Document peO86z8Ran2r8RvQLq252v6dX

/. WATSON" 'BMr, H. s.. Am.sOil Popula. i, I ^P^l Orypt. Oil r, v, A. K. Si Ari-uiTO.v. D. I{, to Duration of Mitosis in the i of thi Hut; a <`ytokini-fio it*. Oil TUnur Kinft., 5, I, i.v, A. .1., Mobt-EV, A. K.. tttKS. J. (Itf73( Oil Kinotio' icoro of tho. Human Smafl I I f ( / t 1 f t ! i \ i Hr. /. Oawpx (1978) 37, 073 MASS AND NUMBER OF FIBRES IN THE PATHOGENESIS OF ASBESTOS-RELATED LUNG DISEASE IN RATS M. C. DAVIS. S. T. BECKETT, K. E. BOLTON. P. C'OLLINOS and A. P. .MIDDLETON Front the fnMilute of Orru/xdional Mc/Rrine.. Roximryh Flare. Rdinburyh EHtt 9.S7,' Hrcoiv.'il 3 January IW78 Accepted 13 Pebnmry l`J78 Summary.--Five groups of rats were treated by inhalation for 12 months with the U ICC. preparations of the 3 main commercially used asbestos types, chrysotile, crocidolite and amosite. The experiment was designed so that the effects of both fibre mass and fibre number could be examined. The results indicated that chrysotile dust caused far more lung fibrosis than either amphiboie type even when the fibre num bers in the dust clouds were similar. All malignant pulmonary neoplasms found during this study occurred in animals treated with chrysotile. The fibre-number calculations used for the generation of dust clouds were evaluated using the para meters recommended by the Health and Safety Executive in 1976, by which all fibres over 5 itm long are counted using a phase-contrast light microscope. When fibrelength distributions were calculated using a scanning electron microscope, however, it was found that the chrysotile clouds used in this study contained many more fibres over 20 p.m long than either of the amphiboie clouds. The results, therefore, support, previous suggestions that long asbestos fibres are more dangerous than short. They also indicate that neither a single mass standard, nor the present fibre-number standards are satisfactory. The inhalation of asbestos dust may cause both lung fibrosis and neoplasia in those involved, in the industrial processing of this material and for this reason the maximum level of dust in asbestos factor ies is governed by strict standards in most countries. Although the major types of asbestos used commercially differ both physically and chemically, the legislation in many countries lays down one standard which is applied to several or all forms (Zielhuis, 1977). For coal dust, the work of Jacobsen el al. (1970) has shown that the mass of respirable airborne dust corre sponds more closely with radiological change than does particle number. Un fortunately similar data do not exist, for asbestos. This fact was noted by the T.A.R.C. Advisory Committee on Asbestos and Cancer in 1973. Previous work carried oi.r in Edinburgh (Beckett, 1975) has shown that for each type of asbestos there i* a different relationship between the airborne fibre number and mass concen trations. This means that if a gravimetric standard is adopted the permitted fibre number for chrysotile is much higher than for amosite, while if a fibre-number stand ard is operated the permissible mass for amosite is greater than that for chrysotile. This situation has no doubt arisen because reliable evidence relating to the relative pathogenicity of asbestos dust has been difficult to obtain from human epidemio logical studies, since most factories have, in the past at least, used more than one asbestos type. In addition, the information from animal inhalation studies has often been conflicting. Holt, Mills and Young (1965) found no differences in the fibrogenic potential of chrysotile, crocidolite, amosite and anthophyllite. while Wagner (1963) and Wagner and Skidmore (1965) and Morris el al. (1967) suggested that chrysotile produced less fibrosis than amosite or crocidolite for the same mass 674 J. DAVIS, S. BECKETT, R. BOLTON. P. COLLINGS AND A. MIDDLETON dose. In a later inhalation study, using the U.I.C.C. standard reference samples, Wagner et al. (1974) reported that amosite dust invariably gave the least fibrosis and Canadian ehrysotile the most. Croddolite and Rhodesian chrvsotile were inter mediate. Knowledge of the relative importance of the different asbestos tyes in the produc tion of neoplasia is no more precise and, although erocidolite has been specially linked with the production of mesothelio mas (Wagner, Sleggs and Marchand, 1960), subsequent epidemiological studies have indicated that at least some of the other asbestos types may also cause this type of tumour (McDonald, 1973; Selikoff, Ham mond and Seidman, 1973). Similarly, since the report of Doll (1966) showing a greatly increased risk of bronchial carcinomaamong asbestos workers, no reliable human epidemiological data have been produced indicating whether or not all industrially used asbestos types are equally potent in the production of these lung tumours. This is due to the fact that most workers in factories handling asbestos have been exposed to more than one type during their working lives.. Most early animal inhalation studies produced no lung tumours, and those later ones which did result in the production of bronchial carcinomas and mesotheliomas gave positive results with different asbes tos types in each experiment (Gross and De Treville, 1967; Reeves et al., 1971). However, both Wagner et al. (1974) and Reeves, Puro and Smith (1974) published the results of studies in which all the major asbestos types had been administered to rats. Wagner used amosite, anthophyllite, erocidolite and 2 varieties of ehrysotile. He found the highest number of malignant tumours in animals treated with Rhodesian ehrysotile, and the lowest number in those treated with amosite. Anthophyllite, eroci dolite and Canadian ehrysotile gave about the same number of tumours. Reeves used ehrysotile, erocidolite and amosite, and obtained similar tumour incidences with all 3. Since both these authors used gravimetric dust estimations, the results could indicate that more fibres of chryso- tile are required for tumour production than any of the amphibole types. Ho y- ever, it appeared desirable to reappraise this problem with a series of experiments in which the effects of both fibre mass and fibre number could be compared in the same study, _ MATERIALS AND METHODS For any given mass, the U.I.C.C. sample of amosite has the fewest fibres, and ehrysotile the most, with erocidolite fibres being some where between the 2. It was decided therefore to use amosite as the reference dust, and to compare its pathological effects with those produced by both, erocidolite And ehrysotile clouds of equal fibre mass in one instance, and equal fibre number in the other. The equal-mass concentration clouds had a target mean concentration of It) mg/m3, which was considered to be high enough to cause significant pathological change (Wagner etal., 1974). This figure is more than ICO X the present British hygiene standard. Higher concentrations were avoided, because ehrysotile asbestos tends to produce more "thistledown" floes, and so form clouds which have a high proportion of non-respirable material. As this does not occur for amphibole asbestos it is very difficult to draw a direct comparison between the different types at higher con centrations. The ehrysotile and erocidolite clouds calculated to have the equivalent number concentrations had 2 mg/m3 and 5 rag/'m3 respectively (Beckett, 1976). This study was undertaken using white SPF rats of the Han strain. 'The 6 groups each consisted of 48 animals aged 3 months at the start of the experiment. They were exposed to asbestos fibre for 7 h/day, 5 days a week. for a total of 224 days during an elapsed time of one year. Twenty rats of similar age were maintained in the same unit as controls. So that a comparison could be made with previous experiments. U.I.C.C. samples of amosite, ehrysotile A and U.I.C.C. erocidolite asbestos dust samples (Timbrell, Hvett and Skidmore. 1968) were used. The clouds were generated with a modified Timbrell dust generator (Timbrell d al., 1970) and the inhalation chambers were of design similar tTimbrells but- with dimensions modified ; ' * [ ) 1 > | / ( I ) I ,, | I , * luilitlv to fit. the -.size-selected b 11 before lu-ir ream. This er i nimble dust i iiiu'mtoring was i ;iiul daily mass < obtained for all M.R.E. sampler more. Hamilton e measure t he cone and amosite chi ehrysotile t his in imdersamplftj am (Beckett, 1976) monitor the ehn shown to give a with l>oth crock meats were also t-l mtnber. -or the eh8 ,med to be < ..oiial monitorir standard aampl Asbestosis Res membrane-filtct open Gelman ti. at a How rate ai give an uptimur examination (1 The filters wi contrast micro and Beckett" e Beckett, 1977) of view being * wen* taken ii inhalation peri iii(* fihres cou - ter than 5 . an aspect , .,-yth and obtained part! and partly by (Beckett, 1971: Four anima were killed on and 4 more t animals were allowing then in order to tumour deve1 of the popuk. animals were start of dusti the t-xperime ' maiding an . MIDDLETON ions, the re*.,. , fibres of chry.su>r^HTmm our production itphibole types. Howesirable to reappraise' series of experiments of both, fibre mass and be compared in the .AND METHODS .. the U.I.C.C. sample of -;t fibres, and chrysotile elite fibres being none . 2. It was decid ' to as the reference <iu. athological effects win. both erocidolite aim qual fibre mass in one ire number in the other, ucentr&tion clouds had ntration of 10 mg/m3, bo be high enough to ilogical change (Wagner e is more than 100 x the iene standard. Higher voided, because ehrysoproduce more `"thistleown clouds which have jtfh^spirable material. phibole asbestos u^a direct comparison types at higher convsotile and crocidoiii have the equivalent ns liad 2 mg/m3 and (Beckett, 1975). adertaken using white rain. The 5 groups eacii - la aged 3 months at the at. They were exposed it/day, 5 days a week, during an elapsed time 'fits of similar age were ne unit as controls. So could be made with . U.I.C.C. samples of md U.I.C.C. erocidolite (Timbrel!, Hyett and used. The eiouds were edified, Timbrell du.-! et al., 1970) and tl>p ere of design similar t- dimensions modified ) i * i < \ t f FIBRE NUMBER FA* MASS IN ASBESTOS BIOEFFECTS 675 ihtly to tit the available space. The dust Tissue used for histological examination v. ,ssize-selected by a c.vclonesystem (Beckett, was fixed with 10% formal saline solution J:,75) before being added to the chamber and embedded in paraffin wax. Lungs were iiirstrcam. This ensured a high proportion of fixed by inflation. Sections were stained wish respirable dust in the clouds. Gravimetric cither haematoxylin and eosin (H. and E.), monitoring was carried out during dusting, Van Geison's method for collagen or Gordon and daily mass concentration measurements Sweet's stain for reticuiin. obtained for all the chambers. The N.C.B.- For the quantitative estimation of fibrotic M.R.E. sampler (C&sctla Type 113A; Dun- lesions produced in the rat lungs by the more, Hamilton and Smith, 1964) was used to different asbestos clouds, the following method measure the concentrations in the erocidolite was adopted. Lung tissue was examined from and amosite chambers. At 10 mg/m3 with all the animals killed at the first 2 intervals ehrysotile this instrument had been found to 12 and 18 months after the start of dusting. tindersample, and a vertical etutriator Bystem Of the animals that survived until the final (Beckett, 1975) was therefore used to killing date at 860 days, 6 were examined monitor the chrysotile eiouds. This had been from each group. The remaining animals were shown to give similar results to the M.R.E. examined only for the presence or absence of with both "erocidolite and amosite. Measure tumours. The quantitative estimations of the ments were also made of the total dust in the fibrous lesions produced in rat lungs by the chamber. different asbestos clouds were undertaken F'or the chambers whose eiouds were using the following procedure. The entire lung planned to he of equal fibre number, addi tree with the heart was embedded together, tional monitoring was undertaken, using the and sections were cut in the coronal plain to standard sampling method described by the include parts of all lobes. Sections were cut Asbestosis Research Council (1971). Each at 4 different levels in each block, and were membrane-filter sample was taken using an at least 1 mm apart, and groups of serial open Gelman filter holder facing downwards, sections were mounted from each of these at a flow rate and sampling time calculated to levels for use with the different staining give an optimum density tor the microscopical techniques. For all lesions, the H. and E'. examination (1-3 fibres per graticule area). sections from each animal were scanned with The filters were counted with a phase- the light microscope using an eyepiece contrast microscope containing a ``Walton graticule consisting of a 1cm square sub and Beckett" eyepiece graticule' (Walton and divided into 1.00 units of 1 ram3. Viewing Beckett, 1977) to define the area of the field magnification was X 60. The area of large of view being evaluated. At least 50 samples regions of interstitial fibrosis was estimated wore taken in each chamber during' the for each slide by counting the number of grid inhalation period, not more than one per day. squares involved and presenting the results The fibres counted were those with a length as a percentage of the total lung tissue in the greater than 5 pm, a diameter less than 3 pm section. An average figure for the animal was and an aspect ratio of more than 3:1. Fibre produced by combining the result from all length and diameter distributions were 4 sections. The very early fibrotic lesions obtained partly by phase-contrast microscopy were usually much smaller than one grid and partly by scanning electron microscopy square at the magnification involved and (Beckett, 1973). since they were associated with the respiratory Four animats from each inhalation chamber bronchials they were also widely scattered. were killed one year after the start of dusting, For this type of small lesion, the calculations and 4 more 6 months later. The remaining were based on the number of squares that animals were left with the intention of contained the small areas of fibrous tissue and allowing them to survive their full life-span, the results from all 4 sections were again in order to study the frequency of lung- presented as a percentage. r.umour development. However, the survival Asbestos retained in the lungs of selected "f the population was extremely good and 56 animals was recovered by a low-temperature nimals were still alive 860 days after the ashing process. This was conducted in a fart of dusting. It was decided to terminate stream of Ou excited by a radio-frequency the experiment at this point, and all the discharge (Gleit and Holland. 1962). Any remaining animals were killed. residual lung salts were removed by washing f)76 .1. DAVIS, S. BECKETT, R. BOLTON, t>. COLLING8 AND A. MIDDLETON the samples in .1 ml of cold ( --- 2()''C1) (i*2m HC1 before gravimetric estimations of the amounts of asbestos recovered were made using the infra-red spectrephotometric tech niques described by Middleton, Beckett and Davis (1977). To determine the percentage retention of the different dust types it was assumed that the rats breathed at the rate of 100 om3/min during dusting. Calculations wore made using this volume and the gravi metric levels of each dust cloud. Bust retention estimations were under taken on the left lungs of animals, the right lung being retained for histological study on each occasion. At the first killing date (12 months after the start of dusting) 2 left lungs were analysed from each group of animals, but on the second occasion 6 months later 4 left lungs were available from each group. Because of t.he suggested association between laryngeal carcinomas and asbestos in humans (Steli and McGill,- 1973) the larynxes were examined from" all "animals, both in the 5 experimental groups and in the controls. For histological examination the larynx was serially sectioned in the longitu dinal plane and approximately 8 evenly spaced sections were mounted for examina tion from each specimen. RESULTS The dust parameters for the 5 chambers over the period are given in Table I. The mass concentrations were very close to the target set at the beginning of the experiment. More than. 50% of the daily concentration measurements in the equalmass chambers were within 3 mg/m3 of the target concentration. The 3 equalnumber chambers were dosed at gravi metric concentrations determined by a number cs mass correlation obtained during previous short-term experiments (Middleton et al., 1977). This correlation was based on 30 membrane-filter samples for each type of asbestos and had a large uncertainty (coeff. of variation ~70%). This was due to the fact that the mass concentrations were integrated measure ments taken over 7 h. The counting samples, on the other hand, were limited to a few minutes, owing to the high dust concentrations giving deposits which were too dense to evaluate for the Jargervolume samples. As fluctuations in concen tration occur during the day, and mem brane filter samples cannot be. evaluated with a reliability better than 30% (National Health and Medical Research Council, 1970), uncertainties of this order are inevitable. In this present study, between 50 and 100 membrane-filter samples were evalua ted during the 12-month inhalation period to check this correlation, and gave mean fibre concentrations of 550 fibres/mi for amosite, 390 fibres/ml for chrysotile and 430 fibres/ml for croeidolite. This meant that 0-1 mg of dust/m3 of air was equival ent to 19*5, 8-6 and 5*5 fibres/ml for chrysotile, croeidolite and amosite respec tively. The uncertainty in the measure ments was of a similar order to that in the previous experiments. There was no significant difference between the fibrenumber concentrations in the croeidolite and chrysotile chambers (P = o-4), but the amosite chamber was significantly Table I.--The Mean Mass and Fibre-number Concentrations oner the Exposure Peiiod Asbestos Typo Type of cloud Target concentration (mg/m3) Moan mass concentration (mg/m3) Mean ratio of total to respirable dust Mean fibre-number concentration (fibros/ml>5 jun| Mean fibre-number concentration (fibres/ml>20 pm) (estimated from use-distribution data) * Estimated figure Chrysotilo Equal mass 10-0 SM) 1-4 : 1 Chrysotile Equal fibre number 2-D 2-0 1-3 : 1 Croeidolite Croeidolite Amosito Equal mass Equal fibre Equal mass number anti fibre number I (HI KM) 1-2 : 1 .-(> 4-S) 1-1 : 1 (()() 10-0 MS : t :!(> X0* 431) 550 .`1150 72 :i4 17 K lifferent from I " w-animals inti ibabiy dosed verage tiumbei between the fib was, however, v the equal mass c Taking a serf to monitor the although subjee does in fact c industrial situat are frequently t exposure (Bep and Productive A series of sa were taken in t : ute filters. T. size distrib u'lning electr cant difference different samp!' The iength dii than 0-fi jam an uf fibres broadc Pig. I and Fig. The surviva from the five shown in Tab there were nc survival times with the differ t be average ' " K II.- Su in the. I ihe Experii Months. Cro Chamber me months lo mp/m5 IChrysotile * - iru/ma > lilnm/rn3 \inriNih* , linir/rnJ *t* tUto . MIDDLETON tn^'letermined by . ation obtained .rA^PFrm experiments 977). This correlation m bra me fil tor sam pin* u-stos anrl hud. a large of variation ~7l>-;.). ie foot- that the mass integrated rneasure- 7 h. The counting er hand, were limited wing to the liigh dust g deposits which wen1 uate for the largerduet nations in conoen t the day, and men- cannot be evaluate better than .'}0o-pl nd Medical Research rtaintios of this order udy. between 50 and samples were evaluamth inhalation period ition, and gave mean of 550 fibres/ml for ill for chrysotile and ividolite. This meant tidUif air was equivali^^K-5 tibres/ml for amosite respectnfy-. in- the -measure" ir order to that in tho ds. There was no between the fibre>n in the crocidolite ibers (P -- 0-4), but er was significantly ' the. Exposure Period CVocKlofifco Amosite (ual fibm E<(ual mass itnbcr utii I fibre number .>() -Ml M:1 Ill-O lll-O 1-15 : 1 4.`l<> r,so 17 ti ( / ) ft ltt ( 9 FIBRE NUMBER i'S MASS IN ASBESTOS BIOEFFECTS (J77 t. fferent from the other two (P<0-01). 's he animals in this chamber were therefore probably dosed with a slightly higher average number of fibres. The difference between the fibre-number concentrations , vcK'doM* CVmhfc *3 Incttlt was, however, very much smaller than for the equal mass chambers. Taldng a series of short-period samples 'Vi to monitor the fibre number exposure, although subject to this large uncertainty, does in fact correspond closely to the industrial situation, where JO min samples are frequently taken to monitor a person's exposure (Department of Employment v *>0 i 40 * jO lv 20 , ' . and Productivity, 1970). to A series of samples on Nuclepore filters were taken in addition to those on mem brane filters. These were used to measure the size distribution of t-he fibres using a scanning electron microscope. No signifi 'i cant difference was found between the different samples from the same chambers. The length distribution of fibres longer than 0*5 /am and the diameter distribution of fibres broader than 0*2 pm are shown in Fig. I and Fig. 2 respectively. i A ' 4 to V> <0 M so ;o too in mie/onh Ftc. U--Length <iL>tributions of fibres longer than 0-6 ftrn. (Scanning electron microscope measurements.) The survival times from the animals from the five inhalation chambers are different groups was considered, however, shown in Table II. These indicate that some differences were noticeable. At the there were no significant differences in end of the I2-raonth inhalation the rats survival times between animals treated from the 3 amphibole chambers averaged with- the different asbestos clouds. When between 500 and 510 g-each. Those from the average' weight of animals in the the high and low chrysotile chambers, however, averaged 465 and 467 g respec Table II.--Survival patterns for the Anim als in the. Different Inhalation Groups. The Experiment urns Terminated at 29 Months. Groups of 4 animals from Each Chamber were Kitted at both 12 and 18 months Months after start of exposure 12 IS 24 29 tively. This differential was gradually reduced with time, until at 20 months after the start of dusting all groups averaged slightly over 500 g per animal, with the exception of the low-crocidolite group where the average was 494 g per animal. Subsequently, with advancing age, all animals gradually lost weight, but there were no significant differences between the different dust groups. 10 ing/rn3 Chrysotile 3 mg/m3 ('fr-yaotilo ].* .ng/m3 Li mg/m3 1 `rocidoUte mg/m3 Crocidolite 48 40 21 12 Light-microscope examination of lung tissue from animals in the 5 dust groups 48 40 26 ; killed 12 months after the start of dusting 47 39 2'Z ii showed 3 distinct types of lesion that could lie associated with asbestos dust. None of 47 41 25 a these lesions were seen in control animals. 46 37 22 K The first type of lesion consisted of I 67K J. DAVIS* S. BECKETT. R. BOLTON. t\ POLLINGS AND A. MIDDLETON V9.Q9 99 1 W1 * "H A CrOcfctolttt t ChryvotiU Q Amo'tiU 1<o 4 " 30 . I. i to . 5 0.5 J 0:y " } "I ' ^ -- I---1 " ' T1 0.2 0.3 04 0-5 0 9 to 19 pi*mUt in micron* 3.0 Flo. 2.--Diameter lintribuf.ion of fibres i>roader than 0-2 fim. (Scanning electron microsoope measurements.) aggregates of dust-containing macro phages, giant cells and fibrous tissue in association with the respiratory bronchi oles and alveolar ducts (Fig. 3). These areas stained strongly positive for reticulin and more weakly for collagen, although some collagen was always present at this stage. The second type of lesion consisted of the replacement of the epithelial lining of many respiratory bronchioles, alveolar ducts and associated alveoli by epithelium of bronchiolar type, it was not possible, however, to determine whether this was due ---perplasia of the bronchiolar lining or iu . .aplasia of the alveolar epithel ial cells (Fig. 4). Both these types of lesion were frequently found together around any one respiratory bronchiole, but either could appear on its own. The third type of lesion consisted of the thickening of alveolar septa over quite large areas- of lung tissue (Figs. 5 and 0). The alveoli involved were lined with rounded epithel ial cells, probably Type 2 pneumocytes, and Cordon Sweet's stain showed an increase in the reticulin network in the septa walls although no collagen was present in the early stages. While most sections of alveoli in each animal contained only an occasional macrophage packed with asbestos fibres, those from areas of interstitial fibrosis were often filled with dust-containing cells. In these cases, however, it was noticeable that each cell contained relatively little dust. The areas of interstitial fibrosis could become quite large, often 4--5 mm in diameter, especially in the oldest animals, but early lesions were small, and appeared to be centred on one bronchiole. With the increasing age of the animal the depositions of fibrous tissue in the interstitial space was often greatly increased, so that the total alveolar wall thickness could become as much as 50 100/am (Fig. 6). In these advanced cases, the thickened septa stained positive for both reticulin and collagen. An. alternative to advanced fibrosis, however, was the continued growth of the rounded epithelial cells, with the subse quent compression of the alveoli to produce an adenomatous appearance. In some cases, positive adenomas were found forming in these areas. In a few animals small areas of squamous metaplasia of the alveolar epithelium were also found. Quantitative estimations of these 3 types of lesions are shown in Table III. It was found that both of the chrysotile clouds had produced much more of the early granulomatous deposits around ter minal bronchioles and alveolar ducts than any of the amphibole dusts (P<CM)01). The 10 mg/m3 chrysotile cloud had pro duced significantly more peribronchial fibrosis than the 2 mg/m3 chrysotile cloud {P< 0-001). These lesions showed no further increase in numbers after the end of the inhalation period. Subsequent studies of tissues taken at either 0 or l" M.,inths after tl wel a slight lie lesions. i .-reased areas i i * V Kki. 3.---Depos blasts and re lesion develo .< 230. > 'id. 4.--Tir P'Utlltl Wl MIDDLETON I thickening large areas of und ti). The alveoli a'ifrh rounded epifcheb ',ypt` pneumocytes, s sfain showed an ulin network in the -h tto collagen wnu stages. While most !U-h animal contained macrophage packed those from areas of wre often tilled with is. In these cases, ioeable that each cel i little dust. The area? s could become quite n diameter, cspeciall\ `Is, but early lesions aied to bo centred on > the increasing age of itions of fibrous tissue ace was often greatly ui total alveolar wall ome as much us 50hese advanced cases, stained positive for <^^fcvanced fibrosis, iWPed growth of the ells, with the subse- of the alveoli to utous appearance. In adenomas were found `as. In a few animals toua metaplasia of the .cere also found, rnatjons of these 3 shown in Table III. oth of the chrysotiie d much more of the : deposits around terd alveolar ducts than de dusts (P<Q-00l). sot ile cloud had pro- more peribronchial g/m3 chrysotiie cloud lesions showed no umbers after the end period. Subsequent ken at either 6 or 17 FIBRE NUMBER r.S' MASS in asbestos kioeffects ijiy t> uifchs after the end of dusting in fact snowed a slight decrease in the frequency of the lesions. However, this was due to the increased areas of interstitial fibrosis that had developed by these times, which reduced the area of tissue in which the peribronchial lesions could be recognized with certainty. The amphibole dusts "! . ' S3* sm * . ----- -------- ..___ Fig. 3.--Deposits of granulation tissue, consisting of dust-containing macrophages, giant cells, fibro blasts and reticuiin fibres, associated with a terminal bronchiole) and several alveolar ducts. This lesion developed in a rat treated for 12 months with a cloud of ohrysotde asbestos of 10 mg/m1. X' 250. EfeMialStSBEPW * . * .' s - i . __ Fro. 4. --Tissue reaction to asbestos dust around terminal and respiratory bronchioles in an animal treated with ohrvsotile asbestos. Bronchial epithelial cells now line some alveolar spaces. : 260. liKll J. DAVIS.. S. BECKETT, R. BOLTON, P. CODLINGS AND A. MIDDLETON T Table Hi.--Levels of Lung Fibrosis Produced, by the. /HJfJ ll) mg/m3 Ghrysotile 2 mg/m3 Chryaotilc ) i 'mills (Pornmeter, > - ii1 Anuiaito Tfmo uflersf net of exponunt (months) 12 18 29 P.-ribiimchiolaJ: fibrosis Extension of bronchial epithelium to alveolar oiucta and alveoli Interstitial fibrosis No. of rats in aamplo 19-3 17*l 15-0 (12-7-24-5) (15-1-19-2) (T2-7-2U-I) 2-68 2-4 I'43 (1-28-4-4) 12-3-2-6) (0-7-1-9) 0-48 0-9 9-16 (0-1-8) (0-26-1-86) (3-8-14-4) 446 12 18 10-7 9-9 (7-8-12-7) (7-5 -11-77) 1-7 (1-1-2-5) 0-35 (0-1-2) 4 4-03 12-8-0-6) 0-83 (0-2-9) a - l-sfr1- (5-J.W' V'" jW ; 2i (0-6.: Mf jh (0-' Is i.:- .vmi :i-u (l-u H-m u-12 ill l>-4) 29 4-2 (2-5-.V."i) 3-1)5 (1-8- 5-5) 2-58 iH-5-11 Fio. 5.--An area of interstitial fibrosis from an animal treated with ehryaotile asbestos for 12 months. The alveolar septa are thickened aud they aro surfaced with rounded epithelial cells. Moat alveolar spaces contain aggregates of cells, many of which are duat-oontaining macrophages, x 250. produced relatively little early peri bronchial fibrosis (Fig. 7) but dustcontaining macrophages still aggregated around all the terminal bronchioles. For the most part- they did not appear to be held in place by any reticulin network, and yet some aggregates were still present without associated fibrosis in the oldest animals examined (Fig. 8). The extension of bronchial epithelial cells in alveolar ducts and alveoli varied much less between the different dust douds than tV peri bronchiolar fibrosis. In common w- hese fibrotic areas, however, there appeared to be no long-term progression of the lesions after 12 months from the start of dusting. While areas of peribronchiolar fibrosis and peribronchiolar alveolar epithelialization appeared evenly distributed through the lungs of any animal examined, areas of widespread interstitial fibrosis were much more haphazardly arranged, and these areas were completely absent from some animals examined at between 12 and 2!) months after the start of the experiment. Consequently the figures for interstitial fibrosis at the 12-month stage based on only 4 animals in each group are not considered to show significant differences between the asbestos types. Only 4 animals were included in the groups taken at 18 months, so that the same considera tion might apply although by this time most animals treated with chrvsotile had I t rP Km. 0.--Advs Sumo aivcolt -'i'fubly m< treatei . ;ilc. For iiiei- rhe end available froi .-how that al developed sij fibrosis than earlier. The produced lari animals exar more than < croup (P<li the other 4 dramatic, hi definite grad -l.uifl havinj ) A. iVUIMJLKTON r fr'UJKK NUMBER C.V MASS IN ASBESTOS HtOEFFECTS Fihroulf Produced by l, i,k,Ao* f/- '<& (Parameters ax Described in Methods section) 2 mg/m9 Chry^ttii,. I 1 1,1 iiJ Amomtn 10 mg/m3 < 'roculolitc 5 mit/m3 (.'rocimlolitr- 12 IS .i.|> Hi-7 !MI (7-8-12-?) (7-5-l 1-77) 1-7 H>: !) (1-I-2-5) <2-8- -K) 0-35 (S3 1-4) (ti-1-2) (o-2-U) 44 Jli ili i IS H (HI ) 0-12 (I) 0-4) 4 2D 4'2 2-5-0-5) 3-05 1-8 s-.r, 2-58. l-t-5-1) 8 12 2-BX 11-25-4-05) 11-85 (0-47-1-27) 0 (0) 4 18 4-25 (2-H-K-4) 2-15 U-3-4-8) i'i-07 (0-0-27) 4 2!) :mi (2-5 -H-0) 1-08 (1-1-3-U) 1-38 (0-4-1) a 12 2-8 12-2 3-8) I -35 (0-U- l-(i) 0-42 (d-t-7) 4 18 2-3 (2- l-2-li) 1-25 (1-2-1-8) 0-04 (0-1-71 4 20 2-47 11-25-4-1) 1-80 (0-W7-3-4) o-78 (0--2-23I 8 tiS i [jo asbestos for 12, -1 ('p;thfi!ial (M'lls. Moat ' macrophages, v 250. tlveolar epithefializa distributed through ial examined, areas of d fibrosis were much i trange.d, and these ly absent from some between 12 and 29 i t of the experiment. Aii res for interstitial >nth stage based on each group are not ignificant differences os types. Only -1 l in the groups taken the same eonsidera'hough by this time with chrysotile had ^ j , 1 I ' 1 Fig. 6.---Advanced interstitial fibrous in a 32-month-old rat after the inhalation of chrysolite dust. . Some alvooiar. s$<?pfc are >400 ftm*in*fchi'elcnosfi and stein strongly posit-ivo for coltagen.- 2/30. noticeably more interstitial fibrosis than those treated with either atnphibole sample. For the last sample, 17 months after the end of dusting, 6 animals were available from each group. The figures show that all groups had by this time developed significantly more interstitial fibrosis than had been present 11 months earlier. The high chrysotile cloud had produced large areas of fibrosis in all the animals examined, with average figures more than double those for any other group (P< 0-001). The differences between the other 4 asbestos clouds were less dramatic, but there appeared to be a definite gradation, with tfye low chrysotile cloud having produced more interstitial disease than any of the amphiboles (/<0-01). At the same time, amosite appeared to have produced more damage than the 2 erocidolite clouds and the high erocidolite had produced more fibrosis than the lower cloud of the same material. The levels of significance of these latter observations is, however, low. The incidence of neoplasms of the lung and mesotheliomas that were found in the different experimental groups is shown in Table fV. The incidence of lung tumours closely follows the level of lung fibrosis, and all the malignant lung tumours were found in animals that had inhaled chrysofcile dust (P<!.M)0I). Even benign pulmon ary adenomas were more frequent in these 62 .r. DAVIS. S. BECKETT. K. BOLTON, 1'. CULLINGS AND A. MIDDLETON Table IV.--Lung Tumours and Mesotheliomas Tumour Adenoma Adenooarcinoma Squamous carcinoma Plourat mesothelioma Peritoneal mesothelioma 10 mg/m1 Chrysotile 40 animals 7 (5 u 0 0 d mg/m3 Chrysotile 42 animals 6 1 1 0 1 10 mg/m1 Amosito 43 animals 2 0 0 0 0 10 mg/m1 Croeidolite 40 animals I 0 n fi 0 5 mg/'m" Croeidolite 43 animals 4 u ti 1 a Control 20 animal] 0 0 (l ll 0 > t* ns. S.--Aegregs / Table Etc. 7.-- Small deposits of granulation tissue associated with respiratory bronchioles in a rat treated with ainosito dust. These areas contained dust-laden macrophages, fibroblasts and reticulin fibres, i.riant-cell formation was rare in the lesions caused by both amosito and croeidolite asbestos, r 250. 2 groups than in animals treated with either variety of amphibole (P -- 0*006). Four of the adenocarcinomas had meta stasized to the pleural cavity (Fig. 9). The typical histological pattern of one of the squamous tumours is illustrated in Fig. 10. Neither had metastasized to the pleural cavity, although they had reached di&r meters of 6 and II mm respectively and had caused marked swelling of the lung lobes involved. Both showed evidence of direct invasion into the surrounding tissues. Only 2 mesotheliomas were found in this study, one solitary spindle-ceii tumour in the pleural cavity of an animal treated with croeidolite, and an abdominal mesothelioma in an animal that had inhaled chrysotile. This latter tumour showed the histological pattern previously described (Davis, 1974). No pulmonary tumours were found in control animals. The tumour incidence from sites other than the Jung, and excluding mesotheliomas. is shown in Table V. If the tumour totals for each group are compared, the high chrysotile group and the amosite group appear to have more evidence of neoplasia than controls. However, with the relatively small groups of animals these differences are not significant. Of interest was the finding of relatively large numbers of peritoneal connective-tissue tumours. One was a I**'---myofibroma that had developed on the . . ! of the small intestine. The remaining tumours, however, were malignant and multiple, and macroscopically were very similar to peritone;11 mesotheliomas. Histological examination ) ' , > ' > Site of tumour tvp Submitatieous eonr tiMHue tumours IVntoneal conncofc tissue tumours < t-ti'OKftpcomos Testicular tumours S((Hiim<niH tumour 'bi> Epidermis I i*>il tumours il rumours ,.l tumours ..;.immu/leuka< I .tiii-icttrift tumoui Totals however, show the mesothelio Some appearer fibrosarcomas, and nuclear p ing one found tamed large with small spit 11 istological the anirr urx. in t 1. MIDDLETON ia>/ iii/m3 _ F'ldolito 43 animals (l li ! 0 ('ontrol 20 an i multi 0 0 0 0 0 - V" FIBRE NUMBER IV? MASS IN ASBESTOS BrOEFFECTS 6H3 ncbiulos m a rat treated i^Mund retieulin fibres, asbestos. x 250. 1974)'. No pulmonarv id in control animals, i fence from sites other excluding mesothelioable V. If the tumour up are compared, the. oup and the amosite ave more .evidence of itrols. However, with groups of animals these significant. Of interest datively large numbers eetive-t-issue tumours, ivofibroma that had li of thesmail intestine, nours, however, were iple, and maeroscopicdmilar to peritoneal tological examination, Kid. 8.--Aggregation!* ot` maerophagefi packed with amosito fibres around alveolar ducts in the lungs of 32*Tnonth-okl rot. > 250. i [ Table V.--Sites of Tumours other than Lung (B, benign, M, malignant) Chrysotilo Chrysotile Amosite Crooidolite Crooidolite f Site of tumour type 10 mg/m3 40 animals B M. 2 mg/m3 42 animals BM 10 mg/m3 41? animals BM 10 mg/m3 40 animals B .U 5 mg/m3 4.1 animals BM Controls 19 animals BM Subcutaneous oonncctivo- tissue tumours 1 I 23 1 L l Peritoneal connectivo- tissuo tumours . . Osteosarcomas : ' Testicular tumours" ' 2 1 l1 -j.. _ .l.. 4 ' " o 1 3 1 .Squamous tumours of the Epidermis Parotid tumours 1 2i I 1 1 Adrenal tumours Thyroid tumours 11 o L1 11 Lymphoma/leukaemia Pancreatic tumours 2 i Totals 4 S 2 5 9 5 I 4 0 fi 0 3 however, showed marked differences from the mesotheliomas normally found in rats. Some appeared to be poorly differentiated fibrosarcomas, others showed gross cellular and nuclear pleomorphism and 2, includ ing one found in a control animal, con tained large muitinucieate cells mixed "'ith small spindle cells. Histological examination of larynxes from the animals in this study showed no tumours. In the oldest animals that had inhaled asbestos dust, some small areas of epithelial hyperplasia were found involv ing squamous cells, usually at the bases of the vocal cords. However, similar areas of hyperplasia were found in control animals, and it was assumed that these changes were associated with advanced age. The weights of asbestos dust extracted from the lungs of animals in the different inhalation groups is summarized in Table <)4 DAVIS. S. BECKETT, R. BOLTON", P. POLLINGS AND A. MTDDLETON VC. Although oil this occasion only left lungs were available for dust estimation, previous short-term inhalation studies had involved dust estimation from both lungs taken separately, and these had indicated that the asbestos content ratio between the left and right lung was 0-6 to l. Figures in Table VI, therefore, indicate actual left lung content and the estimated total lung content calculated from the above ratio. These calculations indicate that, for a given dust cloud, far more amphibole asbestos is deposited and retained in the lung than is the case with Fro. in. Push. *\ mid H>.---Histological patterns of an adenocarcinoma and a squamcnts-O'll carcinoma that developed in tho lungs uf rats treated with ehrysntiW* asbestos. 2J50. I :'ype of asbestos i licysutila (lu-ysutilc (`rnciilolitc Crooitlolito Anionite chrysotile. Foi the lung cont crocidolite at period was ver ((intent was i ,*.;re. A com" c ,ids indica retention after f wire as high t lo mg/m3. \Vi" percentage re cloud was sdig mg/m3 cloud, asbestos conte after the end that chrysotilt lung much me lioles. During tile groups si dust content c m Durable l ('cils were oi In tltis stuc .`(ltd libre nun disease. it wai given airborn chrysotile pre than the sam samples of e This indicate.for all types appropriate.' of the dust was spoilt- b} omosite ctouc chrvsol IDDLETON T FIBRE NUMBER l'A' MASS IN ASBESTOS BIOEFFECTS tj^hfore. indicai. 1 Table VI.--Levels of Asbestos Recovered from Lung Tissue iflHnc estimate I cmatetl from the filiations indicate olotid, far more < deposited and m is the case with Type of uxliesros Chrysotilo ehrysotile Croeidolitc LVoeiiioIift' Amonitf Respirable concentration r----------------- *----------------- > Target Actual (me/m3) 10 9-9 > 2-0 10 10-0 r 4-9 10 10-0 Recovered asbestos Days after rxponliro 7 7 182 7 182 7 182 7 182 ftglleft long 1 Actual) 520 228 198 (ifi .'1212 2731 1279 978 8386 o77 gg/rat (Estimated! 1417 048 526 ISO 8750 7440 3484 265!) 9 1 (ill 7020 Estimated retention 1-5 0-7 2'8 I'D 11-3 7-9 7'0 5-8 11-7 7-6 ll enromoma that 'SO. ehrysofcile. For the clouds of 10 mg/m3, the lung content of both, a mosite and crocidolite at the end of the dusting period was very close, while the ehrysotile content was only 15 t-o t% of this figure. A comparison of the 2 ehrysotile clouds indicated that the percentage retention after i.2 months of dusting was twice as high for the 2 mg/m3 cloud as for 10 mg/m3. With crocidolite, however, the percentage retention for the 10 mg/m3 cloud was slightly higher than for the 5 mg/m3 cloud. The differences in lung asbestos content between 7 and 182 days after the end of dusting would indicate that ehrysotile had been cleared from the lung much more quickly than the amphiboles. During this period both the chrysotile groups showed, a. reduction In. lung dust content"'Of 50 to 70% while the comparable figures for the amphibole clouds were only 15 to 25%. DISCUSSION In this study of the effects of fibre mass and fibre number on asbestos-related lung disease, it was clearly demonstrated that a given airborne mass of U.I.O.C. Rhodesian ehrysotile produced far more lung fibrosis than the same airborne weight of U.I.O.C. samples of either amosite or crocidolite. This indicates that a single mass standard for all types of asbestos would be in appropriate. To some extent, a comparison t>: the 3 dust types on a fibre-number basis was spoilt by the high fibre count of the amosite cloud. However, the figures for the 2 mg ehrysotile and the 5 mg crocidolite clouds were extremely close, 300 and 43o fibres/ml respectively Once again the animals treated with ehrysotile had devel oped significantly more lung fibrosis than those treated with crocidolite. All these results could be taken to indicate that ehrysotile is much more fibrogenic than either of the amphiboles, and they might be considered to agree with in intro cyto toxicity studies which have reported that ehrysotile causes greater cell damage than either amosite or crocidolite (Klosterkfttter and Robock, 1975). From this it might be suggested that the standards for ehrysotile should be more strict than for cither amosite or crocidolite. In fact, however, consideration of the fibre length distribu tion of the various dust clouds given in Fig, 1 suggests another possibility. Fibre counting for monitoring the dust clouds supposed to have equal fibre numbers was undertaken using the procedure laid down for the present hygiene standards (Health and Safety Executive, 1976) which records all fibres over 5 fim in length but does not allow for fibre variations above this length. A complete fibre-length distribu tion produced by scanning electron micro scopy showed that the ehrysotile clouds used in the present study had many more fibres over 20 fan in length than either of the amphibole clouds (Fig. I and 2). No reliable estimates are available relating fibrogenic potential to fibre length, but a uuraber of authors have suggested that for mesothelioma production at least, carcino genicity depends on fibre lengths in excess of 10 to 20 pin (Maroudas. O'Neal and 686 ,T. DAVIS, S. BECKETT, R. BOLTON. P. COLLINGS AND A. MIDDLETON Stanton, 1973; Stanton et at,, 1977). Since Whether the increased fibrogenic and > 'ninths has mar in the present study the only malignant neoplastic effect of ehrysotile found in this ' study of : lung tumours were produced by ehrysotile study was due to ehrysotile itself, or to and this was also by far the most fibrogenic, increased fibre lengths in the ehrysotile these results could support the long-fibre clouds, it does not change the position ! -ntion that ...ii't-term admii i Middleton at < theory of carcinogenesis in general, and regarding human hazards from ehrysotile confirmed that t also indicate that tho same parameters are exposure, since the present British indust l ion of ehrysotile involved in the fibrogenic response. This rial asbestos dust standards were based on amphibole types would indicate that the present protocol epidemiological data from ehrysotile- ' reduced when tl for fibre counting is inadequate, and that exposed working populations. The human increased. With either the 5 fim. limit for counting should position regarding the types of ehrysotile percentage reten be raised, or counts should be broken cloud met with in industry is, therefore, s double that for down into different fibre-length groups. already known, but the new data indicate retention of crc Since, however, biological knowledge on that some amphibole clouds may be less 1 the reverse, ant the exact lengths of fibre that cause dangerous than previously expected. higher with the damage is still not definite, a suitable At present erocidolite is considered in reasons for this 1 compromise might bo to retain the present most countries to be the most dangerous with certainty, t 5 /*m lower limit, but to include an asbestos type and its use is banned in some , 1 "'rth distribute additional count of fibres > 20 fim long. cases. However, this situation is largely i is in prog It might,.be-found that this..latter figure due to the association of erocidolite with the' i.- ..ate the rea: correlates better with epidemiological data production of mesotheliomas in humans. for asbestosis and bronchial carcinomas This connection is well documented, 3 Tbix work was un than the 6 fim counts. but there are no epidemiological data | I [ii-iigrammo fancied f I'oiineil. The finding that ehrysotile asbestos indicating that croeidolifce Is worse than produced far more lung fibrosis and pul the other asbestos types at producing lung monary neoplasia than the amphibole fibrosis or bronchial carcinomas. The asbestos types was not expected from present study would indicate that as far as r: previous animal inhalation experiments. Wagner et al., in a large study published in 1974, had included groups of rats treated for 12 months with 10 mg/m3 clouds of U.I.O.n. samples of Rhodesian ehrysotile, lung pathology is concerned erocidolite is the least dangerous of the absestos types examined, even though as much as 6 X more erocidolite than ehrysotile was retained after one year of dusting. Because AxRKRTOStS Kesear iiinnt of Airborne Kilter Method. I Tn Uniral .Vote .Vo He< KK'rr. r>. T. (19' Axlxntns Using a I nnttle Orrirjytt. { amosite and erocidolite, so that the results mesothelioma production in response to should have been directly comparable asbestos inhalation is a very rare event in with those of the present study. However, both animals and humans, animal studies Hm KKTT, S. T. I III' turn of (f.r.C.(.`. ' lu.-uro Ohamhi I -I. M. tl. they reported similar levels of lung fibrosis for all groups, and the number of malignant lung tumours produced by the ehrysotile and erocidolite clouds were closely comparable although the amosite so far undertaken have been unable to produce statistically significant numbers of these tumours for an accurate comparison between the various forms of asbestos. Wagner et al. (1974) reported 5 mesothelio uuu of Vet i mix bv rnjoc" - . . 52. 182:5. litreiii'.mknt or E Will H.M. Eaci AxIxwcoa Oust *.' Axl x'xi_o*i Reeulat cloud produced only one such tumour. The reason for this discrepancy between the 2 studies is difficult to determine, since the dust-retention figures in both studies are extremely close. It may be that the elutriation systems used in the 2 studies differed. Wagner did not give fibrelength distributions for the dust clouds used and it seems likely that the ehrysotile clouds used in the present study had a higher proportion of long fibres. mas from 76 animals with tumours in both the ehrysotile and erocidolite groups. We found only 2 mesotheliomas in 123 animals but again both erocidolite and ehrysotile were implicated. No mesothelio mas were produced by amosite in either study after 12 months, but Wagner did find one mesothelioma in an animal treated for only one day with amosite dust. The use of asbestos clouds of difFerhg density over a long inhalation period of - Vote i:. London: llol.i., K. IlHflS) i Aslx'xtox Workers Dl N'MOKK,.f. H.. K 09041 An Instrur iihld Dust for Sitb f. -triritt, lnatnim. 1 Ii.kit, O. F,, A-. " Kliftncally Ex.. lure Di-composit-i l 'Item.. 34, 14o4. iliioxx. I*, or Th Axbcxtimi, III ANT. SAKF i,.. Stutitiur. A.. MIDDLETON r FIBRE NUMBER CN MASS IN ASBESTOS BIOEFFECTS na 7 creased fibrogenic am (^gsotile found in this J^Butile itself, or to gfra in the ohrysotile t change the position ilizards from chrysotile present British indust- randards were based on ata from chryaotileopulations. The human the types of chrysotile industry is, therefore, i- the new data indicate )Ie clouds may be less -viously expected, dolite is considered in he the most dangerou.s use is banned in some lis situation is largely n ofcroeidolite with the itheliomas in humans, is well documented, epidemiological data icidolitc is worse than ypes at producing lung lial carcinomas. The l indicate that as far as oncerned croeidolite is of the absestos types . as much as fi x chrysotile was earof dusting. Because uction in response to is a very rare event in limans, animal studies have been unable to significant numbers of n accurate comparison us forms of asbestos. } reported 6 mesothelios with tumours in both croeidolite groups. We esotheliomas in 123 both croeidolite and licated. No mesotheliol by amosite in either uths, but Wagner did lioma in an animal ilay with amosite dust t'OS clouds of differing inhalation period of 12 i-miths has made it possible to continue i (.' study of asbestos rleposition and u-i cation that- was commenced using short-term administration of asbestos dust (Middleton til nl., 1977). It. has been confirmed that the percentage lung reten tion of chrysotile is much lower than cither atnphiboie types and also that retention is reduced when the density of the cloud is increased. With a 2 mg/m3 cloud the percentage retention of chrysotile is almost double that for a Id mg/m3 cloud. The retention of croeidolite, however, show's the reverse, and retention is marginally higher with the denser dust cloud. The reasons for this have not been determined with certainty, but measurements of fibrelength distribution of retained lung dust in rats is in progress. These results rfiav indicate the reasons for these differences. This work was undertaken os part of tho research programme funded by the British Ashest.os Research Council. REFERENCES Asbestosis Research Council (1971 )jThe Measure ment of Airborne Asbestos Dust by the Membrane Filter Method. Rochdale (Lancashire). A.R.C. Technical Nate No. I. Beckett, S. T. (11)73) The Evaluation of Airborne Asbestos Using Scanning Eleotron Microscope. . 1 nnals Occupat. Hyg., 16, 406.- .. . Uri KKTT, S. T. <1975) Tho Generation and Evalua- ',ion of U.I.C.C. Asbestos Clouds in Animal Exposure Chambers. Ann. Occupat. Hyg., 18, 187. Davis, S. M. G. (1974) Histogenesis and Fine Structure of Peritoneal Tumours Produced in Animals by Injections of Asbestos. J. nntn Cancer fast., 52, 1823. Department of Employment and Productivity (1970) H.M. Factory Inspootorate Standards for Asbestos Dust Concentrations for Use with the Asbestos Regulations, 1969. DEJi1. Technical Data. Note 13. London: H.M. Stationery Office. Doll, R. 11965) Mortality from Lung Cancer in Asbestos Workers. Br. J. ind. Med., 12, 81. Ounmork. J. H.. Hamilton, R. .1. & Smith, D. 8. C. 11964) An. Instrument for the Sampling of Respir able Dust for Subsequent Gravimetric Assessment. J. orient. Instrum., Al, 669. Olbit, C. E. & Holland. W. D. (1962) Use of Elootrioally Exeitod Oxygen for tho Low Tempera ture Decomposition of Organic Substances. Anal. Chem., 34, 1464. .Ross, P. & dk Trevillk, R. T. P. (1967) Experi mental Asbestosis. Arch. Eno. Health, 15, 038. Health and Safety Executive (1976) Asbestos Hygiene StandartIs and Measurement of Airborne 45 Dust Concentrations. HJS.E. Outdance Note. Environmental Hygiene 110. London: H.M. Stationery Office. Holt, P. F., Mills, .1. & Vhiino. D. K. (1905) Experimental Asbcstosis with Four Types of Fibers. Ann. .V. 1'. Acad, Sri.. 132, 87. i.A.R.C. (1973) Report of tho Advisory Committee on Asbestos Cancers to the Director of tho Inter national Agency for Re.-A-arch on Cancer. In Biological Effects of Asbestos, I.A.lt.O. Ncirntijie Publication No. , p. 346. Jacobsen. M.. Rae, S,, Walton, W. H. St Rooan, M. (1970| New Dust Standards for British Coalmines. Nature, 227. 445. Klohterko iter, \V. A Roboi k. K. (1975) New Aspects on Dust and Pneumoconiosis Research. ,4 m. induct. Hyg. Am. 36, 059. McDonald, J. 0. (1973) Cancer m Chrysotile Minos and Mills. Biological effects of asbestos, I,A.Ii.C, Nckntific Publication No. 8, p. 189. Maroudas. N'., O'Neal. C. & .Stanton. M. (1973) Fibroblast Anchorage m Carcinogenesis by Fibres. Lancet, i, 807. Middleton, A. P.._Beckett. S. T. it Davis, .1. M, G. (1977) A Study of tho .Short-term Retention and Clearance of inhaled Asbestos by Rats, using ir.T.C.O. Standard Reference Samples, in Intuited Particle* IV Ed. W. II. Walton. London: Perga- mon Press, p. 247. Morris, T. G., Roberts. W. H., Silverton, R. E * Wagner. J. C. (1967) In Inhaled Particles and Vapours. II. Ed. C. N. DaviOR. London: Porgamon Press, p. 205. National Health and Medical Research Council (1976) Membrane Filter Method for Estimating Airborne Asbestos Dust. Canberra (Australia): N.H.M.R.C. Reeves, A. L., Puro, H. E. & Smith, R. G. (1974) Inhalation Carcinogcnosis from Various Forms of Asbestos. Environ. Res., 8, 178. Reeves. A. L., Pcro. H. E., Smith, K. G. & Vorwald, A. .1. (1971) Experimental Asbestos Carcinogenesis. Environ. Res., A, 496. Seukoff. X. J,,.Hammond. . C. & Seipman, H. (1973) Cancer Risk of Insulation Workers in the Unitod States. In.. Biological Effects of Asbestos. I.A.R.C. Ncieni, Publ. No. 8, p. 21)9. Stanton, M. F., Layard, M., Teoeris, A., Miller, E,, May. M. JSt Kent, E. (1977) Carcinogenicity of Fibrous Glass. J. natn Cancer Inst.. 58, 587. Stell, P. M. & McGill, T. (1973) Asbestos and (Cancer of Head and Neck. Lancet, i, 678. Timbrell, V,, Hyetx, A. W. & Skidmore, J. M. (1968) A Simple Dispenser for Generating Dust Clouds from Standard Relereneo Samples of Asbestos. Ann. of Occupat. Hyg., 11, 273. Timbrell, V., Skidmore. ,r. W.. Hyett. A. W. 4 Wagner, ,f. C. (1970) Exposure Chambers for Inhalation Experiments with Standard Reference Samples of Asbestos of the International Union against Cancer (U.I.C.C.). J. Aerosol Net.. I, 215, Waqnbr, .T. C. (1963) Asbestos in Experimental Animals. Br. J. ind. Med., 20, 1. Wagner, .f, C,, Berry, G., Skidmore, ,1. W. 4 Tthbrell, V. (1974) Tho Effects of tho Inhalation of Asbestos in Rats. Br. J. Dancer, 29, 252. Waoner, J. 0. * Skidmore, J. W. (1965) Asbestos Dust Deposition and Retention in Rats. .Inn, N.Y.Acad.Sci., 132.77. Wagner, J. Slf.qos, C. A. & Marchand. P. <5SS X. DAVIS, S. BECKETT, R. BOLTON, P. COLLINGS AND A. MIDDLETON (1000) Diffuse Pleural Mesothelioraata ami Aibe-sroi? Exposure in the North Western I 'ape Province. Br../, Ind. Med., 17, 260. Wai.tow, \V. H. A Beckett. ,S. T. (1077) A Micro scope Eyepiece Gfuticulo for the Evaluation of Fibrous Dust. Ann. (keupat, Hyg..20. Ziei.hi-ih, R. (,. 11077) f'ublic Health [Hake of F.xpoaure In Aaheatns. Report- of a Workma (,}roup of Experts preparotl for the Commission of the European Communitiea. Directorate general for Social Affairs, Health anti Safety Directorate. Oxfonl: Pcrj(amon Press. fir../. ('finr.fr (I SACCH/ WITH A Fr/mt t/i Summary.--R< dimethylnitros j maintained, in plasma membi renal tumour . lectrophoresis '.isma membi . .d3decreaaec ' is one of the glj , in the tumour Fluorescein upon transfori (neuraminidas in the nuclei oi Dimethylnitr ubiquitous care induce oarcino (Mague, l!)7()). nitrates and n linn uf food \vi 'niiiont can le ' :-<isamines, ] an stomad uldition, tl '.M-i-o-smoke lulinson. 10721 biologically inf 'I nl.. 1977) se qiicucy of hum carcinogen. Alt between DMN meat uf hum stablished, it man alone won if tliis carcin animals arc rt l'7(i). w limn