nature of the tumour is known, and much of the relevant history of the animal is known. In human " spontaneous" tumours the nature of the tumour is, in general, known, the tissue affected is known, a little of the history of the animal is known, but of the inducer and how it entered, became active or was manu factured in the body, nothing is known. I11 human occupational tumours a considerable amount is known about the inducer of the tumour as well as bow it entered the body, in addition to the other elements referred to. It therefore follows that, just as in the experimental animal, the development of tumours can be prevented by not carrying out the experiment, so the occupational neoplasms need not de velop 11 the " experiment" of exposing the worker to industrial carcinogens is notj'dftrricd otit. S P O N T A N E O U S T U M O U R S : P Hi - 1) I S P OS t N G F A C T O R S Nothing is positively known, on the oilier band, about bow to prevent tlic ......itk-i.iis" human tumours. Since turnouts of this kind develop in 406 Tin; PRACTITIONER every class of human society, no absolute case can be made out for any obvious environmental factors, although statistical evidence may show a pre ponderance of certain forms of neoplasm in the poorer strata of society. It is probable that if a cross-section of society could he taken and wrapped in what might be called economic and physical cotton-wool, the incidence of tumours would in course of time he found to be quantitatively lower but qualitatively very similar to that found in norma! conditions, i.e., there would be a cancer age-group, certain forms .of tumours would be pre dominantly male or female in their sex incidence, and so on. Genetic Jacton appear to enter in the appearance of " spontaneous" human tumours and there can be no doubt about them in certain animal strains. Attempts to demonstrate the weight of genetic character on the develop ment of occupational tumours are not easy, in spite of the frequent presence of members of the same family tree in an industry with a carcinogenic hazard. That dietary jactors may enter as accessories in the etiology of human cancer has been considered in great detail but results have been inconclusive. Studies of records of American Insurance Companies led to the conclusion that there was a correlation between overweight and an increased liability to cancer. The same investigator, five years later, showed that both spon taneous anti artificially induced tumours in mice can be delayed or even prevented when a carbohydrate-deficient diet was fed containing all the other recognized food constituents. Of course, such animals are stunted and there follows great endocrine atrophy. In this state the animals become re fractory to the most active carcinogenic compounds. The recommendation by this investigator that diminished food intake should be practised as a prophylactic measure against cancer is perhaps a little naive. When, as in the case of occupational tumours, the cause of the neoplastic process is known, no influence on the liability to develop the tumours which can be attributed to dietary habit has been discovered: such tumours develop in ihe well-nourished and in the relatively undernourished. Age as a factor in the incidence of "spontaneous" tumours in man has for long been held to have a special significance, in spite of the apparent con tradiction between the extreme vitality of a malignant process and the diminished vitality of the host body in which all passion appeared to be spent. Indeed, in some sense a cancer seems to be a process of local re juvenation refusing to be bound by the general senescence. Age as a factor in occupational tumours is of secondary importance. Whilst it is true that occupational tumours often take many years to develop, the determinant as to the age at which they appear depends upon the age at which the tumourinriurmg hazard was first met and how intensely the exposure continued. Cl . I N I C A 1. A N D I* A T II 1) 1 .0 C l C M . F E A T U R E S 'li.w rv rr liillcicut the causes of occupational turnouts may lie from the OCCUPATIONAL CARCINOGENESIS l7 features are more or less identical. This is the more evident in that the occupational tumour may appear many years after all demonstrable vestige of the carcinogenic hazard lias long since disappeared. There can lie no doubt that before the importance of occupational exposures was fully appreciated, tumours were seen in hospital which were in all respects identical and yet belonged in some cases (unknown to the surgeon or pathologist) to the occupational and in others to the non-occupational s groups. This fundamental fact must be appreciated in order to realize fully the non-specific character of carcinogens. T he manner in which the affected tissue reacts to the carcinogenic insult is non-discriminating so far as the latter is concerned. Thus, a bronchial carcinoma is the sa^ne whether it arises from the action of the mysterious "spontaneous" agent or from ex posure to chromates; a carcinoma or papilloma of the bladder arising from exposure to certain aromatic amines is indistinguishable from the bladder tumours arising in everyday life; cancer of the lung is indistinguishable whether it arises from radium emanation or from the unknown spontaneous agents. Hence it will be apparent that no better prognosis can be given for occupational than for non-occupational tumours. An advantage lies with the occupational case inasmuch as, the hazard being known, appropriate protective measures can be taken and earlier diagnosis may be possible. Although the tissue that reacts to potent carcinogenic agents docs not discriminate between one or other of them, so far as the qualitative nature of the reaction is concerned, there must exist some difference as between in dividual tissues and between individual members of a species in the "mechanisms" of response to, or detoxication of, carcinogens. T h e samfc principle applies in occupational carcinogenesis. Thus, absorption o f any of several aromatic amines over a period of years leads to tumours of only one organ, the bladder; great variation between individual members of a working team occurs, inasmuch as some develop tumours in a few years, others after many years, some after having left the industry altogether, and still others not at all. Such individual variation is a commonplace in experimental work with animals bred with every precaution to approach genetic uniformity and used only at the same age, of the same sex, fed with the same diet, o f the same weight, and so on. In spite of these precautions against variability of re sponse to any given agent, toxic, carcinogenic, pharmacological, and the like, responses are obtained which require statistical treatment in the same way as a population of individuals. In other words, if a single individual is taken at random from the group of animals to be used for an experiment, it is noV possible, however carefully uniformity has apparently been assured, to for^lBl with certainty what the response of the individual will he; we can only express the response as a probability. What lies at the root of this peculiar hut universal individual variation nobody can say, hut it clearly lends colour to the life of even a humble mouse or rat. In the case of man it./ .ilier arrani'cmri'tt is unthinkable. If all men reacted in the same way. .|o8 THE I'HACTITIONEH whether psychologically or physically, to external forces, it would require inly one severe epidemic to wipe out the species. It may be permitted to piotc, in this connexion, a great Trench mathematical-physicist, Poincare, vim said: "C'cst Pasymmetrie qui crce le phenomine", and to translate it rccly thus: " Uniformity is death; to have life there must be variation" . A notable respect in which occupational and " spontaneous1' tumours liffcr is that whereas in the latter it is as yet impossible to foretell when or vhere a tumour is likely to arise, in the former the clinical eye can be guided o the region ill which tumours are likely to occur. Human mammary cancer las been shown sufficiently often to have familiar characters to render ational a look-out for its occurrence in sisters and close female relatives in he appropriate family. But it is rarely that a medical attendant inquires into he family history of a member of his panel before disease begins. On the other hand, the medical observer in an industry involving a arcinogenic hazard knows or suspects where tumours will occur and directs iis scrutiny to those parts. Thus, whereas in the manufacture of briquettes patent fuel) attention will be directed to the incidence of skin tumours on lie head and neck due to pitch; in the cotton industry with a hazard due to nincral oil the part to observe is the scrotum. Accepting the general principle that the anticipation of the development if an occupational tumour permits of much earlier diagnosis, it is manifest dial for any real value to attach to it, it must be possible to say that the disease thus diagnosed can be cured pr the condition' improved. S E L E C T IO N OF WORKKHS before proceeding to consider some aspects of occupational carcinogenesis more closely, a word must be said about the selection of workers for in dustries entailing a cancer hazard. Nothing is easier than to compose a pontifical statement on the medical requirements to be fulfilled before a worker is engaged to work in an industry in which sooner or later there is a danger of his developing a tumour in some region or other. But too much dogmatism in laying down criteria of employability is ill-advised. The genetic factor.--A common demand is that there should be no History of cancer in the prospective employee's family. Indeed, in respect of one type of occupational tumour, I have made such a statement. But to what extent is there evidence that the liability to cancer in certain kinds of in dustry is increased by a family history of the disease? The answer must depend to a great extent upon one's own experience and presuppositions. Thus, l have observed the same occupational cancer develop in a father and ,i .on, in two brothers, and in two or three members of the same line, all employed at different times in the same factory. This is a very different tlnng from saying that there is a family tendency unless a great deal more evidence is obtained, for each of these observations may have been due to .................. . '>r C"uilv tmilrl.no unless the 0 C Cl) r A T l O N A 1. C A H C I N O <5 F. N E S I S >9 improvement of conditions of work is so great as to have reduced tin. nazard to vanishing point. The age factor.--Another criterion often insisted upon is that the prospec tive employee should be a young, healthy adult. A formula of this kind is liable to he given as almost self-evident. Assuming that a considerable car cinogenic hazard still exists in a factory, the presumption in the formula is that a young, healthy man is less likely to develop an occupational tumour than a middle-aged or not completely fit person. We know of no evidence for such a presumption. The determinants of the development of occupational tumours arc length and intensity of exposure to the carcinogenic agent, and the individual power to detoxicate or inactivate the carcinogen. Consider the experience in this country in the case of pitch and tar epithelioma and in shale and mineral oil epithelioma of the skin. Of the former actual cases some 6o per cent, occur after 20 or less years of exposure, whereas of the latter only some 3 per cent, occur after 20 years or less of exposure, and some 64 per cent, occur after 55 years or less of exposure. Thus if a young man, of say twenty years of age, enters operations involving contact with pitch and tar or tar products and if he develops a tumour, he has a 60 per cent, chance of doing so before lie is forty. If, on the other hand, he undertakes work involving contact with shale or mineral oil, and if he develops a tumour, the corresponding chance is only 3 per cent. Thus, it docs not seem unreasonable to say that it is less fair to put a young man on to pitch and tar work than on to shale and mineral oil work. This docs not mean that we regard middle-aged men as something to be sacrificed. Far from it. It means that a middle-aged man may run most of his life's course before he gets a tumour at all. Latency of the occupational tumour in ordinary industrial conditions should always be fully understood before any statement is made on the desirable age for new starters in the hazardous operations. The health factor.-- It will he clear that " health" in a prospective employee must be interpreted in the light of the hazard into which he is about to enter. In the first place, if the carcinogenic hazard is one affecting the skin, it is apparent that skin diseases (e.g. warts) or sensitivities should preclude from entry; if the hazard is one involving a carcinogenic dust, it is clear that any condition which induces any measure of dyspnoea must he precluded, e.g. asthma, bronchitis, obesity, cardiac conditions, even when these arc not stillicienlly troublesome to prejudice ordinary working capacity; if the hazard involves a carcinogenic effect on the urinary tract, it is obvious that . any degree of urinary dysfunction or infection must he precluded; if wc may include cleanliness as an adjunct of health, then a dirty worker should never baipcrmitted to continue working in a dangerous toxic process; finally, since the health of a man may have much to tlo with his habits, it is essential to prohibit nail-hiicrs, tobacco-chcwcrs, snufT-takcrs, surreptitious s suckers of sweets ami partakers of snacks. Intelligence in the toother, - ft is best to employ intelligent workers on 410 TIM- P R A C T I T I O N E R processes which involve the subtle 3nd furtive hazard of an industrial carcinogen. The overt manifestations of acutely toxic effects will more or less readily impress themselves on the mind of the least endowed workman, especially if there is danger of what is called " gassing" or if there is evidence of irritant fume or dust. When dealing with liquids or non-irritant dusts or solids, it is much more difficult to impress the average workman. Now, a carcinogenic hazard must involve an employer in the moral obligation to inform his workers minutely on what is involved. The reaction of a worker to such information depends upon his intelligence, for, provided the em ployer goes to the limit in precaution and the informed workman exerts all iiis faculties in careful operation and good housekeeping, there is good and reasonable ground for confidence in his safety. But it will not be overlooked that we have used the phrase, " provided the employer goes to the limit" . Mere adherence to statutory requirements in the Factories Act, Regulations anil Orders is not enough; these are minimum, not optimum. carcinogenic agents 1 proceed now to consider occupational carcinogenesis more especially in relation to the industries or occupations in which it is observed. If we ask the question, what are the occupational carcinogens, the answer must satisfy criteria that the substances indicated are carcinogens experi mentally and that they produce cancer in man in industry. To demonstrate unequivocally that a substance is a carcinogen we must have recourse to animal experiment. Thus, although the relation between coal tar and occupational skin tumours was discovered as long ago as 1875, there was a hiatus of forty years before Japanese investigators produced definite malignant epitheliomas by applying coal tar to the ears of rabbits. Since those days a host of compounds has been tested on animals, and unless a material suspected of producing occupational tumours yields positive re sults in animals of one species or another we cannot say more than that it is suspect. This is the position at the present time with chromates; evidence in favour of their producing occupational cancer of the lung is strong, but they have never been shown to do so experimentally. It must he emphasized, however, that we are far from being able to say that because a particular substance is a carcinogen to animals it is also a carcinogen to man, and it would be dangerous to conclude that because it has not been shown to he a carcinogen in animals that it is therefore not a catcinogcn to man. The demonstration of carcinogenic properties in animals engenders a reasonable fear that the material involved may also be carcino genic 10 man. Hut (lie demonstration of carcinogenicity in man may never be possible. Thus, the notorious dye " butter yellow" , p-dimelhyinininonzobcnzcnc, is indubitably a liver carcinogen to a variety of species and has long been iorhidden as a food colour, but no evidence that it lias ever had carcino genic effects in man lias hern, or ever will lie, obtained. Many azo com- line lu-rii clinivn in nnssrss carciiu>genie properties in animals, tint O C C U P A T I O N A L C A R C ! N O Cli N KS ! S ,|U no reasonable evidence has ever been obtained that men in a dye factory handling only uncontaminated azo compounds, including azo dyestuffs, are especially liable to develop tumours. On the other hand, it has never been shown that the dye intermediate, benzidine, is a carcinogen to animals, but there is no reasonable doubt that it is a carcinogen to man. It has recently been pointed out by Salter (1948) that ``despite a wealth of data indicating carcinogenesis in man from crude mixtures, only three unadulterated agents have thus far been proved carcinogenic for man. These arc (a) radiation, (b) beta-naphthylaminc and (c) arsenic". RADIATION The evidence for skin carcinoma in white workers much exposed to sun light in tropical or semi-tropical lands is widely accepted. The effect of intense ultra-violet light on the much thinner skin of rodents is to produce sarcoma dr mixed tumours, the difference probably being that whereas the relatively thick horny layer of man absorbs the ultra-violet radiation, in the rodent the latter penetrates to the subepithclia! layers and acts on the connective tissue there located. X-rays.-- Statistics of the incidence of cancer and related conditions (e.g. leukaemia) among radiologists, X-ray tube makers and technicians show a definite occupational hazard for these workers. The direct demonstration of the development of tumours of the skin from exposure to X-rays has been made during the treatment of certain skin conditions with this agent, e.g. hypertrichosis of the face, psoriasis, lupus. The final confirmation of the carcinogenic action of X-rays was made by exposing the skin of rodents for many months to this action; spindle-cell sarcomas developed. Radioactive substances.--As long ago as 1879, it was shown that the rapidly wasting disease with manifest pulmonary damage which had for centuries been observed in miners in the Erz Mountains (between Saxony and Ilobemia), was due to a circumscribed lung tumour. In subsequent years there were various theories of the ultimate cause of the condition (e.g. uranium, cobalt, arsenic), but the measurement of the radioactivity in these mines in the early 1920's left no doubt that the miners were inhaling very large amounts of radioactive material, whether as actual radium compounds or as radon gas. An interesting feature of these tumours was, ami still is, that the lung tumours may not appear until many years after the miners have left the mines altogether. This latency is so frequent an accompaniment of occupational tumours as to be almost a specific character of them. The alarming recognition years ago, both in this country and in the United States, that persons engaged in the application of radioactive coinponjUfjs to surfaces anil in the introduction of such compounds into glass tubing (luminizing operations) were liable to develop severe antrinia of _ aplastic type and sarcoma of bone, led to the important I.uminizing Order of 1942. This order is meticulously designed to prevent the ingestion am! 4*4 Till! IMIACTITIONKR dost for forty-three years, and to one of skin cancer complicated by lung cancer in a worker in arsenical insecticides. The clinical evidence in favour of the carcinogenic action of arsenic is undeniably strong, but the experimental evidence with animals is sur prisingly slight. I.ritcb nd Kennaway'a frequently quoted experiments were on painting mouseskin 3 times daily with 1.8 per cent, alcoholic solution of potassium arsenite. This highly toxic solution had to be replaced by a o.ix per cent, solution. After 8b days, the experiment yielded one animal out of too with a aquamous-celled epithelioma and bench's own later experiments yielded no tumours at ail. There were many fatalities. Hyperkeratosis and hyperpigmentation produced by arsenic have been demonstrated by other investigators. Reviewing the evidence, it seems reasonable to agree with Salter (1948) that radiation, p-naphthylamine and arsenic can be identified as carcino genic agents in man, and that there is varying weight of evidence that they are also carcinogenic in animals. OTHER OCCUPATIONAL CARCINOGENS Wc must now consider whether or not in other cases of known or alleged occupational carcinogens an identifiable cause can be held to be responsible. When in 1915, Yamagawa and Ichikawa produced skin carcinoma on the external ear of rabbits by repeated applications of coal tar, and Passey in 1922 induced malignant growth,on the Bkin of mice with ether extracts of soot, the stage seemed to be set to identify the active compounds in these complex mixtures. From this period began the remarkable work of Kennaway and his school: the demonstration by ICennaway of the carcino genic properties of tars obtained by pyrolysis from mineral oil, coal, skin, yeast, hair, and by subjecting relatively simple saturated and unsaturated hydrocarbons to beat in the presence of hydrogen; the study of the charac teristic fluorescence spectra of carcinogenic tars and oils and the later recognition that the polycyclic hydrocarbon 1 :2-benzanthracene possessed a fluorescence spectrum of a similar character; the syntheses of homologucs of benzanthracene leading later to a great increase of synthetic carcinogens; the isolation from coal tar pitch of a pure hydrocarbon which showed the characteristic spectrum and was a potent carcinogen, and its later identifica tion as 3 ^-benzpyrene; ami the relatively recent recovery of about 75 mg. of pure benzpyrene from to g. of crude tar distillate. It may safely be supposed that the actual benzpyrene content of tar is very much more than this having regard to inevitable losses in purification. The chain of evidence sceins complete that 3:4-benzpyrene is the con stituent in tar responsible for human cancer but there is no direct evidence that this is so. No doubt evidence will appear that the omnipresent town-soot contains similar or identical carcinogenic hydrocarbons, hut its reference to the incidence of, say, cancer of the lung (which is certainly on the increase) will he hypothetical until the relation between animal carcinogens and O C C U P A T I O N A L C A H C 1N O C. I! N E S I S 4 5 Aibeslos.-- As long ago as 1938 the suspicion arose that asbestos workers might be more than normally prone to lung cancer. Nordmann (1938) analysed aix coses of lung cancer and allowed that the range of exposure periods was 7 to zi years, and the range of intervals between entering the industry and death waa 15 to 21 years. The malignant disease in some cases occurred years after leaving the industry. Half of these cases were comparatively young, 3s to 41 years of age at death. In one remarkBldc case a seventy-one year old woman had worked in asbestos for only 19 months. Later in the some year lie referred (igjH) to a further seven cases of associated asbestosia and lung cancer, including (.Moyne's finding of six cases of carcinoma of the lung in 50 autopsied cases nf asbestosia. In the 1947 Annual Report of the Chief Inspector of jFactorics, the Senior Medical Inspector tabulated the age incidence anwlpg 235 deaths caused by asbeatosis: in 13.2 per cent, of these cancer of the lung was present, and it is especially important to note that 4.8 per cent, of the agegroup twenty-five to thirty-four, and 5.6 per cent, of the age-grotip thirty- five to forty-four had this condition. It will be seen that the over-all figure is closely in agreement with Gloyne's findings. The establishment of the well-known difluse pulmonary fibrosis induced by exposure to asbestos fibres is usually regarded as due to mechanical action, and some evidence of this is presented by finely grinding the as bestos to particles less than 2p in length, whereby the fibrosis efiect is eliminated in animal experiment. The reverse would be the case with silica. But when so much has been said, it still remains a puzzle why cancer should favour the asbestosis lung and not the silicotic lung. Silica is held to act by local solution and chemical action: asbestos (a hydrated magnesium silicate) is held to act as an irritant in virtue of its physical form. The next step, to bring cancer in the asbestosis lung into line with a chemical theory of cancer, has not been taken. Nickel carbonyl.-- It has been known for a long time that workers in nickel refineries develop remarkable tumours of the ethmoid region of the nose and bronchogenic carcinoma of the lung. Like so many other occupa tional cancers it may occur many years after exposure to the carcinogen has ceased. Amor1! (1939) interesting analysis of ten cases of cancer of the ethinoids anil four of lung cancer ahowa that tile average period from first exposure to tlie hazard (tmle injra) until death was some 17 years (range rr-z8) for cancer of the etlunuids and zo years (range 9-27) for cancer of the lung. In tins scries the nasal tumours attacked the middle turbinalea and the ethmoidal air cells. These were all carcinomas of various kinds, 9 <|uamous, columnar-cellcd and undifferentiated. I introduce nickel into this short discussion mainly because its associa tion'with the problems of occupational cancer is a very good example of a cryptic ktause. In tls<|eparation of nickel from its ores the problem before the technologist is to fractionate theorem such a way that not only nickel but several other valuable metals always found in association with nickel will also he recovered, It is the m.inrirr in which this fractionation is carried out which determines whether or not the jnnress will include the hazard of cancer. Nickel is usually extracted in this country from Ciitiadinn ore and die hitter must undergo a considerable number of proersses tirfoie