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
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.
Hexamethylenetetramine was evaluated for acceptable daily intake
by the Joint FAO/WHO Expert Committee on Food Additives (see Annex 1,
Refs No. 6, No. 9 and No. 27) in 1961, 1964 and 1971. Since the
previous evaluation, additional data have become available and are
summarized and discussed in the following monograph. The previously
published monographs have been expanded and are reproduced in their
Under acid conditions, or in the presence of proteins,
hexamethylenetetramine (HMT) decomposes gradually yielding ammonia
and formaldehyde (Hutschenreuter, 1956).
There is a theoretical possibility that bis-chlormethyl ethers
may be formed in the stomach from the reaction of formaldehyde with
chloride ions. However, whilst this occurs readily in the gaseous
phase it is much less likely in liquid phases (Conning, 1973).
Because of these chemical and biochemical properties it is
pertinent to consider biochemical information on formaldehyde and its
metabolite formic acid in relation to the toxicological evaluation of
Formaldehyde and formic acid occur in fresh unpreserved codfish
muscle (Amano & Yamada, 1964), also in honey, roasted foods and some
fruits (Malorny, 1969a).
Formaldehyde and formic acid are excreted in the urine of dogs
and cats fed a meal of fish not treated with preservatives, the source
of the substances being trimethylamine oxide occurring naturally in
fish muscle (Malorny & Rietbrock, 1962; Malorny et al., 1963;
Schassan, 1963). The concentration of free formaldehyde in the urine
of these dogs and cats given unpreserved fish was four to five times
higher than that found in human subjects who had consumed fish
preserved with hexamethylenetetramine. Humans fed almost exclusively
on unpreserved fish did not show an increase in the excretion of
formaldehyde or formic acid (Malorny & Rietbrock, 1962). Formic acid
present in the diet is mostly metabolized by man, only 25 mg is
excreted daily in the urine (Malorny, 1969b).
Hexamethylenetetramine liberated free formaldehyde in the stomach
(Malorny & Rietbrock, 1963).
I.v. infusion in dogs of formaldehyde, 35 mg/kg as a 0.6%
solution during eight to 10 minutes, was carried out. Two minutes
after completion of the infusion there was no increase in formaldehyde
but a high increase in the formic acid level in the blood. The formic
acid level decreased by 50% in one hour, by 75% after two hours, and
was back to normal after four hours. In vitro human blood oxidizes
30% formaldehyde (0.7 mg to 4 ml blood) to formic acid within four
hours. Addition of methylene blue (0.05% to 0.2%) results in a
complete oxidation of the formaldehyde (Malorny et al., 1964).
Similar maximal plasma concentrations of formic acid were found
after i.v. injection of equimolar solutions of formaldehyde, formic
acid or sodium formate, and the elimination constants and biological
half-lives of formic acid in these different cases were also similar
(Malorny et al., 1965). In another study, it was found that
elimination of formic acid from the body is much more rapid in rats
than in dogs, the biological half-lives being about 12 min and 77 min
respectively, while those of the guinea-pig are 12 min, rabbit 32 min
and for the cat the value is 67 min. The rate of oxidation of
formaldehyde to formic acid is comparable in all these species, with a
half-life of only 1 min (Rietbrock, 1965).
The kinetics of the absorption of formaldehyde by erythrocytes
and its conversion to formic acid were studied in rabbits, guinea-
pigs, rats and cats. The elimination of formic acid was proportional
to the concentration present but formaldehyde had the same biological
half-life of 1 min irrespective of species. Using three different i.v.
infusion rates (8, 2 or 0.5 ml/min) for formaldehyde showed that only
at 0.5 ml/min there was no free formaldehyde present in the blood but
only formic acid. Formic acid was eliminated from erythrocytes,
plasma, liver, kidney, testes, muscle and brain in an exponential
manner. The half-life varied with species but was always shortest in
the liver. No relation to catalase activity of the blood was
established. Pre-treatment of the cat with folic acid reduces the
half-life for formic acid. Treatment with methotrexate, a folic acid
inhibitor, prolongs the half-life in the dog from 61 to 480 min
Formaldehyde, administered orally to dogs, was rapidly absorbed
and converted to formic acid. No formaldehyde was detected in the
plasma and only traces occurred transiently in the erythrocytes. In
cats the production of formic acid and influx of lactic acid into the
blood caused a temporary acidosis. When formaldehyde was added to
human blood in vitro it was rapidly absorbed on to the erythrocytes
and then oxidized to formate. Both NAD and NADP-independent
formaldehyde dehydrogenase and catalase appeared to be involved in
this oxidation, so that both erythrocytes and the liver played an
important role in the rapid detoxication of orally and parenterally-
administered formaldehyde (Malorny et al., 1965).
The elimination of formate is folic acid dependent, since in the
cat daily injections of folic acid for several days before giving
formate have been shown to reduce the biological half-life of formate
considerably, from 67 min without folic acid pre-treatment to 46 min
after four days' treatment and 17 min after 10 days' treatment
(Rietbrock, 1965). Interference with liver function prolongs the
biological half-life of formic acid in rabbits from 35 to 130 min
(Malorny, 1969a). In vitro experiments with liver homogenate showed
that after 30 min 31.6%, 3.9% and 8.0% of formaldehyde was found in
the case of rat, guinea-pig and rat liver respectively (Neugebauer,
Experiments with 14C-labelled formaldehyde given by stomach tube
to rats and mice showed that 5 min after application the radioactivity
was distributed over the total body. After 12 hours, approximately 40%
had been expired as 14CO2, 10% had been excreted in the urine and 1%
in the faeces. The homogenized whole animals contained 20% of the
radioactivity after 24 hours and 10% after 96 hours (Buss et al.,
When female rats were given 70 mg/kg of 14C-labelled
formaldehyde i.p., 82% was detected as 14CO2 and the urine
contained 13 to 14% of the isotope in the form of methionine, serine
and a formaldehyde-cysteine adduct (Neely, 1964).
The yellow discolouration observed in rats treated with oral or
i.m. HMT was shown to be due to a reaction between formaldehyde in the
urine and kynurenine in the rat hair (Kewitz & Welsch, 1966).
Special studies on mutagenicity
Both HMT (Auerbach, 1951) and formaldehyde (Rapoport, 1946) have
been shown to act as mutagens in Drosophila. Stumm-Tegethoff (1964)
has demonstrated that formic acid, present as an impurity in
formaldehyde, is the causative agent in producing the mutagenic effect
in Drosophila melanogaster. Nafei & Auerbach (1964) have shown that
mutations only occur in the larval spermatocytes and that the
conditions necessary for the change only exist for a short time.
Special studies on reproduction
In a five-generation study lasting three-and-a-half years, a
total of 80, 80 and 245 rats were fed 0, 5 and 50 mg/kg daily of HMT
in the drinking-water. Animals, including pregnant dams, were selected
from each group at half-yearly intervals, starting at one-and-a-half
years, for pathological study. No changes attributable to HMT were
found in test animals or foetuses and placentas at either level.
Tumours were observed on three of 48 animals at the high dose level
Groups of 10 male and 10 female rats were fed 0, 400, 800 and
1600 mg/kg of hexamethylenetetramine in a normal basic diet for two
years. The 10 pairs were mated at the age of 20, 28 and 35 weeks. No
effect was found at all levels on growth, two-year survival,
reproduction and viability of offspring. Post-sacrifice examination of
survivors (organs examined and type of examination not stated)
disclosed no specific pathological changes (Berglund, 1966).
Twelve female and six male Wistar rats were given 1% HMT in the
drinking-water starting two weeks before mating. The females were kept
under treatment during pregnancy and lactation. A similar untreated
group of 12 females and six males served as control. Twelve treated
females and 11 controls became pregnant and gave birth to 124 and 118
babies respectively; no malformations were noted. From these animals,
24 for each sex were continued on the 1% HMT up to the twentieth week
of age or were kept untreated. The body weight of treated animals was
significantly lower than that of the controls, only up to the ninth
week of age for the males and up to the thirteenth week for the
females. At the end of the treatment both groups were sacrificed; the
weight of organs was identical in the treated and control animals;
there was no gross or histological pathology. In a second experiment,
rats were given 1% HMT in the drinking-water for three successive
generations, up to the age of 40 weeks in the F1 and F2 groups and
of 20 weeks for F3. The three groups were composed of 13 male and
seven female, 15 male and 11 female, 12 male and 12 female,
respectively. In addition, a group of 16 male and 16 female
descendants of 2% HMT-treated parents were given 2% HMT for 59 weeks.
A group of 48 males and 48 females served as untreated controls. All
groups were kept under observation for over two years. No evidence of
carcinogenicity was found in any of the HMT-treated groups (Della
Porta et al., 1970).
A five-generation study using eight male and 24 female rats as
P generation is in progress using eight controls. Test animals
received 0.2% calcium formate in their drinking-water. No
abnormalities were detected regarding growth and fertility. No
disturbances of organ function or retinal or fundal abnormalities were
noted. Gross and histopathology of 250 rats from three generations
revealed no abnormal findings related to the administration of calcium
formate. Tumour incidence was no different from controls. Pregnancy
and foetuses as well as numbers of pups per litter were similar to
controls. No malformations or placental changes were detected. Similar
groups were given 0.4% calcium formate in their drinking-water for two
generations and will be continued. No abnormalities, including
histopathology, related to the test substances have been seen.
A group of two male and four female mongrel dogs was fed from
1250 to 1875 mg/kg of HMT and a group of four males and four females
was fed from 125 to 375 mg/kg of formaldehyde for 32 months. The
control group consisted of two males and one female. Groups formed
from litters of these animals were also fed test diets; three males
and two females were fed 1250 mg//kg of HMT, three males and two
females were fed 375 mg/kg of formaldehyde and two animals of each sex
served as a control group for 22 months. At the end of the test
period, the test animals were placed on a normal diet and the controls
on 1250 mg/kg of HMT for one year. No effect of formaldehyde or HMT
was found on food consumption, growth, reproduction or litter numbers
and weights or in monthly observations on blood chemistry and cell
total and differential counts, and periodic urine examinations.
However, of 30 litters in the HMT group, 66.7% were noted as
unusual in that there were stillborn and eaten animals as well as five
cases of defective animals born; in the formaldehyde group, 60% of 34
litters were noted as unusual, with 10 cases of defective animals. Of
16 litters in the control group, 7.7% were noted as unusual, with no
malformations (Kewitz, 1966).
According to a preliminary report, groups of eight pregnant
female beagle dogs received from the fourth to fifty-sixth day after
mating either 0, 0.0125% and 0.0375% of formalin or 0.06% and 0.125%
of HMT in their food. Many litters are still in the pre-weaned stage
at present. So far no adverse effects have been reported with regard
to rate of pregnancy and duration, behaviour and weight of the mothers
or litter size, pup weight, incidence of congenital defects or
survival to weaning of the pups (Tierfarm, 1969).
Special studies on teratogenicity
Two groups of 39 male and 44 female CTM mice received five s.c.
injections on five alternate days of 5 g/kg HMT, a 30% aqueous
solution, beginning on day 10 of age. They were observed for 100
weeks. No pathological findings related to treatment were noted nor
was there any difference from controls regarding tumour incidence
(Della Porta et al., 1968).
Twenty rats were given s.c. doses of 1 ml of a 0.4% aqueous
solution of formaldehyde, in the dorsal region, weekly for 15 months.
Half the animals died during the experiment without developing
tumours; of the remaining 10, four showed sarcomas three to five
months after cessation of treatment. Two of these tumours developed at
the site of injection and were diagnosed as spindle-cell sarcomas. One
tumour developed in the liver (male, 21 months) with metastases to the
colon; the other was in the colon (female, 21 months) (Watanabe et
al., 1953; Watanabe et al., 1954).
Twenty rats were injected s.c. on the left dorsal side once or
twice weekly with 1-2 ml solution of hexamethylene-tetramine, using
at first a concentration of 9.23% but, after the middle of the
experiment, a concentration of 35-40%. In addition, shortly before
each injection of hexamethylene-tetramine, an injection of 0.5 ml 0.1%
formic acid was given on the right dorsum. Of 14 survivors, eight
showed tumours at the sites of injection of hexamethylene-tetramine.
At the sites of formic acid injection small ulcers sometimes formed
but no s.c. thickening or nodules. Of the eight tumours seven were
sarcomas and one syringadenoma (Watanabe & Sugimoto, 1955).
Groups of 15 male and 15 female albino rats were injected s.c.
with 1 ml of 40% HMT solution and 1 ml of 0.1% formaldehyde solution
weekly for one-and-a-half years. Control groups of seven males and
seven females and eight males and eight females were similarly
injected with doses of sodium chloride or sucrose (equiosmotic with
40% hexamethylenetetramine) for one-and-a-half years, by which time
the last animal had died. In the HMT group, one injection-site
spindle-cell sarcoma, one distal spindle-cell sarcoma, one alveolar
mammary carcinoma and one fibrosarcoma, in addition to two benign
tumours were seen. In the formaldehyde group, two malignant tumours
and three benign fibro-adenomas were seen. In the control group, one
blastoma in a sodium chloride-injected animal) was found (Kewitz,
Two groups of 20 male and 20 female rats received five s.c.
injections of 5 g/kg HMT as 30% aqueous solution on alternate days
starting at day 10 of age and were observed for 104 weeks. No
treatment-related abnormalities were seen regarding growth, mortality,
gross or histopathology or tumour incidence. The test animals had a
lower tumour incidence than the controls (Della Porta et al., 1968).
Groups of from nine to 11 mated beagle bitches were fed 0, 3.1 or
9.4 mg formaldehyde (FA) per kg bw, or 15 or 31 mg hexamethylene-
tetramine (HMT) kg bw from the fourth to fifty-sixth day after mating.
These dose levels were based on an average body weight of 12 kg.
Implantation, maintenance and duration of pregnancy and litter size
were unaffected by any treatment. At 31 mg HMT kg per day, incidence
of stillbirths was slightly increased, and post-natal survival to
weaning was decreased. Birth weight in all treated groups was slightly
reduced as was post-natal growth. These effects were most noticeable
at 9.4 mg FA/kg and 31 mg HMT/kg dose levels. No gross abnormalities
were observed in any offspring either at birth or weaning. Stillborn
pups were examined for skeletal and tissue malformations but none were
observed (Hurni & Ohder, 1973).
Special studies on chick embryo
Two hundred fertilized chicken eggs were injected with 5 mg HMT
after 96 hours incubation. The distribution of HMT and formaldehyde
was determined between the sixth and nineteenth day of incubation in
the various components of the egg. HMT is continuously metabolized. No
toxic effect was noted in the chick embryos examined (Pilz, 1967).
Formaldehyde was injected in another experiment into fertilized
eggs. There was rapid oxidation to formic acid. After 48 hours some
90% had been converted to formic acid. The remaining formaldehyde was
probably bound to the albumin and the shell proteins. Injected sodium
formate is metabolized at a much slower rate but had disappeared by
the twelfth day of incubation. Practically no formaldehyde metabolism
occurs in the extra-embryonal components but embryonic liver and
erythrocytes rapidly oxidize formaldehyde to formic acid (Lange,
Sodium formate did not show any toxic or teratogenic effect when
submitted to the chick embryo test (Malorny, 1969a).
Animal Route (mg/kg bw)
(formic acid) 1 100 Malorny, 1969a
(sod. formate) 11 200 "
(pot. formate) 5 500 "
(amm. formate) 2 250 "
(calc. formate) 1 920 "
(formic acid) 145 "
(sod. formate) 807 "
(pot. formate) 95 "
(amm. formate) 410 "
(calc. formate) 154 "
Rat i.v. (HMT) 9 200 Malorny et al., 1965
One group of 14 six-week-old male and female rats were given a
single i.v. injection of 10 g HMT/kg bw as an 80% aqueous solution.
All animals survived without evidence of toxic effect. Two groups of
five adult rats were given 10 or 20 g HMT/kg bw as an 80% aqueous
solution by oral intubation. All animals survived (Della Porta, 1968).
Groups of 13 mice were treated daily by cutaneous application 10%
HMT in chloroform, 1.5% aqueous formaldehyde, and chloroform alone,
for 300 days. No malignant tumours were found in any group (Kewitz,
One group of five male and five female rats was given by gavage
400 mg HMT daily for 90 days. A second group of five male and five
female rats was given daily intramuscular injections of 200 mg HMT for
90 days. A third group of 15 male and 15 female rats was given
repeated oral doses of 400 mg HMT by gavage over 333 days. No adverse
effects were noted in any group except for a citrus-yellowish
discolouration of the fur (Brendel, 1964).
One male and three female cats were fed approximately
15 000 ppm (1.05%) of formaldehyde (375 mg/kg/day) and two males and
three females were fed approximately 50 000 ppm (50%) of HMT
(1250 mg/kg/ day) for two years. The control group consisted of three
males and three females. One female in the formaldehyde group died in
the seventh month of pleuritis and a female of the HMT group died in
the twenty-third month of a pyogenic infection of the nasal cavity and
paranasal sinuses. No effect of formaldehyde or HMT was found on food
consumption, weight gain or appearance. Preliminary microscopic
examination of tissues from test and control animals show no changes
attributable to formaldehyde or HMT (Kewitz, 1966).
Sixteen groups of 30, 50 or 100 males and an equal number of
females of CTM, SWR or C3Hf strain mice received in their drinking-
water either 0, 0.5, 1 or 5% hexamethylenetetramine over 60 weeks. The
group receiving 5% was treated for 30 weeks only. After the end of
treatment animals were observed for the remainder of their lifetimes.
The coats of treated mice showed no discolouration. The group on 5%
HMT showed slightly reduced growth rate and survival and the SWR group
on 1% showed also slight growth retardation. Gross and histopathology
did not show any pathological lesions related to treatment. As regards
tumour incidence there was no significant difference between treated
and control groups (Della Porta et al., 1968).
Groups of 30 male and 30 female NMRI/Ban albino mice were fed 0
and 1% of HMT and 0.15% formaldehyde for two years. Benign and
malignant tumours were found in a total of 43 animals: 20 in the HMT
group, 12 in the formaldehyde group and 11 in the control group.
Except for one control male and two males receiving HMT all tumours
occurred in females. Twenty-nine of 36 malignant tumours were s.c.
carinomas and adenocarcinomas. The author concluded from the data that
the possibility of an increased tumour incidence effect by HMT could
not be ruled out. A further study was instituted using groups of 50
female mice at levels of 0, 0.1, 0.5 and 1% hexamethylenetetramine,
which, after 31 weeks, showed no difference in tumour incidence
between the groups (Kewitz, 1966).
Four groups of 48 male and 48 female rats received for 104 weeks
either 0 or 1% HMT in their drinking-water while another two groups of
12 male and 12 female rats received 5% in their drinking-water for
only two weeks. The coat of treated rats showed a yellowish
discolouration due to a reaction between formaldehyde present in the
urine and kynurenine in the rat hair. 50% of the rats given the high
dose level died within the two weeks of administration, the remainder
survived for 104 weeks. Growth, mortality experience, gross and
histopathology of the treated animals showed no specific changes due
to administration of HMT. Tumour incidence was essentially similar in
controls and treated animals at 1% level and lower than in controls at
the 5% level (Della Porta et al., 1968).
Two groups of 16 male and 16 female rats were given either 0 or
0.16% HMT in their diet and were observed over 28-30 months. General
health and behaviour as tested by activity in the revolving drum at 1,
3, 7 and 14 months showed no significant differences between controls
and test groups. Body weights of male rats were slightly lower in the
treated animals. Mortality experience was similar for all groups. The
addition of HMT had no effect on palatability of the diet. Organ
weights showed no gross differences and tumour incidence was not
different between controls and test animals, although not all rats
were examined by autopsy. Separate mating of the test and control
groups revealed no differences in fertility as measured by numbers of
pups produced. The young showed no difference in activity as measured
by the revolving drum, nor did they show any abnormalities or
differences in body weight. Organ weights and gross and histopathology
were unremarkable. Some test rats showed yellowish staining of hair
(Natvig et al., 1971).
OBSERVATIONS IN MAN
Twelve human volunteers received orally 1.48-2.96 g sodium
formate. They excreted on average 13 mg formic acid per 24 hours from
control diet and only 23% per 24 hours of the additional formic acid.
Most of the additional load was excreted within six hours and all had
been eliminated after 12 hours. No cumulation was observed. Excretion
of formate is accompanied by alkaline urinary pH and mild diuresis.
The biological half-life of orally administered sodium formate was
found to be 46 min as determined by plasma levels in three volunteers.
Four to 7% of the administered dose was excreted in 24 hours. No
metabolic acidosis was found in man because of rapid metabolism of the
formate anion but the excess of bicarbonate is excreted as alkaline
urine. Formic acid is absorbed rapidly from the stomach in
undissociated form (Malorny, 1969b).
As the toxicological effects of hexamethylenetetramine appear to
be due to the liberation of formaldehyde and its oxidation product,
formic acid, much work has been done on formaldehyde and formates as
well as on hexamethylenetetramine. The metabolic studies point to
rapid conversion of formaldehyde to formate and then at a slower rate
to water and carbon dioxide particularly by the liver and the
erythrocytes. In man, very little formate is excreted in the urine
after its oral ingestion. Only one of many long-term studies in
mice conflicts with other studies in this species as well as in
the rat. These long-term studies point to a dietary level of 1%
hexamethylenetetramine as causing no effect in rodents. The
teratogenicity study in dogs revealed a no effect level of 15 mg/kg.
Other reproduction studies in rats using dietary levels from 0.1-1%
hexamethylenetetramine revealed no adverse effects. Although there is
production of local sarcomata in rats at the site of repeated
injections, the probability of carcinogenic potential in food additive
use appears to be excluded as a result of the findings in adequate
studies using oral administration. In addition, the experiments on
pregnant and lactating rats indicate the absence of carcinogenic
effects in the offspring. However, in vitro formation of nitrosamine
has been reported as a result of the interaction of nitrite with
hexamethylenetetramine at pH 1-3.
Level causing no toxicological effect
Dog: 15 mg/kg bw.
Estimate of acceptable daily intake for man
0-0.15 mg/kg bw.
Amano, K. & Yamada, K. (1964) FAO Symposium on the Significance of
Fundamental Research in the Utilisation of Fish, Paper WP/II/9
Auerbach, C. (1951) Cold Spr. Harb. Symp. quant. Biol., 16, 199
Berglund, F. (1966) Unpublished report
Brendel, R. (1964) Arzneimittel-Forsch., 14, 51
Buss, J. et al. (1964) Arch. exp. Path. Pharmak., 247, 380
Conning, D. (1973) Personal communication
Della Porta, G., Cabral, J. R. & Parmiani, G. (1970) Tumori, 56, 325
Della Porta, G., Colnaghi, M. I. & Parmiani, G. (1968) Fd. Cosmet.
Toxicol., 6, 707
Hurni, H. & Ohder, H. (1973) Fd. Cosmet. Toxicol., 11, 459
Hutschenreuter, H. (1956) Z. Lebensmitt.-Untersuch., 104, 161
Kewitz, H. (1966) Unpublished report
Kewitz, H. & Welsch, F. (1966) Arch. exp. Path. Pharmak., 254, 101
Lange, G. (1968) Thesis submitted to Hamburg University
Malorny, G. (1966a) N.-S. Arch. exp. Path. Pharmak., 255(1), 42
Malorny, G. (1966b) Unpublished report
Malorny, G. (1969a) Zeitschr. Ernahrungswiss, 9, 332
Malorny, G. (1969b) Zeitschr. Ernahrungswiss, 9, 340
Malorny, G., Netter, K. J. & Seidel, G. (1964) Untersuchungen über die
Hemmbarkeit enzymatischer Reaktionen durch Formaldehyd und einige
seiner Homologen, Arch. ext. Path. Pharmak., 247, 381
Malorny, G. & Rietbrock, N. (1962) Naturwissenschaften, 49, 520
Malorny, G. & Rietbrock, N. (1963) Vortrag auf der X. Tagung des
Ernährungswissenschaftlichen Beirats der deutschen
Malorny, G., Rietbrock, N. & Schassan, H. H. (1963) Arch. exp. Path.
Pharmak., 246, 62
Malorny, G., Rietbrock, N. & Schneider, M. (1965) Arch. exp. Path.
Pharmak., 250, 419
Nafei, H. & Auerbach, G. (1964) Z. Vererbunsl., 95, 351
Natvig, H., Andersen, J. & Wulff Rasmussen, E. (1971) Unpublished
report submitted to WHO
Neely, W. B. (1964) Biochem. Pharmacol., 13, 1137
Neugebauer, G. (1967) Thesis submitted to Hamburg University
Pilz, R. (1967) Thesis submitted
Rapoport, I. A. (1946) C. R. Acad. Sci. USSR, 54, 65
Rietbrock, N. (1965) Arch. exp. Path. Pharmak., 251, 189
Schassen, H. H. (1963) Inaugural dissertation, Hamburg University
Stumm-Tegethoff, Bertha (1964) Naturwissenschaften, 51, 646
Tierfarm, A. G. (1969) Unpublished report
Watanabe, F. et al. (1954) Gann, 45, 451
Watanabe, F., Matunaga. T. & Yamabe, T. (1953) Trans. Jap. path. Soc.,
Watanabe, F. & Sugimoto, S. (1955) Gann, 46, 365