NATURE. VOL. 223. AUGUST 16.

1969 753

Biological Methylation of Mercury

in Aquatic Organisms 100 __
,
-...--
FRESHWATER fish, especially pike (Esox lucius), from I
Sweden sometimes contain abnormally large amounts of ~
mercury'. It was initially concluded to be either inorganic 5"" I
>,
I
mercury or phenyl mercury, which are known to be !3
~ I
released as industrial wastes, but later it was shown that I
the mercury was present almost entirely as methyl mer- s"' f
>, 50
cury (CH 3Hg+) 2 • A possible explanation is that living ,s I
I
organisms have the capacity to methylate mercury s"'
compounds present in pollution. We now report that """ I
both mono and dimethylmercury (CH 3 Hg+ and
CH 3HgCH 3 ) can be produced in bottom sediments and "0
I'-< I
I

in rotten fish, and relate the findings to the hazards ),

of mercury pollution.
Monomethylmercury was analysed by conventional
gas-chromatographic detection of CH 3HgX (X = halogen) 0 5 10 15 20 25 30
by means of an electron capture detector. Confirmation Days of incubation
analyses were performed on an LKB 9000 combined
Fig. 2. Concentration of methyl mercury in bottom sediment after
gas-chromatograph mass spectrometer, with the instru- addition of 10 ( - - ) or 100 ( - · - - ) p.p.m. of inorganic mercury
ment set for detection of m/e 215 (CH,2° 0 Hg+). The followed by variable times of incubation. Lines drawn between mean
valnes from five and three samples respectively.
ionization potential was 20 eV. The column (length
180 cm, internal diameter 0·18 cm, temperature 150° C)
contained 10 per cent 'Carbowax 1500' on 'Chromosorb W' used as controls and placed in 250 ml. Erlenmeyer flasks
60-80 mesh. containing 25 ml. of water. In two other series the sus-
Gas samples to be tested for the presence of CH 3 HgCH 3 pensions were sterilized by autoclaving. All samples were
were bubbled through 1 ml. of toluene which was then incubated at 24 ° C for five or ten days and then analysed
shaken with 1 ml. of a solution of HgC1 2 (10 mg) and for CH 3Hg+. After ten days one of the series of sterilized
KBr (llO mg). In this solution CH 3HgCH 3 will form two samples was inoculated with microorganisms from the
molecules of CH 3HgBr which is estimated as above. original bottom sediments and incubated for a further
For mass spectrometric confirmation the gas sample was five or ten days.
bubbled through ethyl ether and the resulting solution Samples were analysed for CH 3Hg+. The blank showed
analysed as described. The column temperature was 60° C, background values corresponding to 40 ng of CH 3Hg+
and the mass spectrometer was set for the detection of per g of sediment after either five or ten days. Where
m/e 230 (CH 3 200 HgCH 3 ): 50 pg could be detected. Rotten 100 p.p.m. HgC1 2 had been added the corresponding values
fish to be tested for the presence of CH 3 HgCH 3 were ex- were 180 and 440 ng/g. The amounts of CH 3Hg+ in the
tracted with benzene and the extract was boiled under sterile specimens were no higher than those in the blank.
reflux with 4 M KOH. After washing with water, the After re-infection, 90 and 330 ng/g were found after five
extract was pure enough to be analysed mass spectrometri- and ten days respectively. CH 3 Hg+ added to the sedi-
cally as described. The amounts of CH 3 HgCH 3 were large ment could be recovered almost quantitatively.
enough to give a detailed mass spectrum, corresponding To determine formation of CH 3Hg+ as a function of
essentially with that of an authentic sample. HgC1 2 concentration, 1 g samples of bottom sediments
To study the methylation of HgC1 2 bottom sediments (upper 10 cm) from Lake Langsjon near Stockholm were
from freshwater aquaria, 1 g samples were treated with treated with HgCl 2 in two series. To the first series 0,
HgC1 2 or CH 3HgCl (100 p.p.m.); untreated samples were 0·l, 1, 10, 100 and 1,000 p.p.m. HgC1 2 were added and to
the other 1, 5, 10, 50, 100 and 500 p.p.m. After seven
days all samples (five or ten at each concentration) were
150 analysed for CH 3Hg+ (Fig. 1). Another set of samples
from the same lake was treated with HgC1 2 to 10 or 100
p.p.m. After 1, 2, 4, 7, 10, 14, 21 and 28 days the samples
were analysed for CH 3Hg+ (Fig. 2).
To investigate the formation of CH 3 HgCH 3 from

}/ I
/I
I
A '

\
\
\
\
\
CH 3Hg+, two dead fish (Xiphophorus maculatus), weighing
about 5 g and each containing about 125 µg of CH 3Hg+,
were placed in open glass beakers with water. After 7
weeks the whole content was analysed for total mercury
(by means of the dithizone method) and for CH 3Hg+.
A third dead fish (also containing 125 µg of added CH 3Hg+)
was placed in water in the first of two coupled washing

</
/1 \I flasks. The second flask contained 7 5 ml. of 1 M HCI.
After 4 weeks the system was flushed with nitrogen and
jl \ closed. The acid solution was analysed for CH 3Hg+.
After a further three weeks (in anaerobic conditions) the
/ \
contents of the first flask were extracted with 10 ml. of
'l \
/j \ benzene which was analysed for presence of CH 3HgCH 3 •
/// \ Results are in Table 1. Of the original 125 µg Hg, only
- __- ___/ / \ about 20 µg remained in the sample after seven weeks in
l;;:::·_:_.::_ --- - -4
-1--- -------- --- t- -
Table 1. FORMATION OF
CH,Hg+ CH,HgCH, FROM
0 0·1 1·0 10 100 1,000
Experiment
Total Hg (µg)
Added inorganic mercury (µg/g)
Original CH,Hg+ 125
Fig. 1. Concentration of methyl mercury in bottom sediment after addi- Remaining CH3 Hg+ after 7 weeks in open beaker 20
tion of inorganic mercury followed by incubation for 7 days. Lines Remaining Hg total after 7 weeks in open beaker 18
are drawn between mean values from five samples in two parallel CH,Hg+ found in washing flask 2 (HCl) after 4 weeks 38
experiment series. CH,HgCH, formed in flask 1 between weeks 4 and 7 53

© 1969 Nature Publishing Group

 754
the open beaker; in the closed system at least 53 µg, but
pos;;;ibly 91 µg, wa;, converted tu CH 3 HgCH 3 •
To study the formation of CH 3 HgCH 3 from Hg 2 + an
entire Xiphophorus helleri (about 5 g) was homogenized
togothor with about 10 g of fillet from Gadus colja. Tho
homogenate was transferred to two 50 ml. flasks. Water
containing 60 µg of HgCl 2 was added and the volume
adjusted to 25 ml. The flasks were flushed with nitrogen,
made airtight with a membrano of silicone rubber and
shaken for -four days at 20° C. Aliquots of tho gas in the
flasks wore analysed as described. The gas-chromato-
graphic analysis of tho benzene gavo no value for CH 3Hg+
before treating with HgBr 2 and KRr, indicating that the
mercury was originally present as CH 3HgCH,; the mass
spectrogram of the ether solution supported this. Qmmti-
tation by comparison with synthetic CH 3HgCH 3 yielded
a total value of 10 µg (calculated as Hg) in the head-space
gas.
In conclusion, tho sediment from tho aquarium has a
higher background value for CH:,Hg+, but also a higher
biological activity than tho sediment from Lake Langsjon.
This may be the reason for tho larger amount of CH 3Hg+
after ten days in sediments from aquaria. The biologieal
mothylation of moreury compounds provides an explana-
tion for the fact that CH 3Hg+ is found in fish, even if all
known sources of mercmy in the environment arc in the
form of inorganic m<,rcury or phenyl mercury. Tho
formation of the volatile CH 3 HgCH 3 (b.p. 94° C) may be a
factor in the redistribution of mercury from aqmmus
industrial wastes. In Sweden emphasis has been placed
on the study of the turnover of morcury in the aquatic
environment. The process of methylation is fundamental
to a knowledge of the turnover of mercury; it may be
significant in the uptake and di;:;tribution of mercury in
fish and in tho mobilization of mercury from deposits in
bottom sediments into tho gmwral environment.
We thank Leif Bergstedt for help ·with t.110 mass spectro-
metric analyses.
S, JENSEN
Institute of Analytical Chemistry,
University of Stockholm.
A. JERNELOV
.Swedish Water and Air Pollution
Research Laboratory,
Stockholm.
Received iVlarch 2-i; revised June 30, 1000.
1 Westermark, T., Kricksifre,jrdgan i Sveri(Je (Stockholm, 1D65), 25 (1064 ars
Nat.111TPs11rs11lri-~tlning, Jordbruksdcpartcmentet, KvieksilverkonfrrP118eII,
l!Hi5).
'\\'estiio, ll., Acta Clwm. &and., 20, 2131 (1066).
3 Jensen, I:\., and .TerneWv, A., Riosyntes av ,11etylkvicksili'er I, Biocid-
information, 10, e>ord!'orsk, May 1967.
4
Jensen. S .. arnl JerneWv, A., Rfrn;yntes av .1lfetylkvicksifoer II, Biocid-
informal'ion, 14, ?\ ordforsk, .Fch. 1 H68.
5 Ha.nncrz, L., Annual report from FrPRh ,vatPr Research T,ahoral'ory, DroU~
11i11µ:liolm, 1968. '
6
.Tenu-~rnv, A., Procrcdings of the first l{.ochel-iter Conti:>renee on Toxicity,
J"tme 1968 (in th<' press).
Presumed Super-foetation in an
Erythrocebus patas Monkey
ON November 22, 1964, a female specimen of Erythrocebns
patas monkey waR recoived at. Tigoni Primate Research
Centre (now the National Primate Resoareh Centro). Her
exact, provonance was unknown, as she had been rescued
from Romo poasant. farmors and brought in by her rescuer.
At the time of her arrival at the cent.rn, hor second milk
molars wm-e just erupting. In 1968, after reaching adult
status, Rho Rhared a cugo with an adult male of the same
species for several months, but mating was not actually
observed.
© 1969 Nature Publishing Group
NATURE, VOL. 223, AUGUST 16, 1969
On January 17, Hl69, the male was removed from her
cage. On March 8, 1969, she gave birth to a stillborn
male, which was, however, "full term"_ Three 1nonths
later on June 8, 1969, she gave birth to another stillborn
baby, this time a female, which was also "full term".
Because the average gestation period for Erythrocebus
patas is 170 days' this would appear to be a case of super-
foetation; tho first infant was sired around the end of
September 1968, and t.he second sometime in December_
I have not been able to trace any similar record of super-
foetation in this or allied species of monkey, but our
literature hero is very incomplete. Super-foetation in
wornen is, of course, a well recognized phenornenon 2 •
L. s. B. LBAKEY
National Primate Research Centre,
Nairobi.
Received June 25, 1969.
1Napier, J, R., :<nd Napier, P.H., A Handbook of Living Primates (1969).
'Kerr, M .. and Moir, C., Operative Obstetrics, 217 (1969),
Evidence for Extraterrestrial Life:
Identity of Sporopollenin with the
Insoluble Organic Matter present in the
Orgueil and Murray Meteorites and
also in some Terrestrial Microfossils
Wl'l'H very few exceptions 1 the insoluble organic matter
present in both Pre-Cambrian sediments and carbonaceous
chondrites has boon neglected and organo-geochomical
studies of these materials have been largely devoted to
the readily solvent extractable soluble organic substances 2 •
This is in some ways unfortunate, for by far tho greater
proportion of carbonaceous mat.tor in both Pre-Cambrian
sediments (up to 95 per cent)" and in carbonaceous
chondrites (up to 70 por cont)• is insoluble and tho soluble
mattor is frequently of a very minor nature. In addition,
the solublP, and so potentially more mobilB, orgR,nic
chemic,-Js aro more likdy to have moved in total or in
part from their point of origin, and problems of rock
contamination with such substancos, either ovor long
periods through soopngo or, in the case of motooritos, at
impact•, are especially acute.
\-Ve }mve shown• that sporopollcnin which forms a
major part of pollen and spore exines is an oxidative
polymer of carotonoids a,nd rn•,rotenoid esters, and have
;:;uggcstccl' that it is identical with oldm korogen derived
fro-m terrestrial, especially Pre-Cambrian, sodimcnt,s. Wo
now pre,tnt briefly somo rnsults of experiments carried
ont 011 tho Orgueil and Murray motoorites which in our
opinion clC'nrly establish that tho in,mluble mattnr tlwy
oont&in is identical with sporopollenin.
Tho insoluble organic matter was isolated from samples
of the Orgueil (O·l g) and Murray (0·9 g) motooritos as
brown amorphous Rolids by repeated digestion with
hydrofluoric and nitric acids and potassium hydroxide in
the usual manner, taking maximum care in manipulation.
The relatively largo amounts of organic matter (Orgueil
3·5 por cent, Murray 4·4 per cont) would make trace
contamination oflittle or no consequence to onr sub;:;oqucnt
chemical studies. The solids were examined by infrared
spectroscopy, pyrolysis gas chromatography, potash
fusion followed by thin-layer chromatography of the
product,s and by somo elemental analyses•, and results
compared with similar examinations of sporopollenins
from many modern pollen and spore exinos, some
synthetic analogues, some microfossils and artificially
metamorphosed (by heating with sand) spore exines.