Toxicological evaluation of some food additives including anticaking agents, antimicrobials, antioxidants, emulsifiers and thickening agents WHO FOOD ADDITIVES SERIES NO. 5 The evaluations contained in this publication were prepared by the Joint FAO/WHO Expert Committee on Food Additives which met in Geneva, 25 June - 4 July 19731 World Health Organization Geneva 1974 1 Seventeenth Report of the Joint FAO/WHO Expert Committee on Food Additives, Wld Hlth Org. techn. Rep. Ser., 1974, No. 539; FAO Nutrition Meetings Report Series, 1974, No. 53. CARRAGEENAN AND FURCELLARAN Explanation These substances have been evaluated for acceptable daily intake by the Joint FAO/WHO Expert Committee on Food Additives (see Annex 1, Ref. No. 20) in 1969. Since the previous evaluation, additional data have become available and are summarized and discussed in the following monograph. The previously published monograph has been expanded and is reproduced in its entirety below. Native carrageenan is a mixture of highly sulfated polygalactosides and is extracted from seaweeds. The detailed structure varies slightly between samples depending on the source but all samples have a sulfate/galactoside ratio of approximately one and a molecular weight of 800 000 to one million. Native carrageenan is used by the food and toiletry industries. Degraded carrageenan is prepared from the extract of Eucheuma spinosum by partial hydrolysis, using dilute HCl, followed by purification. The sulfate/galactoside ratio is the same as in the native form but the molecular weight is only 20 000. This material is not used by food manufacturers but is sold as an antipeptic agent on the continent. It has, however, been used as a model compound in many of the investigations into the mechanism of action of native carrageenan since it produces the same lesion in guinea-pigs as the native form but much more rapidly and consistently. BIOLOGICAL DATA BIOCHEMICAL ASPECTS Absorption Native carrageenan. Rats probably do not absorb carrageenan (2% to 20% in diet) but excrete it quantitatively in their faeces (DHSS, 1972). Guinea-pigs absorb native carrageenan in the caecum and proximal colon and its presence can be demonstrated in the macrophages of the lamina propria by staining with toluidine blue. Absorption is also indicated by the presence in the same sites of Perls' positive granules when iron-labelled native carrageenan is fed for five days. No absorption from the gastrointestinal tract could be demonstrated by these techniques in the rat and mouse (BIBRA, 1971). Degraded carrageenan. Two baboons excreted in the urine 2-6 mg/kg of degraded carrageenan following a single oral or parenteral dose of 3 g/kg. No material could be detected in the blood (DHSS, 1972). Degraded carrageenan is absorbed by the caecum and proximal colon of the guinea-pig and rabbit where it can be demonstrated in the macrophages of the lamina propria by staining with toluidine blue or by newly developed Alcian blue technique which distinguishes degraded carrageenan from mucopolysaccharides. It has also been demonstrated in these tissues by an electrophoretic technique (BIBRA, 1971). Absorption is also indicated by the presence in the same sites in the guinea-pig and rabbit of Perls' positive material when iron-labelled degraded carrageenan is fed. No absorption from the gastrointestinal tract could be demonstrated either histologically or better use of iron-labelled material in the rat, mouse, squirrel monkey, hamster and ferret. Five patients suffering from carcinoma of the transverse colon were given 5 g degraded carrageenan each day for 10 days before partial colectomy. At operation specimens of colon were obtained. Histological sections from all five resections were stained by the toluidine blue and Alcian blue techniques and one specimen was analysed by the electrophoretic technique. No degraded carrageenan could be detected in the tissues either histologically or chemically (the chemical method is sensitive to 0.5 µg/g tissue) (BIBRA, 1971). Rats excrete dietary carrageenan quantitatively in the faeces over levels of 2% to 20% and it therefore has no direct nutritive value (Hawkins & Yaphe, 1965). Groups of five rats received 0.5% native carrageenan, or 5% degraded carrageenan for 10 days. Faecal excretion and weight gain of the two polymers were similar (Dewar & Maddy, 1970). Guinea-pigs failed to excrete in their urine up to 158 mg orally administered undegraded carrageenan. If given intraduodenally to guinea-pigs with ligated pylorus, it reduces gastric secretion. Two adult baboons received single doses of 240 mg/kg (more than three times the therapeutic dose in man) or 3 g/kg of degraded carrageenan while a third baboon received both the high and the low dose levels with a 14-day rest period between doses. Only at the high level were 3-6 mg/kg excreted in the urine after 24 hours and no carrageenans could be detected in the blood up to 18 hours after dosing (Beattie et al., 1970). Groups of six rats were fed diets containing 17.4 or 34.8% carrageenan for three weeks. Weight gain was significantly reduced especially at higher levels. Food efficiency showed interference with utilization of other nutrients in the diet. Only 10% to 15% appeared digestible as estimated from faecal examination (Carey, 1958). Single intravenous injections of degraded carrageenan into nine groups of three, four or five guinea-pigs produced a dose-related increase in urinary polysaccharide excretion. Oral or intraduodenal administration of degraded carrageenan gave detectable residues in the urine and reduced gastric secretion in guinea-pigs with ligated pylorus. However, the amounts absorbed were very small (Anderson & Soman, 1966). Subcutaneously administered degraded carrageenan lowers significantly the volume of fasting gastric secretion in guinea-pigs and tends to reduce the total acid concentration of fasting secretion. When administered parenterally it has a pronounced and prolonged inhibitory effect on histamine - stimulated gastric secretion (Eagleton et al., 1968). The prevention of histamine-induced gastric and duodenal ulceration in the guinea-pig has been demonstrated using a series of undegraded and degraded carrageenans. The degree of antipeptic activity is probably determined by a combination of structural features which includes molecular size and polyanionic properties (Anderson & Soman, 1967; Anderson & Watt, 1959). Artificially produced gastric ulcers (30% acetic acid injection) in rats did not show any significant acceleration in healing when carrageenan was administered concurrently (Takagi et al., 1969). Inhibition of peptic activity was obtained in the Shay rat preparation (Levey & Sheinfeld, 1954). Carrageenan inhibits pepsin action by binding the substrate (Anderson & Watt, 1959; Piper & Fenton, 1961; Houck & Bhayana, 1960 and Bonfils et al., 1966). Interference with pepsin activity depends on concentration of carrageenan and protein. It occurs only over a narrow range (Marquez & Garcia, 1960; Vaughan et al., 1961; Vaughan et al., 1962 and Bonfils et al., 1960). 5% carrageenan in the diet had no effect on utilization of cassein, soybean protein, or of other proteins by rats (Friedman & Douglass, 1960). Carrageenan complexes with mucoproteins in the human stomach with a consequent increase in viscosity and eventual precipitation (Anderson & Watt, 1959 and Marcus & Watt, 1965). No difference in protein economy of the growing rat was observed with a diet of 0.5 to 5% carrageenan and either a high quality protein (casein) or a low quality vegetable protein (USFDA, 1969). Subcutaneously injected carrageenan stimulated the biosynthesis of collagen and the formation of connective tissue (Robertson & Schwartz, 1953). 3% carrageenan in the diet has been reported to reduce the plasma cholesterol level in chicks by 50% (Riccardi & Fahrenbach, 1965). Carrageenan acts as an antithrombic in both human and dog blood plasma (Hawkins & Leonhard, 1962). The presence of carrageenan in infant feeds does not affect the availability of any added iron (Gorton et al., 1963; Gorton & Cross, 1964; Owen & Fomon, 1963; Fomon et al., 1961). A seven-month infant feeding study of the nutritional value of Formil, an infant formula containing carrageenan (concentration not stated), showed normal results in feeding pattern, growth, bowel function, and serum calcium, phosphorus and cholesterol levels (Marine Colloids, Inc., 1969). It has been suggested that the presence of carrageenan in infant food may lead to the selective presence of intestinal organisms capable of forming galactose from carrageenan. This may be important for galactosaemic infants whose food should therefore be free from carageenan (Davis, 1970). In 50% of patients suffering from chronic peptic ulcer and treated with degraded carrageenan, it was noted that carrageenan was still present in gastric juice 7 to 10-1/2 hours after a single dose (Marcus & Watt, 1965). Pharmacodynamic properties Carrageenan is rapidly taken up by mouse peritoneal macrophages in tissue culture with subsequent slow interaction between the engulfed carrageenan, the lysosomal enzymes and the lysosomal cell membrane; the enzymes eventually pass into the cytoplasm of the macrophages which are then destroyed. The action of silica is similar to carrageenan except that the cells more rapidly succumb and finally a fibrous response is seen because the lysosomal cell membranes are quickly destroyed. If no lysosome destruction occurs the foreign material is merely stored in the intact macrophages (Allison et al., 1966). Monocyte infiltration, histiocyte proliferation and macrophage differentiation can be induced in the rat by a single subcutaneous injection of carrageenan. A dual origin of macrophages has been observed: through migration of blood monocytes and by cellular multiplication in situ. The evolution of the granulomatous reaction was followed by the histochemical demonstration of some hydrolytic enzymes of lysosomes (Monis et al., 1968). Intravenous injection of carrageenan into normal rats lowered the bradykininogen levels acutely, and this effect was not prevented by indomethacin. It is considered improbable that bradykinin could be an important mediator of the inflammation caused by carrageenan (Arman & Nuss, 1969). Carrageenan injected intravenously lowers the arterial blood pressure of rats. This hypotensive effect is unaffected by histamine antagonists and is abolished by protease inhibitors and thus seems to be due to kinin release from plasma substrates (Di Rosa & Sorrentino, 1970). Tissue cultures of embryonic rat-kidney fibroblasts exposed to a 1% solution of native or degraded carrageenan for four hours showed metachromatic granules when stained with toluidine blue. These granules persisted for 120 hours following exposure without cytopathological changes or suppression of growth rate (Sharratt et al., 1970). TOXICOLOGICAL STUDIES Special studies on large-bowel ulceration in various species Native carrageenan Mouse Minimal changes comprising focal cellular infiltration of the colonic mucosa were observed in mice treated with 1% carrageenan in drinking water for 30 days to one year (Marcus & Watt, 1969). However, no ulceration was found in the gastrointestinal tract of mice receiving up to 4 g/kg/day of carrageenan (derived from Eucheuma spinosum) for six months. Mice fed on diets containing up to 25% carrageenan for their life-span showed no pathological changes in the gastrointestinal tract (DHSS, 1972). Rat Ulceration of the large bowel followed by scarring was seen in rats treated with 1% carrageenan in drinking water for 30 days to one year (Marcus & Watt, 1969). However, no ulceration of the gastrointestinal tract or other untoward effects were seen in rats receiving dietary carrageenan (derived from Chondrus crispus and Gigartina mamillosa) at a level of up to 12.5 g/kg/day for two years (DHSS, 1972). No ulceration was found in rats fed on diet containing 5% carrageenan (derived from Eucheuma spinosum) for eight weeks (BIBRA, 1971) or on diets supplying up to 4 g/kg/day for periods up to six months (DHSS, 1972). No abnormality was found in rats receiving 1.3 to 1.8 g/kg/day of carrageenan (derived from Chondrus crispus) in drinking water for six months (Golberg, 1971). Guinea-pig When fed on a diet containing 5% native carrageenan (derived from Eucheuma spinosum or Chondrus crispus) they develop ulcers of the caecum within two to four weeks (BIBRA, 1971). After administration as a 1% solution (from Eucheuma spinosum) in drinking water ulceration occurred in eight out of 10 animals in 20-30 days (DHSS, 1972). No ulceration was seen, however, in guinea-pigs which received 1% native carrageenan (from Chondrus crispus) in drinking water for six months (Golberg, 1971). Gerbil No untoward effects were found in gerbils which consumed a diet containing 5% carrageenan (derived from Chondrus crispus) for six months (Golberg, 1971). Dog No untoward effects were found in four dogs which received 1 g/kg/day of carrageenan (derived from Eucheuma cottonii and Chondrus crispus) for six months (DHSS, 1972). Pig No gastrointestinal ulceration was found in pigs fed for four weeks on diets supplying carrageenan (derived from Eucheuma spinosum) at a level of approximately 1.7 g/kg/day (Tournot, 1970). Four groups of six pigs received native carrageenan jelly for 12 weeks at levels of 0, 50, 200 and 500 mg/kg bw per day. Another group received 500 mg dry carrageenan in a powdered diet. No adverse effects were seen in behaviour, weight gain, haematology, blood chemistry, urinalysis, organ weights. No ulcerative colitis or erosions of caecum and colon were seen (Poulson, 1972). Monkey Six rhesus monkeys received 1% carrageenan (derived from Chondrus crispus) in drinking water (1.3 g/kg/day) for a period of seven to 11 weeks. Of two monkeys killed at 11 weeks one had minimal changes in the colon (minute areas of capillary hyperaemia and oedema of mucosa) the relationship of which to carrageenan ingestion is questionable. After an 11-week period without treatment the remaining four received, by gastric intubation, up to 1.25 g/kg/day of carrageenan for 84 days. No abnormalities were observed in these animals and at post mortem examination there were no visible changes in the gastrointestinal tract (Golberg, 1971). Degraded carrageenan Mouse Marcus & Watt (1969) reported minimal changes, consisting of focal cellular infiltration of the colonic mucosa in mice receiving 0.1 to 5% degraded carrageenan in their drinking water for 30 days to one year. Other workers have found no ulcerative lesions in the gastrointestinal tract of large numbers of mice receiving up to 4 g/kg/day for up to six months. At high dosage levels (3 and 4 g/kg/day) Maillet (1970a) found strongly metachromatic material to be present in the Kupffer cells. Rat Ulcerative lesions were seen in the caecal, colonic and rectal mucosa of rats given a 0.1 to 5% solution of degraded carrageenan in drinking water for 30 days to one year. The early changes included pin-head sized haemorrhagic and ulcerative lesions, along with cellular infiltrates and some crypt abscesses. In later stages there was loss of mucosal folds and, in some animals, there was scarring of the bowel which led to the formation of constriction rings (Marcus & Watt, 1969). Guinea-pig Many workers have shown that degraded carrageenan causes ulceration of the large bowel when fed to guinea-pigs (DHSS, 1972). While Marcus & Watt report ulceration of the colon and rectum as well as the caecum, other workers have found ulceration only in the caecum and proximal colon (BIBRA, 1971; Golberg, 1971; Maillet, 1970). Two forms of ulceration have been described. In the first, the "acute" lesion, with onset of illness appeared in a few days. Severe diarrhoea and weight loss occurred within two days and in many animals ulceration appeared in four to six days. The ulcers were large, often 5 mm diameter and had the appearance of stripped-off epithelium. The subepithelial tissues were oedematous, markedly so in some cases. While accumulation of macrophages was often seen in the areas of ulceration, this was not always so and frequently the ulcer had the appearance of being the result of mechanical stripping. This type of ulceration occurred only when there was profuse diarrhoea, that is when degraded carrageenan was administered as a 5% (and in some cases 2%) solution in drinking water. It appeared that the lesion was related more to the osmotic diarrhoea than to carrageenan per se since exactly the same clinical, macroscopic and microscopic picture was found in guinea-pigs given equiosmolar sodium sulfate or cyclamate solutions in place of drinking water (BIBRA, 1971). The second "chronic" form of ulceration was found in animals receiving 1% degraded carrageenan in drinking water (or native carrageenan at any concentration). This ulceration occurred more slowly, the first signs appearing after about two weeks. The ulcers were pin-point in size and star-shaped. The stages of development of this type of lesion have been investigated (BIBRA, 1971). Hamster No gastrointestinal ulceration was seen in hamsters fed on a diet containing 5% degraded carrageenan for five months (BIBRA, 1971). Rabbit Three out of five rabbits receiving as little as 0.07 mg/kg/day of degraded carrageenan in drinking-water developed ulceration of the large bowel in three months. The lesions were said to occur in the caecum, colon and rectum. Rabbits weighing 3.5 kg were given 1.6 g/kg/day of degraded carrageenan by gastric intubation for four weeks. No ulceration was found but granulomas consisting of macrophages were seen in the lamina propria of the caecum. No abnormalities occurred in the distal colon or rectum. In rabbits given iron-labelled degraded carrageenan macrophage accumulation occurred together with ulceration of the caecum but again no abnormalities were found other than in the caecum and proximal colon (BIBRA, 1971). Gerbil No abnormalities were found in gerbils consuming 3.5 g/kg/day of degraded carrageenan for six months (Golberg, 1971). Ferret No ulceration was seen in ferrets fed on a diet providing 1.5 g/kg/day of degraded carrageenan for four weeks (BIBRA, 1971). Pig No signs of ulcerative lesions were found in the large bowel in 10 pigs which received 1.7 g/kg/day of degraded carrageenan for four weeks (DHSS, 1972). Squirrel monkey No ulceration of the bowel was seen in squirrel monkeys which received 1.5 g/kg/day of degraded carrageenan as a 10% solution by gastric intubation for four weeks (BIBRA, 1971). Man Patients under treatment for peptic ulceration and ulcerative colitis have been treated with degraded carrageenan without ill effects (DHSS, 1972). In addition, in France, about 50 000 patients are under treatment with 5 g/day of a mixture of degraded carrageenan and aluminium hydroxide. It is suggested that if degraded carrageenan contributed to the development of ulcerative colitis it might be expected to show up in a population of this size under medical care. Although treatment has been used for 10 years there is, so far, no indication that French peptic ulcer patients are developing ulcerative colitis (Shirlaw, 1971). Administration of 100 ml 5% degraded carrageenan as retention enema two to three times per week for three weeks to several months to patients with ulcerative colitis or Crohn's disease improved four out of five patients as seen on X-ray and sigmoidoscopy (Bonfils, 1972). Histological appearance of ulcers Guinea-pig Ulcers involve mainly the mucosa and show features of both acute and subacute inflammatory infiltration as well as crypt abscesses (Watt & Marcus, 1969a, b). Mottet (1970) who examined their material considered the ulcers to be in many ways morphologically comparable to those of human ulcerative colitis. However, he noted that epithelial necrosis involved all regions of the crypt and not the depth of the crypt as in the human disease. Alterations were noted in the amount of mucus in nearby glands and lymphocytic infiltration and capillary congestion were prominent in the lamina propria as in ulcerative colitis. A typical epithelial hyperplasia was also noted. When neomycin was included in the diet of guinea-pigs the carrageenan-induced lesions had a different appearance, the infiltrate consisting mainly of macrophages (confirmed by their strong acid phosphatase activity and by their appearance under EM) with only a few lymphocytes, plasma cells and polymorphs. The incidence and time of development of lesions was similar in neomycin-treated and untreated animals. No crypt abscesses were seen in neomycin-treated animals. Two small crypt abscesses were found in carrageenan-treated animals not receiving the antibiotic (BIBRA, 1971). Golberg (1971) found the lesion corresponded to that described by Marcus & Watt, but without crypt abscesses or epithelial hyperplasia; no antibiotic was included in the animals' diet. Rabbit Ulceration of the large bowel was described by Watt & Marcus (1970a, b). Gross thickening of the mucosa of the caecum and colon was seen with small sessile polypoid formations on the summit of rugi of the caecum. Ulcers were much as they described in the guinea-pig but glandular polyps and pseudopolyps (oedematous congested mucosa with acute inflammatory infiltration) were prominent features. Ulceration which occurred in the caecum of rabbits given degraded carrageenan by gastric intubation together with dietary neomycin developed ulcers which were in all respects similar to that seen in guinea-pigs similarly treated. Neither polyp formation nor gross glandular hyperplasia was seen (BIBRA, 1971). Development of the ulcer in guinea-pigs A study was carried out in which guinea-pigs were killed and the whole gastrointestinal tract removed and examined macroscopically and histologically after periods of one to 31 days on a dietary regime containing 0.1% neomycin and drinking fluid containing 1% degraded carrageenan. Histological abnormalities were found only in the caecum and proximal colon where the first change was an increase in the number of macrophages in the lamina propria after about 14 days' treatment. In sections stained with toluidine blue these macrophages contained metachromatic material. Alcian blue stained sections showed that this was likely to be carrageenan. These cells were more strongly acid phosphatase-positive than normal macrophages and the lysosomes were larger. Also, at this time, lysosomes, which are normally small and sparse in caecal epithelial cells, became larger, more numerous and strongly staining. The epithelial cells most strongly acid phosphatase-positive were superficial to the most marked accumulations of macrophages. The macrophages continued to accumulate and to form granulomas in the lamina propria. Frequently degenerating macrophages were seen in these granulomas (confirmed by EM). Ulceration was found later in those areas where the granulomas occurred. The granulomas could be seen on gross examination as pale, raised areas under the epithelium. Occasionally these pale areas developed into white plaques on which large numbers of pin-point ulcers occurred. Histological sections were prepared which stained degenerating epithelium overlying a mass of degenerating macrophages superficial to the granuloma. The authors consider it likely that ulceration is a consequence of the granuloma formation which follows carrageenan absorption with the acute inflammatory infiltrate being possibly due to subsequent bacterial infection of the ulcerated granuloma. It did not appear that the macrophage infiltrate was a consequence of ulceration caused by some direct action of carrageenan on the gastrointestinal mucosa; direct contact for four hours with 5% degraded carrageenan had no untoward effect on cells of the stomach, intestine or large bowel in rats or guinea-pigs. The authors point out that the pronounced macrophage response, the limitation of the lesion to the caecum and the absence of epithelial hyperplasia and crypt abscesses (in all neomycin-treated animals and many neomycin-free animals) are features which differ from lesion of ulcerative colitis in man (BIBRA, 1971). Reversibility of the lesion in guinea-pigs Maillet (1970) found that the ulcerative lesion in the guinea-pig caecum was reversible in eight to 12 weeks when animals were given normal diet. This was confirmed by a test in which animals were administered 1% degraded carrageenan in drinking water for three weeks. All animals killed at this time had caecal ulcers. Surviving animals were given normal diet and killed after one, four, eight and 12 weeks and examined. Histological sections of the caecum were prepared. Lesions seen were small ulcers with the base infiltrated chiefly with macrophages but also a few mono-nuclear cells and polymorphs, the margins consisting of a dense infiltrate of macrophages with few other cells. The majority of macrophages were grossly vacuolated and others contained one or more inclusions, probably necrotic debris. The lysosomes (acid phosphatase positive granules) were larger and more numerous than in untreated animals both in the macrophages and the overlying epithelial cells. No ulceration was seen in animals killed after replacement of 1% degraded carrageenan by drinking water. In three of four animals killed one and four weeks after cessation of treatment the number of macrophages in the lamina propria was decreased although a few granulomas were observed. Animals killed later were histologically normal. Histochemical examination of specimens from animals killed after one, four and eight weeks still showed considerable acid phosphatase activity in epithelial cells and macrophages. In all four animals killed 12 weeks after ceasing treatment no differences from controls could be found (BIBRA, 1971). Reproduction and teratology studies Mouse Groups each of 25-40 pregnant mice were dosed orally with either sodium or calcium carrageenate as a 25% suspension in anhydrous corn oil, from day 6 through day 15 of pregnancy. The dose levels administered were 0, 10, 45, 470 and 900 mg/kg/day. Animals were observed daily for appearance and behaviour. On day 17, dams were subjected to Caesarean section, and the numbers of implantation sites, resorption sites, and live and dead fetuses determined. Body weight of live pups were measured. Fetuses were examined grossly for the presence of external congenital abnormalities, and then one-third of the fetuses underwent detailed visceral examination, and the remainder were examined for skeletal defects. Administration of either sodium or calcium carrageenate resulted in an apparent increase in the number of resorptions and/or fetal deaths in utero. There was a dose-dependent decrease in the number of live pups and reduction in pup weight at delivery. In addition, there was a concurrent retardation in skeletal maturation as shown by the increased incidence of missing sternebrae and incomplete skull closures. All other soft tissue and skeletal abnormalities appeared to be within normal range (Food & Drug Research Lab., Inc., 1972). Rat Nine pregnant rats were fed 5% degraded carrageenan in their drinking water and their offspring received the same material for six months. Five out of 12 offspring had bronchiectasis, four out of 12 had ulceration of the colon (Watt & Marcus, 1971; Marcus & Watt, 1971). Groups each of 24-29 pregnant rats were dosed orally with either sodium or calcium carrageenate as a 25% suspension in corn oil, from day 6 to 15 of gestation. The dose levels administered were 0, 40, 100, 240 and 600 mg/kg/day. On day 20 all dams were subjected to Caesarean section, and the parameters described in mouse study were measured. Administration of either sodium or calcium carrageenate resulted in an apparent increase in the number of resorption sites without a decrease in the number of live pups delivered. At the highest dose level there appeared to be a decrease in birth weight of the pups. There was a dose-dependent increase in missing skeletal sternebrae, but no other compound-related abnormalities (Food & Drug Research Lab., Inc., 1972). Hamster Groups each of 26-30 pregnant hamsters were dosed orally with either sodium or calcium carrageenate as a 25% suspension in anhydrous corn oil, from day 6 to 10 of gestation. The dose levels administered were 0, 40, 100, 240 and 600 mg/kg bw per day. At day 14 all animals were subjected to Caesarean section, and the parameters described in mouse study were measured. Administration of either sodium or calcium carrageenate had no significant effect on nidation or on maternal or fetal survival. There was some evidence of delayed skeletal maturation (missing or incomplete centres of ossification) which was dose- dependent. No other compound-related effects were observed.(Food & Drug Research Lab., Inc., 1972). Rabbit Groups each of 15 pregnant rabbits were dosed orally with either sodium or calcium carrageenate as a 25% suspension in anhydrous corn oil, from day 6 through day 18 of gestation. The dose levels administered were 10, 60, and 260 mg/kg bw per day. On day 29 all does were subjected to Caesarean section, and the parameters described in the mouse study were measured. Administration of either sodium or calcium carrageenate had no effect on nidation or on maternal or fetal survival, nor were there any abnormalities in either soft or skeletal tissues that were compound-related (Food & Drug Research Lab., Inc., 1972). Acute toxicity Dosage Measurement Animal Sex Substance mg/kg (1G)* Reference Rabbit M Na+ Carr. 4670-5430 LD50 Food & Drug Res. Lab. , 1972 Rabbit F Na+ Carr. 4670-5430 LD50 " " " " Hamster M Na+ Carr. 7530-8470 LD50 " " " " Hamster F Na+ Carr. 7530-8470 LD50 " " " " Rat M Na+ Carr. 5650-6250 LD50 " " " " Rat F Na+ Carr. 5650-6250 LD50 " " " " Mouse M Na+ Carr. 8730-9670 LD50 " " " " Mouse F Na+ Carr. 8730-9670 LD50 " " " " Rabbit M Ca++ Carr. 2280-3000 LD50 " " " " Rabbit F Ca++ Carr. 2280-3000 LD50 " " " " Hamster M Ca++ Carr. 6180-7320 LD50 " " " " Hamster F Ca++ Carr. 6180-7320 LD50 " " " " Rat M Ca++ Carr. 5140-5660 LD50 " " " " Rat F Ca++ Carr. 5140-5660 LD50 " " " " Mouse M Ca++ Carr. 8710-9590 LD50 " " " " Mouse F Ca++ Carr. 8710-9590 LD50 " " " " * Compounds administered intragastrically as a 25% suspension in anhydrous corn oil. Rat Ten male rats received 12 g/kg/day of degraded carrageenan for three days without mortality or visible adverse effect (Bonfils, 1959). Rats survived 10 mg carrageenan intravenously, but showed histopathological changes in the kidney (Morard et al., 1964). Six groups of five male and five female rats received 0, 3.6, 7.2, 14.4, 21.6 or 28.8 g/kg/day carrageenan for five days. There was no mortality or alterations in body weight gain. All test groups exhibited diarrhoea with grey coloured stools during the dosing period. At the highest dose level two rats had cloudy eyes and one had blood in the urine (Rorer, 1966). A group of 10 male and 10 female rats received a total of 23 g/kg carrageenan over a five-day period with five male and five female rats as controls. There were no deaths during the study (Rorer, 1966). Guinea-pig Guinea-pigs survived 5-10 mg carrageenan intravenously but showed histopathological changes in the kidney (Morard et al., 1964). Intravenous carrageenan in doses of 1 mg/kg kills guinea-pigs within 30 minutes (Anderson & Soman, 1966). Rabbit Intravenous injection of undegraded carrageenan is very toxic to rabbits (Anderson & Duncan, 1965). Intravenous injection of 50 mg carrageenan killed rabbits within 48 hours. These animals showed total obstruction of the glomerular capillaries by a fibrinoid-like substance, and extensive tubular necrosis. Rabbits survived 10 mg intravenously, but showed histopathologic changes in the kidney. (Morard et al., 1964). Undegraded carrageenan given intravenously to rabbits (50 mg) resulted in death within 48 hours associated with diffuse renal cortical necrosis and widespread capillary thromboses in various organs. Plasma fibrinogen disappeared and serum complement decreased; the effects were dose related to some extent (Morard et al., 1964; Anderson, 1967). In four samples of C. crispus carrageenans harvested from four locations, the lowest intravenous doses required to kill rabbits within 24 hours varied from 1 to 20 mg/kg for lambda carrageenans to 1-5 mg/kg for kappa carrageenans (Duncan, 1965). The injection of lambda carrageenan induced collagenous growth in the rabbit cornea (Dass & McCandless, 1966). Dog Five male and one female mongrel dogs received graded amounts of carrageenan orally for seven days. The drug level was increased daily until a level of 3 g/kg/day was reached on day 3. This dose was then continued from day 4 to day 7. Each animal was sacrificed 24 hours after the last dose. Diarrhoea and occasional bloody stools were the only clinical findings and this occurred in two dogs on day 8. Gross necropsy revealed a varying degree of gastrointestinal hyperaemia and also an ischaemic spleen in one dog and a haemorrhagic lung in one dog. Body weight loss was observed in five dogs. There were some superficial erosions of the mucosa without extensive inflammation (Rorer, 1966). Three groups of one male and one female mongrel dogs received 750, 1500 or 3000 mg/kg carrageenan for seven days. Diarrhoea was observed in four dogs. Gross necropsy revealed a varying degree of gastrointestinal hyperaemia with extensive haemorrhages in one dog which also showed a gradual decrease in haematocrit during the two weeks drug and observation period. Histopathological examination revealed a consistent clubbing of the free edge of the villi combined with focal congestion and erosion of the lining mucosa (Rorer, 1966). Seven groups of two mongrel dogs (male and female) received 0, 95, 190, 375, 750, 1500 or 3000 mg/kg carrageenan for seven days. Diarrhoea and loss of weight occurred in 11 of the test dogs, emesis occurred in one dog at 3000 mg/kg and one dog at 1500 mg/kg. Gross necropsy revealed a varying degree of gastrointestinal hyperaemia with no other significant abnormal tissue. Histopathological examination of the tissues revealed no evidence of drug induced lesions (Rorer, 1966). Eight groups of two mongrel dogs (male and female) received 0, 250, 500, 640, 1000, 1280, 2560 mg/kg carrageenan for seven days. All animals had diarrhoea throughout the dosing period. Gross necropsy revealed a varying degree of gastrointestinal congestion. Haematology and clinical chemistry studies were negative as was the histopathological examination of tissues (Rorer, 1966). Two groups of two mongrel dogs (male and female) were given 1500 or 3000 mg/kg carrageenan for seven days. Diarrhoea and melena was observed at the 3000 mg/kg dose level. Gross necropsy revealed congestion of stomach, duodenum and jejunum. Histopathological examination did not reveal any drug induced lesions. Body weight loss was observed in all surviving animals (Rorer, 1966). Monkey Three groups of two rhesus monkeys received 150, 750 or 3000 mg/kg/day carrageenan for seven days. Haematological, blood biochemical and urological parameters were not altered. There was no significant effect on body weight or organ weights. Gross necropsy revealed the presence of numerous petechial haemorrhages scattered throughout the length of the colonic mucosa in the animals given 750 or 3000 mg/kg carrageenan. In one of the latter animals the mucosa was definitely haemorrhagic. Although these lesions resembled those produced by systemic viral infection, no evidence of septicaemia was found (Rorer, 1966). Short-term studies Mouse Mice given degraded carrageenan in their drinking water in concentrations of 0.1-5% daily from 30 days to one year, showed no significant weight loss. Some changes were seen in the mucosa of the caecum, colon or rectum resembling those found in ulcerative colitis in man. The early changes included pin-head sized haemorrhagic and ulcerative lesions, along with cellular infiltrates and crypt abscesses. In later stages there was loss of mucosal folds, and in some animals there was scarring of the bowel which led to formation of constriction rings. The changes were minimal in comparison with other species and comprised mainly focal cellular infiltrations of the colonic mucosa (Marcus & Watt, 1969). Mice were given native or degraded carrageenan for six months at doses ranging from 70 mg to 4 g/kg/day. There was no evidence of any pathological changes (Maillet et al., 1970). Seven groups of 45 mice received 0, 0.4, 1.0, 2.0, 3.0 or 4.0 g/kg/day of degraded carrageenan for six months. The higher mortality at 2, 3 and 4 g/kg was due to gavage. No change was noted in the duodenum mucosa of stomach, colon, small gut or caecum. Histological examination of the liver showed no change in the hepatic cells. At 3 and 4 g/kg/day the Kupffer cells showed presence of a strongly metachromatic material (Maillet, 1970a). Rat Groups of six male rats were fed for a 10-week period 0, 5, 10 and 20% of a refined preparation. The animals grew well except at the highest test level which showed a 50% mortality (Nilson & Schaller, 1941). A group of four male and three female rats received 1 g/kg/day of degraded carrageenan for 40 days without mortality or visible adverse effect. No lesions were apparent at necropsy and pathological examination revealed nothing abnormal (Bonfils, 1959). In feeding experiments using 10 rats there was some growth retardation, decreased food intake and reduced urinary nitrogen excretion at dietary levels over 10% carrageenan (Hawkins & Yaphe, 1965). Degraded carrageenan was added to the drinking water of rats in concentrations of 0.1-5% daily from 30 days to one year. No significant weight loss was observed. There were changes in the mucosa of the caecum, colon or rectum resembling those found in ulcerative colitis in man. The early changes included pin-head sized haemorrhagic and ulcerative lesions, along with cellular infiltrates and crypt abscesses. In later stages there was loss of mucosal folds and in some animals there was scarring of the bowel which led to formation of construction rings (Marcus & Watt, 1969). Six groups of 10 male and 10 female rats received 0, 70, 350, 700, 1000, 2000 mg of degraded carrageenan by gavage for one month. No adverse effects were seen on body weight, haematology or mortality. Some vasodilatation was seen in the colon and caecum of the test groups. Another four groups of 10 male and 10 female rats received 0, 70, 350 or 700 mg of native carrageenan for one month. No adverse macroscopical or microscopical effects were noted. Another 150 rats received 2 g/kg/day of degraded carrageenan for one, two, three or six months at which times at least 20 were sacrificed and examined. No adverse macroscopical or microscopical effects were noted. Three further groups acted as controls (Dubrasquet et al., 1970). Five groups of 10 male and 10 female rats received degraded carrageenan 0, 1.0, 2.0, 3.0 or 3.9 g/kg/day for six months. No significant difference was noted between the test and control groups as regards mortality, weight gain, food intake, haematology, urinalysis, SGPH, alkaline phosphatase or organ weights (Saito & Tokunaga, 1970). Two groups of five male and five female rats received a 10% solution of degraded carrageenan as drinking water resulting in consumption of up to 16.5 g/kg/day for one month. Moderate congestion of the caecal mucosa was noted but there was no evidence of erosions (Maillet et al., 1970; Maillet, 1970c). Two hundred rats were given for 28 days native or degraded carrageenan at doses ranging from 70 mg to 4 g/kg/day. No histopathological or other changes were noted. Some rats received 4 g/kg for six months without adverse effect (Maillet et al., 1970). Six groups of 10 male and 10 female rats received 70, 350 or 700 mg/kg/day of native carrageenan or the same quantities of degraded carrageenan for one month. A seventh group acted as controls. No appreciable difference in weight gain was noted between the different groups. In the two groups on 700 mg/kg moderate vasodilatation of the mucosa was seen but no histological changes (Maillet, 1970b). Guinea-pig Fifty male guinea-pigs were injected daily for one to 41 days with 0.25 mg histamine/100 g body weight in order to induce gastric ulceration. 50% carrageenan in the diet did not accentuate the mucosal lesions (Holzmann & Schott, 1963). Four groups of 10-15 guinea-pigs received 0, 2 or 4 g/kg/day degraded carrageenan in their drinking water or by gavage. All test groups developed diarrhoea and positive occult blood in the faeces. The chorion was infiltrated with macrocytes and other mononuclear cells, and at the highest dose level some superficial ulceration was noted (Maillet, 1969). Groups of 10 guinea-pigs received for 20-30 days 0 or 1% undegraded carrageenan or 5% degraded carrageenan. The overall incidence of ulceration in the three groups were 0, 80% and 100%. In both experimental groups the lesions were multiple, numbering sometimes in the hundreds. Macroscopically, they consisted of pinhead sized ulcers. Some had coalesced to form larger lesions, often linear in shape. Histologically, the ulcerations involved mainly the mucosa and showed features of both acute and subacute inflammatory infiltration, as well as crypt abscesses (Watt & Marcus, 1969a). Groups of 15 guinea-pigs received for five months 0 and 1% undegraded carrageenan. The animals appeared healthy and gained weight satisfactorily. In the test group all animals developed multiple focal ulcers in the caecum and in four the lesions had extended distally beyond the caecum for about 3 cm. Microscopically the ulcers involved mainly the mucosa and showed subacute or chronic inflammatory reaction (Watt & Marcus, 1969b). Samples of the lesions in guinea-pigs noted by Marcus & Watt were examined by Mottet who considered them to be in many ways morphologically comparable to those of human ulcerative colitis. However, he noted that the epithelial necrosis involved all regions of the crypt and not principally the depth of the crypt as is usually seen in the human disease. Alterations were noted in the amount of mucus in nearby glands. Lymphocytic infiltration and capillary congestion were noted to be very prominent features of the lamina propria of the mucosa and the superficial portion of the submucosa, as in ulcerative colitis. A typical epithelial hyperplasia was noted in the vicinity of the ulceration as occurs in ulcerative colitis (Mottet, 1970). However, when dietary neomycin was given advanced lesions were not seen; there were very few polymorphs while the macrophage and lymphocyte responses and the degree of ulceration were not affected. Other processes such as osmotic effects may have been responsible for the early lesions noted. It was therefore doubtful whether the carrageenan in fact simulated ulcerative colitis (Sharratt et al., 1971). Guinea-pigs which consumed between 4 and 5 g of degraded carrageenan daily showed mucosal erosions, but no ulceration in their caeca. In rare instances these erosions extended into the colon. The ulcers were quite different from those of ulcerative colitis in man. It was noted that the drinking solutions were heavily contaminated with microorganisms under the conditions of the experiment (Maillet et al., 1970). Guinea-pigs administered 5% native carrageenan in the diet or 1% degraded carrageenan in drinking water for two to four weeks produced multiple caecal ulcerations. Sequential studies have shown that the lesions first develop as an accumulation of macrophages in the lamina propria and subsequently in the sub-mucosa leading to the formation of pale raised areas which can easily be seen macroscopically. Ulceration of the mucosa then occurs, particularly in these areas. The ulcers are small and superficial and affect only the mucosa. A mixed cellular infiltrate, consisting predominantly of macrophages but accompanied by polymorphs, lymphocytes, and plasma cells, surrounds the ulcerated area. If animals are treated with neomycin (0.1% neomycin sulfate in the diet) in addition to carrageenan, the infiltrate consists only of macrophages (Sharratt et al., 1970). When carrageenan (native or degraded), labelled with ferric iron (stable at pH 2-8) was added to the diet of guinea-pigs for one week, particulate Perls' positive granule appeared within the epithelial cells and within macrophages which normally populate the subepithelial layers in this species. No Perls' positive material was found in the ileum and colon in these animals. Administration of equivalent amounts of ferric chloride failed to produce any Perls' positive material (Clarke & Hardy, 1970). Rabbit Degraded carrageenan was added to the drinking water of rabbits in concentrations of 0.1% to 5% daily from 30 days to one year. Loss of weight was noted and in some rabbits there was blood and mucus in the stools. In concentrations as low as 1% there were changes in the mucosa of the caecum, colon or rectum resembling those found in ulcerative colitis in man. The early changes included pin-head sized haemorrhagic and ulcerative lesions, along with cellular infiltrates and crypt abscesses. In later stages there was loss of mucosal folds, and in some animals there was scarring of the bowel which led to formation of constriction rings (Marcus & Watt, 1969). Groups of five rabbits received in their drinking water 0, 0.1%, 1.0% or 5.0% degraded carrageenan for 12, 12, seven and six weeks respectively. In the group on 5% carrageenan, diarrhoea was associated with visible and occult blood in the faeces by the end of seven days and this persisted. The animals lost weight and there was severe ulceration of the colon. In the group on 1% carrageenan only one animal developed diarrhoea and occult blood was present in the faeces in all the animals after two weeks. There was some weight loss and all the animals in this group showed ulceration of the colon of moderate severity. In the group on 0.1% carrageenan, diarrhoea did not occur but occult blood in the faeces was demonstrable in three of the animals by the end of 10 weeks. One animal lost weight and the remainder gained weight. Multiple ulcers in the colon were found in three of the five rabbits. In the control group there was no diarrhoea or occult blood in the faeces at any time and no ulcerative lesions were seen (Watt & Marcus; 1970a, b). Dog Groups of two male and two female cross labradors received either 0 or 1000 mg/kg/day native carrageenan in their diet over 22 weeks. No significant adverse effects were noted regarding general appearance, growth, food intake, urinalysis, faecal occult blood (negative throughout), haematology, organ weight or histology of all major organs. The intestinal tract showed no evidence of any early or healed ulceration (BIBRA, 1970). Pig Three groups of five male and five female pigs received 40-56 g daily of either degraded or native carrageenan for four weeks. A fourth group acted as controls. The animals were sacrificed under general anaesthesia. There was no sign of any ulcerative lesions in the areas of the colon and caecum. There was a modification of the intestinal flora, but without reaching pathological limits. This modification of the flora induced a lymphoid reaction at the level of the submucosa of the large intestine (Tournot & Labie, 1970). Groups each of six Danish Landrace pigs (equally divided by sex), 61-75 days of age, were maintained on diets to which carrageenan was added as a jelly at dose levels equivalent to 50, 200 or 500 mg/kg/day or 500 mg/kg/day carrageenan and sucrose 1 g/kg/day. The control group was given diets containing a corresponding amount of sucrose. The test period was 83 days. No effect was observed on growth rate or on feed utilization. Blood and urine parameters showed no significant changes from control and were within normal range. At autopsy, there were no changes in relative organ weight. Histopathologic studies were made of principal organs and tissues. No ulceration or erosions were seen in the gastrointestinal tract. Three pigs, one in the 200 mg/kg/day group, and two in the 500 mg/kg/day group, showed slight changes in the mucous membrane of the large intestine, including a few focal areas with irregular surface and some shift in cellular infiltration pattern. Bacteriological examination of the flora of the jejunum, caecum, colon and ampoulla of the rectum from all animals, showed that treatment results in changes in intestinal flora. The total counts of aerobic bacteria were decreased in the colon and rectum and the number of lacto-bacilli was reduced in the rectum (Poulsen, 1973). Long-term studies Mouse Groups of five male and five female mice of two strains were maintained over their life span on diets containing 0.1%, 5%, 15% and 25% carrageenan without apparent ill-effects on growth, survival, gross and histopathology of the gastrointestinal tract, liver and kidneys (Nilson & Wagner, 1959). Rat Groups of five mate and five female rats of two strains were fed 0, 1%, 5%, 15% and 25% carrageenan for their life span. Food consumption increased in proportion to the increased carrageenan content. No effect was noted on mortality. At the 25% level there was evidence of hepatic cirrhosis but all other levels showed no histopathological abnormalities of the intestinal tract, liver or kidneys (Nilson & Wagner, 1959). Three groups of rats were injected subcutaneously with carrageenan alone or with a promoter. Sarcomata had developed at the site of injection independently of the presence of a promoter in 12 out of 88 rats by the 650th day. Dog In the course of some pharmacological studies in dogs carrageenan was given to dogs for prolonged periods. No deleterious effects were noted and one bitch gave birth to healthy puppies which appeared to tolerate carrageenan without trouble (Shirlaw, G., 1971). OBSERVATIONS IN MAN In the nutritional study of tocopherol requirements an infant formula (Similac) containing carrageenan was fed as the basic diet for six months to 44 healthy premature infants (Goldbloom, 1963). Twenty-four patients under treatment with various anti- inflammatory drugs and who had gastroduodenal ulcers were prescribed degraded carrageenan. Haemorrhages then ceased and the radiological and clinical picture improved (Busson et al., 1964). Ninety-five ulcer patients were treated with 0.15 g aluminium hydroxide and 0.5 g degraded carrageenan and a further 15 ulcer patients received degraded carrageenan and an anticholinergic drug. The patients received 6 g carrageenan for the first three weeks, 3 g from the third to the sixth week and 1.5 g from the sixth to the eighth week. The condition of all these patients except for 10 improved at a similar rate to others treated by the classical methods (Roches, 1965). Two blind trials of degraded carrageenan (6 g/day) and of aluminium hydroxide showed both to be of similar efficacy in giving symptomatic relief to peptic ulcer patients. No adverse effects were noted (Evans et al., 1965). Two hundred peptic ulcer patients were treated with degraded carrageenan 5 g/day. Patients were followed up for an average of two years. Radiological controls were performed every two to four months. No ulcerative colitis or other intestinal side effect was seen (Bonfils, 1970). Ten patients suffering from ulcerative colitis were given 10 g a day of degraded carrageenan (twice the clinical dose) for 10 days. Regular sigmoidoscopic examinations showed no modification either of the colonic ulcerations or of the bleeding of the mucosa (Bonfils, 1970). The incidence of ulcerative colitis in the Oxford area diagnosed per 100 000 population over the decade 1951-1960 has risen apparently from about five to about 10. It was concluded that part at least of this reflected a true increase in the incidence of the disease although other factors may have contributed. The disease rarely manifested itself in children or adolescents; the incidence in both sexes then rose steeply with age to a peak in early middle life, falling again in the sixth decade (Evans & Acheson, 1965). The incidence of ulcerative colitis in Norway was studied for the three five-year periods 1946-1960. A total of 407 hospital cases were reported of which 362 were typical ulcerative colitis. The average annual rate for the three respective five-year periods starting in 1946 and ending in 1960 were 10.5, 17.4 and 20.3 per million population, thus showing a definite rise in incidence in each five- year period. The increase was most marked for the age-groups 0-9 and 10-19 years. The rate of reported cases was found to vary with geographical areas (Gjone & Myren, 1964). Comments: The native carrageenan used as food additive has a molecular weight of about 1 million and the degree of dispersion must be specified so as to reduce the presence of low molecular weight fractions to a minimum. The high molecular weight material is probably not absorbed and not associated with the production of colonic lesions. The weight of evidence points to a lack of effect of high- molecular carrageenan on any of the many species studied. Hence the colonic lesions and other pathological changes elicited with a particular form of degraded carrageenan are not relevant for purposes of evaluation. The critical criteria distinguishing between the food additive and degraded forms of carrageenan may be summarized as follows. 1. Physical parameters: intrinsic viscosity, separation by agarose gel electrophoresis and other procedures. 2. Biological evidence of uptake and storage within cells of the reticuloendothelial system in the case of degraded carrageenan and lack of such evidence with food-grade carrageenan. While some degradation of high-molecular carrageenan does appear to take place during processing of acid foods at relatively high temperatures, the extent of such degradation seems too limited to elicit the formation of low-molecular products possessing ulcerogenic potential in animals or man. Furcellaran is derived from other varieties of seaweed and has not been tested as such. However it belongs chemically to the general range of carrageenan and carrageenan-like substances. The reproductive effects elicited in the mouse and rat by gavage of calcium carrageenan as a suspension in oil cannot be interpreted in relation to food additive use in man. EVALUATION Level causing no toxicological effect Rat: 150 000 ppm (15%) in the diet equivalent to 7500 mg/kg bw. Estimate of acceptable daily intake for man 0-75 mg/kg bw.* * As carrageenan or furcellaran or the sum of both. REFERENCES Allison, A. C., Harrington, J. S. & Birbeck, M. (1966) J. exp. Med 124, 141 Anderson W. (1967) Canad. J. pharm. Sci., 2, 4, 81 Anderson W. & Duncan, J. G. C. (1965) J. Pharm. (Lond.), 17, 647 Anderson W. & Soman, P. D. (1966) J. Pharm. (Lond.), 18, 827 Anderson W. & Soman, P. D. (1967) J. Pharm. Pharmacol., 19, 520 Anderson W. & Watt, J. (1959) J. Pharm. (Lond.), 11, 318 Arman, C. G. Van & Nuss, G. W. (1969) J. Path., 99, 245 Beattie, I. A. et al. (1970) Fd. Cosmet. Toxicol., 8, 257 Bonfils, S. (1959) Unpublished report, 27 May, Hôpital Trousseau Bonfils, S. (1970) Lancet, 2, 414 and paper read at R.S.M. on 27 January Bonfils, S. (1972) Report to Glaxo dated 12 January 1972 Bonfils, S., Dubrasquet, M. & Accary, J. P. (1966) Overdruk uit Tijdschrift voor Gastro-Enterologie, 9, 1, 7 Bonfils, S. Dubrasquet, M. & Lambling, A. (1960) Revue Française d'Etudes Cliniques et Biologiques, 5, 1, 71 BIBRA (1971) Personal communication Busson, A. et al. (1964) La Presse Medicale, 72, 2651 Carey, P. L. (1958) Thesis submitted To Purdue University Clarke, R. M. & Hardy, R. N. (1970) J. Physiol. (Lond.), 209, 669 Dass, B. & McCandless, E. L. (1966) Fed. Proc., 25, 706 Davis, J. G. (1970) Lancet, 2, 610 Dewar, E. T. & Maddy, M. L. (1970) J. Pharm. Pharmacol., 22, 791 DHSS (1972) Unpublished work quoted in Report to DHSS, London Di Rosa, M. & Sorrentino, L. (1970) Brit. J. Pharmacol., 38, 214 Dubrasquet, M., Lehy, Th. & Bonfils, S. (1970) Unpublished paper Duncan, J. G. C. (1965) Thesis University of Glasgow Eagleton, G. B., Watt, J. & Marcus, R. (1968) J. Pharm. Pharmacol., 20, 972 Evans, J. G. & Acheson, E. D. (1965) Gut, 6, 31 Evans, P. R. C., Nowell, S. & Thomas, I. A. P. (1965) Postgrad. med. J., 41, 48 Fomon, S. J., Owen, G. M. & Thomas, L. N. (1961) Amer. J. Dis. Child., 108, 601 Food and Drug Administration (1969) Unpublished summary report FDA (1972) Press release dated 6 June 1972 Food & Drug Research Lab., Inc. (1972) Unpublished results submitted to the DHEW/Public Health Service/FDA Friedman, L. & Douglass, C. D. (1960) Unpublished summary report submitted by Marine Colloids Inc. Gjone, E. & Myran, J. (1964) Nord. Med., 71, 143 Golberg (1971) Personal communication Goldbloom, R. B. (1963) Pediatrics, 32, 36 Gorten, M. K. & Cross, E. R. (1964) J. Pediat., 64, 509 Gorten, M. K., Hepner, R. & Workman, J. B. (1963) J. Pediat., 63, 1063 Hawkins, W. W. & Leonhard, V. G. (1962) J. Lab. clin. Med., 60, 641 Hawkins, W. W. & Yaphe, W. (1965) Canad. J. Biochem., 43, 479 Holzmann, H. & Schott, H. J. (1963) Naturwissenschaften, 50, 502 Houck, J. C., Bhayana, J. L. & Lee, T. (1960) Gastroenterology, 39, 196 Levey, S. & Sheinfeld, S. (1954) Gastroenterology, 27, 625 Maillet, M. (1969) Unpublished report Maillet, M. (1970) Personal communication Maillet, M. (1970a) Unpublished report prepared for Glaxo, Evans Maillet, M. (1970b) "Tolerance digestive de la carragenine native et degradée chez le rat" (Unpublished report prepared for Glaxo Laboratories) Maillet, M. (1970c) "Tolerance de la muqueuse digestive chez le rat soumis à l'ingestion de fortes doses de carragenate degrade" (Unpublished report) Maillet, M., Bonfils, S. & Lister, R. E. (1970) Lancet, 2, 414 Marcus, R. & Watt, J. (1965) Archives Maladies de l'Appareil Digestif et de la Nutrition, Paris, 54, 457 Marcus, R. & Watt, J. (1969) Lancet, 2, 489 Marcus, R. & Watt, J. (1971) Lancet, ii, 765 Marine Colloids Inc. (1969) Unpublished report Marquez, V. M. & Garcia, R. G. (1960) Rev. Fac. Farmacol. (Venezuela), 3, 237 Monis, B., Weinberg, T. & Specter, G. J. (1966) Brit. J. exp. Path., 49, 302 Morard, J. C. et al. (1964) Nature, 202, 401 Mottet, N. K. (1970) Lancet, 2, 1361 Nilson, H. W. & Schaller, J. W. (1941) Food Res., 6, 461 Nilson, H. W. & Wagner, J. A. (1959) Food Res., 24, 235 Owen, G. M. & Fomon, S. J. (1963) J. Pediat., 63, 490 Piper, D. W. & Fenton, B. (1961) Gastroenterology, 40, 638 Poulsen, E. (1973) Fd. Cosmet. Toxicol. (In press) Poulsen, E. (1972) Fd. Cosmet. Toxicol. (In press) Riccardi, B. A. & Fahrenbach, M. J. (1965) Fed. Proc., 24, 263 Robertson, W. B. Van and Schwartz, B. (1953) J. biol. Chem., 201, 689 Roches, Ph. (1965) Praxis, 54, 1230 Rorer, W. H. (1966) Inc. USA (Unpublished report) Saito & Tokunaga (1970) January 19th Toxicite Chronique de l'Ebimar Compte-Rendu Préliminaire (Unpublished report) Sharratt, M. et al. (1970) Lancet, 2, 932 Sharratt, M. et al. (1971) Lancet, 1, 192 Shirlaw (1971) Personal communication Takagi, K., Okabe, S. & Saziki, R. (1969) Jap. J. Pharmacol., 19, 418 Tournot (1970) Personal communication Tournot, J. & Labie, Ch., (1970) National Veterinary School, Toulouse (Unpublished report) Vaughan, O. W., Rezabek, H. & Filer, L. J. (1961) Fed. Proc., 20, 863B Vaughan, O. W., Filer, L. J. jr & Churella, H. (1962) Agric. Fd. Chem., 10, 517 Watt, J. & Marcus, R. (1969a) J. Pharm. Pharmacol., 21, 187 Watt, J. & Marcus, R. (1969b) Paper read to the Nutrition Society on 26 September Watt, J. & Marcus, R. (1970a) J. Pharm. Pharmacol., 22, 130 Watt, J. & Marcus, R. (1970b) Gastroenterology, 59, 760 Watt, J. & Marcus, R. (1971) Lancet, ii, 766
See Also: Toxicological Abbreviations Carrageenan and furcellaran (FAO Nutrition Meetings Report Series 46a) Carrageenan and furcellaran (WHO Food Additives Series 19)