INTOX Home Page




    This report contains the collective views of an international group of
    experts and does not necessarily represent the decisions or the stated
    policy of the United Nations Environment Programme, the International
    Labour Organisation, or the World Health Organization.

    Published under the joint sponsorship of
    the United Nations Environment Programme,
    the International Labour Organisation,
    and the World Health Organization

    First draft prepared by Dr. S. Dobson,
    Institute of Terrestrial Ecology, United Kingdom,
    and Dr. R. Cabridenc, Institut National de
    Recherche Chimique Applique, France

    World Health Orgnization
    Geneva, 1990

         The International Programme on Chemical Safety (IPCS) is a
    joint venture of the United Nations Environment Programme, the
    International Labour Organisation, and the World Health
    Organization. The main objective of the IPCS is to carry out and
    disseminate evaluations of the effects of chemicals on human health
    and the quality of the environment. Supporting activities include
    the development of epidemiological, experimental laboratory, and
    risk-assessment methods that could produce internationally
    comparable results, and the development of manpower in the field of
    toxicology. Other activities carried out by the IPCS include the
    development of know-how for coping with chemical accidents,
    coordination of laboratory testing and epidemiological studies, and
    promotion of research on the mechanisms of the biological action of

    WHO Library Cataloguing in Publication Data

    Tributyltin compounds.

        (Environmental health criteria ; 116)

        1.Trialkyltin compounds - adverse effects  2.Trialkyltin compounds
         -toxicity      I.Series

        ISBN 92 4 157116 0        (NLM Classification: QV 290)
        ISSN 0250-863X

         The World Health Organization welcomes requests for permission
    to reproduce or translate its publications, in part or in full.
    Applications and enquiries should be addressed to the Office of
    Publications, World Health Organization, Geneva, Switzerland, which
    will be glad to provide the latest information on any changes made
    to the text, plans for new editions, and reprints and translations
    already available.

    (c) World Health Organization 1990

         Publications of the World Health Organization enjoy copyright
    protection in accordance with the provisions of Protocol 2 of the
    Universal Copyright Convention. All rights reserved.

         The designations employed and the presentation of the material
    in this publication do not imply the expression of any opinion
    whatsoever on the part of the Secretariat of the World Health
    Organization concerning the legal status of any country, territory,
    city or area or of its authorities, or concerning the delimitation
    of its frontiers or boundaries.

         The mention of specific companies or of certain manufacturers'
    products does not imply that they are endorsed or recommended by the
    World Health Organization in preference to others of a similar
    nature that are not mentioned. Errors and omissions excepted, the
    names of proprietary products are distinguished by initial capital



1. SUMMARY         

   1.1. Physical and chemical properties  
   1.2. Analytical methods    
   1.3. Sources of environmental pollution    
   1.4. Regulations on use    
   1.5. Environmental concentrations  
   1.6. Transport and transformation in the environment   
   1.7. Kinetics and metabolism   
   1.8. Effects on microorganisms 
   1.9. Effects on aquatic organisms  
          1.9.1. Effects on marine and estuarine organisms 
          1.9.2. Effects on freshwater organisms   
          1.9.3. Microcosm studies 
   1.10. Effects on terrestrial organisms  
   1.11. Effects on organisms in the field 
   1.12. Toxicity to laboratory mammals    
          1.12.1. Acute toxicity    
          1.12.2. Short-term toxicity   
          1.12.3. Long-term toxicity    
          1.12.4. Genotoxicity  
          1.12.5. Reproductive toxicity 
          1.12.6. Carcinogenicity   
   1.13. Effects on humans 


   2.1. Identity of tributyltin compounds 
   2.2. Physical and chemical properties  
   2.3. Analytical methods    
          2.3.1. Measurement of organotin compounds    
           Extraction of tributyltin derivatives   
           Formation of volatile derivatives   
           Separation of organotin derivatives 
           Detection and measurement of different forms 
                            of organotin    
          2.3.2. Interlaboratory calibrations  


   3.1. Uses                  
   3.2. Production            
   3.3. Regulations           


   4.1. Adsorption onto and desorption from particles 
   4.2. Abiotic degradation   
          4.2.1. Hydrolytic cleavage of the tin-carbon bond    
          4.2.2. Photodegradation  

   4.3. Biodegradation        
   4.4. Bioaccumulation and elimination   


   5.1. Sea water and marine sediment 
   5.2. Fresh water and sediment  
   5.3. Sewage treatment  
   5.4. Biota                 


   6.1. Metabolism of TBT in mammals  
   6.2. Metabolism of TBTO in other organisms 
   6.3. General mechanisms of toxicity of TBTO    
          6.3.1. General toxic mechanisms  
          6.3.2. Toxic mechanisms in bivalve molluscs  


   7.1. Bacteria and fungi    
   7.2. Freshwater algae  
   7.3. Estuarine and marine algae    


   8.1. Aquatic plants        

   8.2. Aquatic invertebrates 
          8.2.1. Trematode parasites of man    
          8.2.2. Freshwater molluscs   
           Acute toxicity  
           Short- and long-term toxicity   
           Factors affecting toxicity  
          8.2.3. Marine molluscs   
           Acute toxicity  
           Short- and long-term toxicity   
           Reproductive effects    
           Effects on growth   
           Shell thickening    
          8.2.4. Crustaceans   
           Acute effects
           Short- and long-term toxicity   
           Reproductive effects    
           Limb regeneration   
           Behavioural effects 
          8.2.5. Other aquatic invertebrates   
           Acute effects   
           Limb regeneration   
   8.3. Fish                  
          8.3.1. Acute effects 
          8.3.2. Short- and long-term toxicity 
          8.3.3. Embryotoxicity    
          8.3.4. Behavioural effects   

   8.4. Amphibians            
   8.5. Multispecies studies  


   9.1. Microcosm studies 
   9.2. Terrestrial insects   
   9.3. Terrestrial mammals   


   10.1. Effects on bivalves   
   10.2. Effects on gastropods: imposex    
   10.3. Effects on farmed fish    
   10.4. Effects of TBT-contaminated sediment  
   10.5. Effects of freshwater molluscicides   
   10.6. Effects from spills   
   10.7. The use of indicator species for monitoring the environment


   11.1. Single exposure       
          11.1.1. Oral and parenteral administration    
          11.1.2. Dermal administration 
          11.1.3. Administration by inhalation  
          11.1.4. Irritation and sensitization  
          Skin irritation 
          Eye irritation  
          Skin sensitization  
          11.1.5.  In vitro studies  
   11.2. Short-term toxicity   
          11.2.1. Oral dosing: general body effects 
          11.2.2. Inhalation studies    
          11.2.3. Histopathological effects
          11.2.4. Haematological and biochemical effects    
          11.2.5. Effects on lymphoid organs and immune function    
          11.2.6. Mechanism of immunotoxicity   
          11.2.7. Effects on the endocrine system   
   11.3. Long-term toxicity    
   11.4. Genotoxicity          
   11.5. Reproductive toxicity 
          11.5.1.  In vivo   
          11.5.2.  in vitro  
   11.6. Carcinogenicity       

12. EFFECTS ON HUMANS          

   12.1. Ingestion             
   12.2. Inhalation            
   12.3. Dermal exposure       
   12.4. Miscellaneous effects 


   13.1. Evaluation of human health risks  
   13.2. Evaluation of effects on the environment  

14. RECOMMENDATIONS            

   14.1. Recommendations for protecting human and environmental health
   14.2. Research needs        






MEDIO AMBIENTE                        




Dr C.  Alzieu,  French  Institute for  Research on Exploi-
   tation of the Sea, Nantes, France

Dr I.J. Boyer, Division of Toxicological Review and Evalu-
   ation, Food & Drug Administration, Washington, DC, USA

Dr A.H.  El-Sabae, Faculty of Agriculture, Alexandria Uni-
   versity, Alexandria, Egypt

Dr B.  Gilbert, Company for the  Development of Technology
   Transfer  (CODETEC),  Cidade  Universitaria,  Campinas,

Dr Y. Hayashi, Biological Safety Research Centre, National
   Institute  of  Hygienic  Sciences, Setagaya-ku,  Tokyo,

Dr R.  Koch, Institute for Geography & Geoecology, Academy
   of Sciences, German Democratic Republic  (Chairman)

Dr E.I.  Krajnc, National Institute for  Public Health and
   Environmental Hygiene, Bilthoven, Netherlands

Dr H.   Schweinfurth,  Schering  AG,   Chemical  Industry,
   Bergkamen, Federal Republic of Germany

Mr D.  Spatz, Office of Pesticide Programs, US Environmen-
   tal Protection Agency, Washington, DC, USA

Dr A.R.D.  Stebbing, Natural Environment Research Council,
   Plymouth Marine Laboratory, Plymouth, United Kingdom

Dr J.H.M.  Temmink, Department of Toxicology, Agricultural
   University, Wageningen, Netherlands

Dr J.E.  Thain,  Ministry  of Agriculture,  Fisheries  and
   Food,  Fisheries Laboratory, Burnham-on-Crouch,  United

Prof  P.N. Viswanathan, Ecotoxicology  Section, Industrial
   Toxicology Research Centre, Lucknow, India


Mr J.  Chadwick,  Health  and  Safety  Executive,  Bootle,
   United Kingdom

Dr R.J.  Fielder,  Department  of Health,  London,  United

Dr R. Lange, Schering AG, Department of Experimental Toxi-
   cology, Berlin, Federal Republic of Germany


Dr S. Dobson, Institute of Terrestrial Ecology, Monks Wood
   Experimental Station, Abbots Ripton, Huntingdon, United
   Kingdom  (Rapporteur)

Dr M. Gilbert, International Programme on Chemical Safety,
   World    Health   Organization,   Geneva,   Switzerland

Mr P.D. Howe, Institute of Terrestrial Ecology, Monks Wood
   Experimental Station, Abbots Ripton, Huntingdon, United


    Every  effort has been  made to present  information in
the  criteria documents as accurately  as possible without
unduly delaying their publication.  In the interest of all
users  of  the  environmental health  criteria  documents,
readers  are  kindly  requested to  communicate any errors
that may have occurred to the Manager of the International
Programme  on Chemical Safety, World  Health Organization,
Geneva, Switzerland, in order that they may be included in
corrigenda, which will appear in subsequent volumes.

                      *    *     *

    A  detailed  data  profile and  a  legal  file  can  be
obtained  from  the International  Register of Potentially
Toxic  Chemicals,  Palais  des Nations,  1211  Geneva  10,
Switzerland (Telephone No. 7988400 or 7985850).


    A WHO Task Group meeting on Environmental  Health  Cri-
teria  for tributyltin compounds was held at the Institute
of  Terrestrial Ecology (ITE), Monks Wood, United Kingdom,
from  11 to 15  September 1989. Dr  M. Roberts,  Director,
ITE,  welcomed  the participants  on  behalf of  the  host
institution and Dr M. Gilbert opened the meeting on behalf
of the three cooperating organizations of the  IPCS  (ILO,
UNEP, WHO).  The Task Group reviewed and revised the draft
criteria  document and made an evaluation of the risks for
human   health  and  the  environment   from  exposure  to
tributyltin compounds.

    The  first draft of this document was prepared by Dr S.
Dobson  (ITE) and Dr  R. Cabridenc (Institut  National  de
Recherche  Chimique Applique, France).  Dr M. Gilbert and
Dr  P.G. Jenkins, both members  of the IPCS Central  Unit,
were   responsible  for  the  technical   development  and
editing, respectively.


AA      atomic absorption
BCF     bioconcentration factor
DBT     dibutyltin
EC50    median effective concentration
EEC     European Economic Community
EQT     environmental quality target
FAA     flameless atomic absorption
FMLP    formyl methionyl leucyl phenylalanine
FPD     flame photometric detector
GC      gas chromatography
GLC     gas-liquid chromatography
HPLC    high-performance liquid chromatography
IC50    median inhibitory concentration
ip      intraperitoneal
IU      international unit
iv      intravenous
LC50    median lethal concentration
LDH     lactate dehydrogenase
LT50    median lethal time
MBT     monobutyltin
MIC     minimal inhibitory concentration
MS      mass spectrometry
ND      not detectable
NOEL    no-observed-effect level
OECD    Organization for Economic Cooperation and Development
PALS    periarteriolar lymphocyte sheath
sc      subcutaneous
T4      thyroxine
TBT     tributyltin
TBTO    tributyltin oxide
TLC     thin-layer chromatography
TLV     threshold limit value


1.1.  Physical and chemical properties

    Tributyltin (TBT) compounds are organic derivatives of
tetravalent tin. They are characterized by the presence of
covalent  bonds between carbon  atoms and a  tin atom  and
have  the general formula (n-C4H9)3       Sn-X (where X is
an  anion).   The  purity of  commercial tributyltin oxide
(TBTO)  is generally above  96%; the principal  impurities
are dibutyltin derivatives and, to a lesser extent, tetra-
butyltin  and  other  trialkyltin  compounds.  TBTO  is  a
colourless  liquid with a characteristic odour and a rela-
tive  density of 1.17 to 1.18.  The solubility in water is
low,  varying between <1.0 and  >100 mg/litre according to
the  pH,  temperature, and  anions  present in  the  water
(which  determine speciation). In sea water and under nor-
mal  conditions, TBT exists  as three species  (hydroxide,
chloride,  and carbonate), which remain in equilibrium. At
pH  values  less  than  7.0,  the  predominate  forms  are
Bu3SnOH2+ and    Bu3SnCl,   at  pH 8, they  are   Bu3SnCl,
Bu3SnOH,   and Bu3SnCO3-,     and at pH values  above  10,
Bu3SnOH and Bu3SnCO3- predominate.

    The octanol/water partitioncoefficient (log Pow)  lies
between  3.19 and 3.84 for distilled water and is 3.54 for
sea  water. TBTO adsorbs  strongly to particulate  matter,
the  reported adsorption coefficients ranging  between 110
and 55 000.  Vapour pressure is low but  published  values
show  considerable variation.  There  was no loss  of TBTO
from a solution of 1 mg/litre over 62 days, but 20% of the
water was lost by evaporation.

1.2.  Analytical methods

    Several  methods  are  used for  measuring tributyltin
derivatives  in water, sediment, or  biota. Atomic absorp-
tion spectrometry (AA) is the most common. AA spectrometry
with  a flame allows  a detection limit  of  0.1 mg/litre.
Flameless  AA,  using  atomization in  an electric furnace
with  graphite,  is  more sensitive  and  allows detection
limits of between 0.1 and 1.0 g/litre   water.  There are
several  different methods of  extraction and for  forming
volatile  derivatives.  Separation of these derivatives is
commonly  done using "purge  and trap" or  gas chromato-
graphy.   The detection limits are 0.5 and 5.0 g/kg   for
sediment and biota.

1.3.  Sources of environmental pollution

    Tributyltin  compounds have been registered as mollus-
cicides,  as antifoulants on  boats, ships, quays,  buoys,
crab pots, fish nets, and cages, as wood preservatives, as
slimicides  on masonry, as disinfectants,  and as biocides
for  cooling systems, power  station cooling towers,  pulp

and  paper mills, breweries, leather  processing, and tex-
tile  mills.  TBT in antifouling paints was first marketed
in a form that allowed free release of the compound.  More
recently,  controlled-release paints, in which  the TBT is
incorporated  in a co-polymer  matrix, have become  avail-
able.   Rubber matrices have  also been developed  to give
long-term slow release and lasting effectiveness for anti-
fouling paints and molluscicides. TBT is not used in agri-
culture because of high phytotoxicity.

1.4.  Regulations on use

    Many  countries have restricted  the use of  TBT anti-
fouling paints as a result of effects on  shellfish.   The
regulations  vary in detail  from country to  country, but
most  ban the  use of  TBT paints  on boats  of  25 metres
length or less.  Some countries have excluded  boats  with
aluminium  hulls from this  ban. In addition,  some  regu-
lations restrict the TBT content of paints or the leaching
rate of TBT from paints (to 4 or 5 g/cm2 per   day, long-

1.5.  Environmental concentrations

    High  levels of TBT in water, sediment, and biota have
been  found close to pleasure boating activity, especially
in or near marinas, boat yards, and dry docks,  fish  nets
and  cages  treated  with antifouling  paints, and cooling
systems.  The degree of tidal flushing and  the  turbidity
of the water influence TBT concentrations.

    TBT levels have been found to reach 1.58 g/litre   in
sea  water and estuaries,  7.1 g/litre   in fresh  water,
26 300 g/kg  in coastal sediments, 3700 g/kg   in fresh-
water  sediments,  6.39 mg/kg  in bivalves,  1.92 mg/kg in
gastropods,  and 11 mg/kg in fish.  However, these maximum
concentrations of TBT should not be taken  as  representa-
tive, because a number of factors may give rise to anomal-
ously  high  values (e.g.,  paint  particles in  water and
sediment  samples).  It has  been found that  measured TBT
concentrations  in  the  surface microlayer  of both fresh
water  and  sea water  are up to  two orders of  magnitude
above  those measured just below the surface.  However, it
should  be noted that  recorded levels of  TBT in  surface
microlayers  may  be  highly  affected  by  the  method of

    Older  data  may not  be  comparable with  newer  data
because  of improvements in the  analytical methods avail-
able for measuring TBT in water, sediment, and tissue.

1.6.  Transport and transformation in the environment

    As a result of its low water solubility and lipophilic
character,  TBT  adsorbs readily  onto particles.  Between
10%  and 95% of TBTO introduced into water is estimated to

undergo particulate adsorption.  Progressive disappearance
of  adsorbed TBT is  not due to  desorption but to  degra-
dation.  The degree of adsorption depends on the salinity,
nature and size of particles in suspension, amount of sus-
pended  matter, temperature, and the presence of dissolved
organic matter.

    The  degradation of TBTO involves the splitting of the
carbon-tin-bond.   This can result from various mechanisms
occurring  simultaneously  in  the environment,  including
physico-chemical  mechanisms  (hydrolysis and  photodegra-
dation) and biological mechanisms (degradation by microor-
ganisms  and metabolism by higher organisms).  Whereas the
hydrolysis  of organotin compounds occurs under conditions
of  extreme pH, it is barely evident under normal environ-
mental conditions.  Photodegradation occurs during labora-
tory exposure of solutions to UV light at 300 nm (and to a
lesser  extent at 350 nm).  Under natural conditions, pho-
tolysis is limited by the wavelength range of sunlight and
by  the limited penetration of  UV light into water.   The
presence  of  photosensitizing  substances can  accelerate
photodegradation.  Biodegradation depends on environmental
conditions  such as temperature, oxygenation, pH, level of
mineral  elements,  the  presence of  easily biodegradable
organic  substances for co-metabolism,  and the nature  of
the microflora and its capacity for adaptation.   It  also
depends  on the TBTO  concentration being lower  than  the
lethal  or inhibitory threshold for the bacteria.  As with
abiotic  degradation, biotic breakdown  of TBT is  a  pro-
gressive oxidative debutylization founded on the splitting
of  the carbon-tin bond.  Dibutyl  derivatives are formed,
which  are more readily degraded  than tributyltin.  Mono-
butyltins  are mineralized slowly.   Anaerobic degradation
does  occur  but there  is a lack  of agreement as  to its
importance.  Some  workers consider  that anaerobic degra-
dation  is slow, others that it is more rapid than aerobic
degradation.    Species  of  bacteria,  algae,  and  wood-
degrading  fungi  have  been identified  that  can degrade
TBTO.  Estimates of the half-life  of TBT in the  environ-
ment vary widely.

    TBT   bioaccumulates  in  organisms  because   of  its
solubility in fat.  Bioconcentration factors of up to 7000
have  been  reported  in  laboratory  investigations  with
molluscs and fish, and higher values have been reported in
field  studies.  Uptake from  food is more  important than
uptake  directly  from  the water.   Higher  concentration
factors  in  microorganisms  (between 100  and 30 000) may
reflect  adsorption rather than uptake  into cells.  There
is  no indication that  TBT is transferred  to terrestrial
organisms via food chains.

1.7.  Kinetics and metabolism

    Tributyltin is absorbed from the gut (20-50% depending
on the vehicle) and via the skin of mammals (approximately
10%). It can be transferred across the blood-brain barrier

and  from the placenta to the fetus.  Absorbed material is
rapidly  and widely distributed among tissues (principally
the liver and kidney).

    TBT  metabolism in mammals  is rapid; metabolites  are
detectable  in blood within 3 h of TBT administration.  In
 in vitro  studies,  it  has been  shown  that  TBT  is  a
substrate  for mixed-function oxidases, but  these enzymes
are inhibited by very high concentrations of TBT.

    The rate of TBT loss differs with  different  tissues,
and  estimates for biological half-lives  in mammals range
from 23 to about 30 days.

    TBT  metabolism also occurs in lower organisms, but it
is slower, particularly in molluscs, than in mammals.  The
capacity  for bioaccumulation is, therefore,  much greater
than in mammals.

    TBT  compounds  inhibit oxidative  phosphorylation and
alter mitochondrial structure and function. TBT interferes
with  calcification  of the  shell of oysters ( Crassostrea

1.8.  Effects on microorganisms

    TBT  is  toxic to  microorganisms  and has  been  used
commercially  as a bactericide and  algicide.  The concen-
trations  that  produce  toxic effects  vary  considerably
according  to the  species.  TBT  is more  toxic to  gram-
positive  bacteria (minimal inhibitory concentration (MIC)
between  0.2 and 0.8 mg/litre) than  to gram-negative bac-
teria (MIC: 3 mg/litre).  The TBT acetate MIC for fungi is
0.5-1 mg/litre  and  the  TBTO  MIC  for  the  green  alga
 Chlorella   pyrenoidosa is 0.5 mg/litre.  The primary pro-
ductivity of a natural community of freshwater  algae  was
reduced by 50% at a TBTO concentration of 3 g  per litre.
Recently   established  no-observed-effect  level   (NOEL)
values  for  two species  of algae are  18 and 32 g   per
litre.   Toxicity  to  marine microorganisms  is similarly
variable  between species and between studies; NOEL values
are difficult to set but lie below 0.1 g/litre   for some
species.  Algicidal concentrations range from <1.5 g  per
litre to >1000 g/litre for different species.

1.9.  Effects on aquatic organisms

1.9.1.  Effects on marine and estuarine organisms

    A  summary diagram relating  lethal and sublethal  ef-
fects  to measured marine and estuarine TBT concentrations
is  presented  in Fig. 1.   Concentrations exceeding those
producing  acute lethal effects  have been found  in  many
different  worldwide  locations,  particularly  associated
with pleasure boating activity.


    The  development  of  the  motile  spores  of  a green
macroalga  was  the stage  most  sensitive to  TBT  (5-day
EC50:    0.001 g/litre).    There was reduced growth of a
marine  angiosperm at TBT concentrations  of 1 mg/kg sedi-
ment but no effect at 0.1 mg/kg.

    Tributyltin  is highly toxic  to marine molluscs.   It
has  been shown experimentally to  affect shell deposition
of  growing  oysters,  gonadal development  and  gender of
adult oysters, settlement, growth, and mortality of larval
oysters  and  other bivalves,  and  to cause  imposex (the
development of male characteristics) in female gastropods.
The  NOEL for spat  of the most  sensitive oyster  species
 (Crassostrea   gigas)   has  been  reported  to  be  about
20 ng/litre.  TBT causes deformation of the shell of adult
oysters  in a dose-related manner. No effect on shell mor-
phology was observed experimentally at  TBT concentrations
of 2 ng/litre.  The NOEL for the development of imposex in
female  dogwhelks is below 1.5 ng/litre.  Larval forms are
generally  more sensitive than adults; in the case of oys-
ters this difference is particularly marked.

    Copepods  are  more  sensitive than  other  crustacean
groups to the acute lethal effects of TBT,  LC50    values
for  exposure  periods  up to  96 h  ranging  from 0.6  to
2.2 g/litre.     These values are comparable  to those of
the more sensitive larvae of other crustacean groups.  TBT

reduces  reproductive  performance, neonate  survival, and
juvenile growth rate in crustaceans. The NOEL  for  repro-
duction  in the mysid shrimp  Acanthomysis sculpta has been
suggested to be 0.09 g/litre.   There was no avoidance of
TBT by the grass shrimp at concentrations up to 30 g/litre.

    The toxicity of tributyltin to marine fish  is  highly
variable,  96-h LC50    values  ranging  between  1.5  and
36 g/litre.  Larval stages are more sensitive than adults
(Fig. 1).   There are indications  that marine fish  avoid
TBTO concentrations of 1 g/litre or more.

1.9.2.  Effects on freshwater organisms

    A   summary  diagram  relating  lethal  and  sublethal
effects to measured TBT concentrations in fresh  water  is
presented  in Fig. 2. Concentrations exceeding  those pro-
ducing  sublethal  effects  have been  found, particularly
associated with pleasure boating activity.


    Fresh-water  angiosperms were killed by a TBTO concen-
tration  of  0.5 mg/litre,  and growth  was  inhibited  at
0.06 mg/litre or more.

    Data  on fresh-water invertebrate species are few, re-
lating  to just three species other than target organisms.
Different salts of TBT yield 48-h LC50 values  for  Daphnia
of   2.3-70 g/litre   and for  Tubifex   of 5.5-33 g/litre.
The  NOEL for  Daphnia has been estimated to be 0.5 g  per
litre, based on reversal of normal response to light.  The
24-h LC50   for the Asiatic clam has been reported  to  be
2100 g/litre,   and for target snail adults in schistoso-
miasis control the corresponding values are 30-400 g/litre.

    Tributyltin  has been shown to be toxic to schistosome
larvae  in the aquatic  stages; the LC50    (TBT fluoride)
was calculated to be 16.8 g/litre   for a  1-h  exposure.
The  TBT dose causing 99% to 100% suppression of cercarial
infectivity of mice was between 2 and 6 g/litre.

    The sensitivity of snails to TBT decreases  with  age,
but  eggs are more resistant  than both young and  adults.
Egg  laying is significantly  effected at a  TBTO  concen-
tration of 0.001 g/litre.

    The  acute  toxicity  of  TBT to freshwater fish in LC50
tests  up to 168 h ranges  from 13 to 240 g    per litre.
The NOEL for the guppy was estimated to be  0.01 g    per
litre, based on histopathological effects.

    No effect on survival was found when eggs  and  larvae
of  the frog  Rana temporaria were  exposed to TBT  concen-
trations  of 3 g/litre  or less, but at 30 g/litre  sig-
nificant mortality was observed.

1.9.3.  Microcosm studies

    Microcosm  studies  modelling  marine ecosystems  have
been conducted with introduced organisms and in conditions
where  inflowing sea water  allowed colonization by  other
organisms.   Results showed decreases  in both numbers  of
individuals  and  in  species diversity  at  TBTO  concen-
trations in water between 0.06 and 3 g/litre.

    Results  from freshwater model ecosystems suggest that
doses  which  kill  freshwater snails  also  affect  other
species, including fish.

1.10.  Effects on terrestrial organisms

    The  exposure of terrestrial organisms  to TBT results
primarily  from its use as  a wood preservative.  TBTO  is
toxic  to bees housed in hives made from TBT-treated wood.
TBT was toxic to bats in a single study, but  this  result
was  not statistically significant  owing to high  control
mortality.   TBT  compounds  are toxic  to insects exposed
topically or via feeding on treated wood. The  acute  tox-
icity of TBT to wild mice is moderate;  estimated  dietary
LC50    values, based on consumption of treated seeds used
in repellency tests, range from 37 to 240 mg/kg per day.

1.11.  Effects on organisms in the field

    Field observations have related high concentrations of
tributyltin  to mortality and settlement failure of larval
bivalves,  reduced growth, shell thickening and other mal-
formations  in developing oysters, imposex  in mud snails,
and  imposex (concurrent with  population decline) in  the
dogwhelk.  Complete failure of oyster fisheries was ident-
ified   initially  in  France  and   afterwards  in  other

countries and related to water concentrations of TBT.  The
effects were most marked in areas close to  pleasure  boat
marinas.  Controlling the use of TBT antifouling paints on
small  boats  has resulted  in  recovery of  oyster repro-
duction  and growth.  However, water concentrations of TBT
are  still high  enough in  some areas  to  affect  marine

    Both  shell growth and  chambering in Pacific  oysters
and  imposex  in dogwhelks  have  been used  as biological
indicators of TBT contamination.

    There have been few studies of the effects  on  organ-
isms  of TBT in sediment,  but there are indications  that
the  TBT is available to burrowing organisms and can cause
mortality in the field.

    Gross toxic effects and histopathological changes have
been  reported in farmed marine fish exposed to TBT by the
use of antifouling paints on retaining nets.

    The  use of TBT as  a molluscicide against the  fresh-
water  snails  that carry  schistosomiasis (bilharzia) has
been proposed. Some field trials have been conducted which
show  that it is difficult  to apply TBT without  damaging
non-target organisms.

1.12.  Toxicity to laboratory mammals

1.12.1.  Acute toxicity

    Tributyltin  is moderately to highly  toxic to labora-
tory  mammals, acute oral LD50   values ranging from 94 to
234 mg/kg body weight for the rat and from 44 to 230 mg/kg
body  weight for  the mouse.   The acute  toxicity to  the
guinea-pig and the rabbit fall within the same range.  The
variation comes from the "anion" component of  the  tri-
butyltin  salt.   These  compounds exhibit  greater lethal
potential  when  administered parenterally,  as opposed to
orally, probably due to only partial absorption  from  the

    Other  effects  of  acute exposure  may include alter-
ations in blood lipid levels, the endocrine system, liver,
and  spleen, and transient deficits  in brain development.
The  toxicological significance of these effects, reported
after high single doses of the compound,  is  questionable
and the cause of death remains unknown.

    The  acute toxicity via the  dermal route is low,  the
LD50    being  >9000 mg/kg  body weight  for  the  rabbit.
"Nose  only"  inhalation LD50    (4 h)  for the  rat  is
77 mg/m3    (65 mg/m3   when only inhalable  particles are
considered).   TBT vapour/air mixtures produce  no observ-
able  toxic effects, even  at saturation. However,  TBT is
very hazardous as an inhaled aerosol, producing lung irri-
tation and oedema.

    TBT  is severely irritating to the skin and an extreme
irritant to the eye.  TBTO is not a skin sensitizer.

1.12.2.  Short-term toxicity

    TBT  compounds have been  studied most extensively  in
the  rat (all the  data in this  section refer to  the rat
unless otherwise indicated).

    At  dietary doses of 320 mg/kg (approximately 25 mg/kg
body  weight), high mortality rates were observed when the
exposure time exceeded 4 weeks.  No deaths were  noted  at
100 mg/kg  diet  (10 mg/kg  body weight)  or  after  daily
administration of 12 mg/kg body weight by gavage.  In rats
dosed  during early post-natal  life, 3 mg/kg body  weight
resulted in increased deaths.  The main symptoms at lethal
doses were loss of appetite, weakness, and emaciation.

    Borderline  effects  on  rat growth  were  observed at
50 mg/kg  diet  (6 mg/kg  body weight)  and  6 mg/kg  body
weight (gavage studies).  Mice are less sensitive, effects
being  observed at 150 to  200 mg/kg diet (22 to  29 mg/kg
body weight).

    Structural  effects  on  endocrine organs,  mainly the
pituitary  and thyroid, have been noted in both short- and
long-term studies.  Changes in circulating hormone concen-
trations  and  altered  response to  physiological stimuli
(pituitary  trophic hormones) were observed  in short-term
tests,  but after long-term exposure most of these changes
appeared  to be absent.   The mechanism of  action is  not

    Exposure  to TBTO aerosol at 2.8 mg/m3   produced high
mortality,  respiratory  distress,  inflammatory  reaction
within the respiratory tract and histopathological changes
of  lymphatic  organs.   However, exposure  to the highest
attainable  vapour  concentration  (0.16 mg/m3)   at  room
temperature produced no effects.

    Toxic  effects on the liver  and bile ducts have  been
reported  in  three  mammalian  species.    Hepatocellular
necrosis  and inflammatory changes  in the bile  duct were
observed  in rats fed TBTO at a dietary level of 320 mg/kg
(approximately  25 mg/kg body weight)  for 4 weeks and  in
mice   fed  80 mg/kg  diet  (approximately  12 mg/kg  body
weight)  for 90 days. Vacuolization of  periportal hepato-
cytes was noted in dogs fed a dose of 10 mg/kg body weight
for   8  to  9 weeks.  These   changes  were  occasionally
accompanied  by increased liver weight and increased serum
activities of liver enzymes.

    Decreases in haemoglobin concentration and erythrocyte
volume in rats, resulting from dosing with  80 mg/kg  diet
(8 mg/kg  body weight), indicate an  effect on haemoglobin
synthesis, leading to microcytic hypochromic anaemia.  The

decrease  in  splenic haemosiderin  levels suggests alter-
ations in iron status. Anaemia has also been  observed  in

    The  formation  of erythrocyte  rosettes in mesenteric
lymph nodes has been observed in certain short-term inves-
tigations  but not in  long-term studies.  The  biological
significance  of  this  finding  (possibly  transient)  is

    The  characteristic  toxic effect  of  TBTO is  on the
immune  system; due to  effects on the  thymus, the  cell-
mediated function is impaired.  The mechanism of action is
unknown,  but  may  involve the  metabolic  conversion  to
dibutyltin  compounds.   Non-specific  resistance is  also

    General  effects on the  immune system (e.g.,  on  the
weight and morphology of lymphoid tissues, peripheral lym-
phocyte  counts,  and  total serum  immunoglobulin concen-
trations)  have been reported in several different studies
with  TBTO using rats and  dogs, but not mice,  at overtly
toxic  dose levels (effects  in mice have  been seen  with
tributyltin chloride at 150 mg/kg).  Only the rat exhibits
general effects on the immune system without  other  overt
signs  of  toxicity  and  is  clearly  the  most sensitive
species.  The NOEL in short-term rat studies  was  5 mg/kg
diet (0.6 mg/kg body weight).  In studies with tributyltin
chloride, analogous effects on the thymus were seen. These
were readily reversible when dosing ceased.  TBTO has been
shown  to  compromise  specific immune  function in rat  in
 vivo host  resistance  studies.  Decreased  clearance   of
 Listeria  monocytogenes was seen after exposure to a diet-
ary  level of 50 mg/kg (the  NOEL being 5 mg/kg per  day),
and  decreased resistance to  Trichinella spiralis was seen
at  50 and 5 mg/kg diet,  but not at 0.5 mg/kg  diet (2.5,
0.25,  and 0.025 mg/kg per day body weight, respectively).
Similar  effects were seen in aged animals, but these were
less pronounced.

    With present knowledge, the effects on host resistance
are  probably of most relevance in assessing the potential
hazard  to man, but  there is insufficient  experience  in
these  test  systems  to fully  assess their significance.
However,  some data on the significance of the  T. spiralis
model  are provided by findings in athymic nude rats after
the  standard challenge.  In  these studies, the  complete
absence  of thymus-dependent immunity resulted in a 10- to
20-fold  increase  in  muscle larvae  counts; by contrast,
exposure  to TBTO concentrations  of 5 and  50 mg/kg  diet
resulted in a 2-fold and a 4-fold increase, respectively.

    Although some data are now available from  studies  on
the  effects  of  tributyltin compounds  on the developing
immune system, there is no information on host resistance.

    It would be prudent to base assessment of  the  poten-
tial  hazard to  humans on  data from  the most  sensitive
species.   Effects on host resistance  to  T. spiralis have
been  seen at dietary levels as low as 5 mg/kg (equivalent
to  0.25 mg/kg  per  day  body  weight),  the  NOEL  being
0.5 mg/kg  (equivalent to 0.025 mg/kg per  day).  However,
the interpretation of the significance of these  data  for
human  risk  assessment  is controversial.   In  all other
studies  a concentration of  5 mg/kg per day  in the  diet
(equivalent  to 0.5 mg/kg body weight, based on the short-
term  studies) was the  NOEL with respect  to general,  as
well as specific, effects on the immune system.

1.12.3.  Long-term toxicity

    A  long-term study in rats indicates a marginal effect
of  TBT  on  general toxicological  parameters (of limited
toxicological  significance)  at  a level  of 5 mg/kg diet
(0.25 mg/kg body weight).

1.12.4.  Genotoxicity

    The  genotoxicity  of TBTO  has  been the  subject  of
extensive  investigations.  Negative results were obtained
in  the vast majority of studies, and there is no convinc-
ing evidence that TBTO has any mutagenic potential.

1.12.5.  Reproductive toxicity

    The  potential embryotoxicity of TBTO  has been evalu-
ated in three mammalian species (mouse, rat,  and  rabbit)
after  oral dosing of  the mother.  The  main malformation
noted  in rat and mouse fetuses was cleft palate, but this
occurred at dosages overtly toxic to the  mothers.   These
results are not considered to be indicative of teratogenic
effects  of TBTO at  doses below those  producing maternal
toxicity.   The lowest NOEL, with regard to embryotoxicity
and fetotoxicity for all three species, was 1.0 mg/kg body

1.12.6.  Carcinogenicity

    One  carcinogenicity  study  has been  carried  out on
rats, in which neoplastic changes were observed  in  endo-
crine  organs  at  50 mg/kg diet.   The  pituitary tumours
reported  at 0.5 mg/kg diet  were considered as  having no
biological  significance since there was  no dose-response
relationship.   These tumour types usually  appear in high
and variable background incidences, and their significance
is,  therefore, questionable.  A carcinogenicity  study on
mice is in progress.

1.13.  Effects on humans

    Occupational  exposure  of workers  to tributyltin has
been  found to result in  irritation of the upper  respir-
atory  tract.  TBT as an aerosol poses a hazard to humans.
TBTO  is a skin and eye irritant and severe dermatitis has
been  reported after direct  contact with the  skin.   The
potential  problem  is  made  worse  by  the  lack  of  an
immediate response to the skin.


2.1.  Identity of tributyltin compounds

    Tributyltins  compounds are organic derivatives of tin
(SnIV)    characterized by the presence  of covalent bonds
between three carbon atoms and a tin atom. They conform to
the  following general  formula ( n-C4H9)3 Sn-X, where X is
an  anion or a group  linked covalently through a  hetero-

    The   nature  of  X  influences  the  physico-chemical
properties,  notably the relative solubility  in water and
non-polar solvents and the vapour pressure.

    These  compounds  differ  from inorganic  tin  both in
behaviour and effects. An important member of the group is
tributyltin oxide (TBTO; RTECS number, JN8750000). Commer-
cial  TBTO has a  purity generally above  96%.   Principle
impurities  are  dibutyltin  derivatives and,  to a lesser
extent, tetrabutyl or dibutylalkyl tin compounds.

    Other  industrially important tributyltin  derivatives
include tributyltin fluoride, tributyltin methacrylate (monomer
or copolymer), tributyltin benzoate, tributyltin linoleate,
tributyltin naphthenate, and tributyltin phosphate.

2.2.  Physical and chemical properties

    TBTO is flammable but does not form explosive mixtures
with  air. It is a mild oxidizing agent. It reacts quanti-
tatively at room temperature with bromide or  iodine  with
cleavage of the Sn-O bond (a reaction that may be used for
quantitative analysis) (Bahr & Pawlenko, 1978).

    In  the presence of oxygen, light or heat, slow break-
down  occurs with the formation of tetra-n-butyltin,   di-
 n-butyltin   oxide, and eventually  tin (IV) oxide  by de-
alkylation (Evans & Karpel, 1985). This degradation may be
inhibited by the addition of 0.1-1.0% of stabilizers (such
as lactic or citric acids).

    It  has been suggested (Maguire et al., 1984; Laughlin
et al., 1986a) that TBTO in aqueous  solution  dissociates
with the formation of a hydrated tributyltin cation, which
can  undergo reaction with  anions present. Data  are  not
available  on  the  equilibrium constants  for these reac-

    Laughlin  et al. (1986a)  showed that TBTO  can  react
with normal constituents of the sea water in the following

    Bu3-Sn-O-Sn-Bu3+ HO  -  2Bu3-Sn-OH
    Bu3-Sn-OH-H+  -  Bu3SnOH2+
    Bu3-Sn-OH + CO32-  -  Bu3SnCO3- + OH-
    Bu3-Sn-OH2+ + Cl-  -  Bu3-Sn-Cl + H2O

    The  predominant  forms are  Bu3SnOH2+     and   Bu3SnCl
at pH < 7, Bu3SnCl,   Bu3SnOH,   and Bu3SnCO3- at    pH 8,
and Bu3SnOH and Bu3SnCO3- at pH > 10.

    Under normal conditions in sea water, it is considered
that  the three species (hydroxide,  chloride, and carbon-
ate) remain in equilibrium.

    The  physical and chemical properties  of some commer-
cially available tributyltin compounds are listed in Table

    Varying data on the solubility of TBTO in water, which
ranges  from < 1.0 to > 100 mg/litre at different tempera-
tures  and pH values,  may be related  to the presence  of
different anionic species as described above.

    In the same way as described in the  reaction  between
TBTO  and water, the TBT group can be transferred to other
oxygen-,  nitrogen-,  and sulfur-containing  groups. Thus,
anaerobically in sediments, TBTO can be transformed to TBT
sulfide.  With amino acids,  or their derivatives  such as
proteins,  reaction can occur  on the nitrogen  and sulfur
atoms,  and, with wood, it has been suggested that the TBT
group  may react with  hydroxylic groups (Blunden  et al.,
1984)  or form tributyltin carbonate (Smith et al., 1977).
Thus  adsorption  on  to particulate  matter could involve
chemical reaction as well as physical adsorption  or  sol-
ution.   TBTO adsorbs strongly to  particulate matter, the
reported  adsorption coefficients ranging between  110 and
55 000.

Table 1.  Identity and physical and chemical properties of tributyltin compounds
               Oxide          Benzoate    Chloride    Fluoride    Linoleate     Methacrylate  Naphthenate
               (TBTO)         (TBTB)      (TBTCl)     (TBTF)      (TBTL)        (TBTM)        (TBTN)
IUPAC name     distannoxane,  stannane,   stannane,   stannane,   stannane,     stannane,     stannane,
               hexabutyl      (benzyloxy) tributyl-   tributyl-   tributyl-     tributyl-     tributyl-
                              tributyl    chloro      fluoro      (1-oxo-9,12-  (2-methyl-1-  mono (naph-
                                                                  octadecadi-   oxo-2-propyl) thenoyloxy)
                                                                  enyl)oxy-     oxy-          derivatives

CAS name       Bis(tributyl-  Tributyltin Tributyltin Tributyltin Tributyltin   Tributyltin   Tributyltin
               tin) oxide     benzoate    chloride    fluoride    linoleate     methacrylate  naphthenate

CAS number     56-35-9        4342-36-3   1461-22-9   1983-10-4   24124-25-2    2155-70-6     85409-17-2

Molecular      C24H54OSn2     C19H32O2Sn  C12H27ClSn  C12H27FSn   C30H58O2Sn    C16H32O2Sn

Relative       596            411         325         309         568.7         374.7         ca.500

Boiling        173            ca.135      140         > 350      ca.140        > 300        ca.125
point (C)     (130 Pa)       (30 Pa)     (1300 Pa)   (extrapol)  (50 Pa)       (extrapol)    (50 Pa)

Melting        < -45         20          -16         240         < 0          16            < 0
point (C)

density        1.17-1.18      ca.1.2      ca.1.2      1.25        1.05          1.14          ca.1.1
(20 C)

pressure (Pa   1 x 10-3       2 x 10-4                            9 x 10-2      3 x 10-2      9 x 10-5
at 20 C)

Refractive     1.4880-
index (20 C)  1.4895       
    TBTO is soluble in lipids and very soluble in a number
of  organic  solvents (ethanol,  ether, halogenated hydro-
carbons, etc.).

    The  octanol/water  partition  coefficient  (log  Pow)
lies between 3.19 and 3.84 for distilled water and is 3.54
for sea water.

    As  shown  in Table 1,  the  vapour pressures  of  TBT
compounds  are low.  The work  of Maguire  et  al.  (1983)
confirmed  this directly by showing no loss of TBTO from a
1 mg/litre  solution after 62 days;  20% of the  water was
lost by evaporation.

2.3.  Analytical methods

    The  control  levels  of  contamination  of  different
environmental  compartments  (water, sediment,  biota) and
the  interpretation  of laboratory  experimental and field
study  results regarding levels, fate, biodegradation, and
bioaccumulation of tributyltin compounds require sensitive
analytical  techniques to allow identification and quanti-

2.3.1.  Measurement of organotin compounds

    These  methods, which are summarized  in Table 2, have
been  applied initially to water and later to sediment and
biota. They must be sufficiently sensitive and specific to
allow monitoring of ng/litre levels, and they need  to  be
able to distinguish between different forms of organic tin
derivatives  present in the environment,  i.e. mono-, di-,
tri-,  or tetra-butyltins and  different species of  alkyl
moieties  (butyl, methyl).  They  have also to  avoid  all
interference  from  other metals  and other organometallic

    Generally  there are four successive  stages to analy-
sis, although some are optional:

*   extraction;
*   formation of volatile derivatives;
*   separation of these derivatives;
*   detection, identification, and quantification.

Table 2.  Sampling, preparation, and analysis of tributyltin compounds
Medium      Sampling method   Sample volume     Analytical method           Detection limit   Reference
Air         adsorption on     50-100 litres     derivatization with                           Zimmerli & 
            Chromosorb,                         RMgX; GC/MS or                                Zimmermann 
            cation exchange                     GC/FPD                                        (1980); 
            resin, or Tenax                                                                   Muller (1987a)

Water                         250 ml            NaBH4 conversion to         0.1-2 ng/litre    Hodge et al. 
                                                hydride; separation by                        (1979); 
                                                fractional distillation;                      Michel (1987);  
                                                AA                                            Donard et al.    
                                                                                              (1986); Braman & 
                                                                                              Tompkins (1979);
                                                                                              Valkirs et al.  
                                                                                              (1986); Weber   
                                                                                              et al. (1986)   

Water and   extraction with   8 litres (water)  derivatization with         1 ng/litre        Maguire & 
sediments   dichloromethane   or 1 g (sediment  C5H11 MgBr; GC-FPD          (water)           Huneault (1981);              
                              dry weight)       or GC-FAA                   or 5 ng/mg        Maguire &     
                                                                            (sediment         Tkacz (1983,   
                                                                            dry weight)       1985); Maguire
                                                                                              et al. (1986) 

Water and   acidification,    1 litre           derivatization with         10 ng/litre       Meinema et al. 
biota       extraction with                     CH3 Mgl; GC-MS or AA                          (1978); Bjorklund 
            dichloromethane                                                                   (1987a)

Water,                        200 ml or         NaBH4 conversion to         5 ng/litre or     Matthias et 
biota, or                     16 litres         hydride; extraction with    0.2 ng/litre      al. (1986a,b);
sediments                                       dichloromethane                               Humphrey & 
                                                                                              Hope (1987)

Water and   adsorption on     60 litres         extraction with dichloro-   0.07 ng/litre     Humphrey & 
sediment    silica            (water) or 10 g   methane/tropolone; deriva-  (water)           Hope (1987)  
            bonded C18        (sediment)        tization with C5H7 MgBr;    0.2 mg/kg
                                                GC-MS                       (sediment)

            macroreticular    1 litre           extraction with  n-pentane   < 1 ng/litre     Muller 
            resin                               (water) diethylether        (water)           (1984)  
            adsorption                          (sediment); derivatization  0.5 mg/kg
                                                with CH3MgCl; GC-MS         (sediment)
---------------------------------------------------------------------------------------------------------  Extraction of tributyltin derivatives

    Extraction  may be independent  of or coincident  with
the formation of volatile derivatives. It is necessary for
sediments and biological tissues and can also  be  applied
in the analysis of water samples.

    Following acidification, various organic solvents have
been  used. The following are most often cited: methyliso-
butylketone,  hexane,  ethyl  acetate, toluene,  methanol,
chloroform,  dichloromethane,  and  mixtures of  tropolone
(2-hydroxy-2,4,6-cycloheptatrienone) with chloroform, ben-
zene, or dichloromethane.

    In  the case of water, liquid-liquid extraction may be
replaced  by adsorption onto  silica gel bonded  with  C18
aliphatic  chains  (Matthias  et al.,  1986a,b; Humphrey &
Hope, 1987).  Formation of volatile derivatives

    Mono-,  di-,  and  tri-butyltins are  not sufficiently
volatile to assure their separation on gas-phase chromato-
graphy; it is, therefore, necessary to prepare  more  vol-
atile  derivatives  to  allow better  separation. Two pro-
cedures have been advocated:

*   formation  of alkyl derivatives (methyl  or pentyl) by
    the  use  of  Grignard's reagent  (reactive organomag-
*   formation  of  hydrides  with  the  general  structure
    RnSnH4-n      by  reaction  with  sodium   borohydride
    (NaBH4) (Hodge et al., 1979).

    These volatile derivatives can then be extracted using
organic  solvents, such as dichloromethane, or purged by a
stream of hydrogen.  Separation of organotin derivatives

    Less sensitive methods for direct separation of mono-,
di-,  and  tri-butyltins  include high  performance liquid
chromatography  (Jewett & Brinckman, 1981)  and thin-layer
chromatography.  The latter method is only qualitative and
little used because of its low sensitivity.  Detection and measurement of different forms of organotin

    Volatile derivatives prepared in the laboratory may be
separated by two procedures:

*   separation as a function of boiling point with collec-
    tion in a cold trap ("purge and trap" procedure);
*   separation by gas chromatography.

    After separation by GLC or by the "purge  and  trap"
procedure, it is possible to detect and quantify,  at  the
ng/litre  level,  different  forms of  organotin using the
following methods:

*   a  flame photometric detector selective  for tin (FPD)
    is considered satisfactory;
*   a  flame atomic absorption (AA) spectrometer or flame-
    less  atomic  absorption  (FAA) spectrometer  using  a
    graphite  furnace  (tin  is detected  at  286.3 nm  or
    244.6 nm);
*   a mass spectrometer (MS); this is useful  for  precise
    identification  of the substance but  has limited sen-

    There  are several methods available for measuring TBT
down to detection limits of 0.2 to 5 ng/litre in water and
5  to 30 g/kg   (in tissues  of biota and in  sediments).
Some  of them can be  adapted for routine monitoring  pur-
poses.  It  is  necessary, however,  to have sophisticated
equipment  and  the  difficulty of  the  methods  requires
experienced laboratories.

    His & Robert (1980, 1985) developed a biological assay
based  on toxic effects on  larvae of the Pacific  oyster,
 Crassostrea   gigas, sensitive only above  20 ng/litre and
nonspecific  between organotin and other  toxic compounds.
Colorimetric  methods (Sherman & Carlson,  1980) have been
based  on forming coloured derivatives with phenylfluorone
(nonspecific and with a sensitivity around 0.1 to 4 g tin).

2.3.2.  Interlaboratory calibrations

    Interlaboratory  comparison of assay methods have been
performed to compare the various proposed methods  and  to
validate their usefulness as standards.

    Young  et  al. (1986)  reported  the conclusions  of a
workshop  held in the USA to examine the problems posed by
the  analysis of organotins in water.  Nine methods, based
on  the principles outlined above, were considered as sat-
isfactory, since the range of results fell within + 15% of
the  mean when the TBT  concentration was in the  order of

    Stephenson  et  al.  (1987) reported  the  results  of
interlaboratory  calibrations  conducted in  1986-1987 and
carried  out on TBT derivatives  in mussel tissues and  in
sediments.  The  measurements  were made  in seven labora-
tories, each using its own technique and  using  different
extraction  conditions,  derivative formation,  and detec-
tion.  A first examination of results showed that they did
not vary by more than a factor of 3. The results were con-
sidered satisfactory.

    Blair  et al. (1986)  took part in  an interlaboratory
calibration  exercise organised by the  National Bureau of
Standards  (NBS) in 1984 in the USA and carried out deter-
minations of TBT in water (at a concentration of 1 g/litre).

    Under  the  auspices  of  the  OECD,  it  was  decided
recently  to organize a new  worldwide intercalibration to
be carried out on:

*   water  samples  containing 10 ng/litre  each of mono-,
    di-, and tri-butyltin;
*   samples  of dried sediment  containing the above  com-
    pounds at a concentration of 100 g/kg;
*   samples of mussel tissue, frozen or freeze-dried, con-
    taining the above compounds at 100 g/kg.

    It  seems  premature  to impose  a  single  analytical
method and preferable to allow a certain freedom of choice
between  methods  to  allow sufficient  sensitivity  to be
attained.  However, control of  the competence of  labora-
tories  that carry out such difficult and complex analysis
is required through new calibration procedures.


3.1.  Uses

    Dutch scientists first recognized the biocidal proper-
ties  of triorganotin compounds  in the 1950s;  major pro-
duction  and  use  of  these  substances  dates  from this
period.  It was found that the different triorganotin com-
pounds  have different toxicities to  different organisms.
Tributyltin compounds were found to be the most  toxic  of
the  triorganotins to gram-positive bacteria and to fungi.
They were also found to have biocidal properties to a wide
spectrum of aquatic organisms.

    In  the early 1960s, both tributyltin oxide (TBTO) and
TBT fluoride were tested, mainly in Africa,  as  mollusci-
cides  against several freshwater  snail species that  are
vectors  of the disease schistosomiasis,  the snails being
the intermediate hosts of the trematode parasite. This use
led to the introduction of TBT, during the mid  1960s,  as
an antifouling paint on boats. At the same time  TBT  com-
pounds  were being registered  as wood preservatives  (the
first registration was in 1958).

    Tributyltin  compounds have been registered as mollus-
cicides,  as antifoulants on  boats, ships, quays,  buoys,
crabpots,  fish nets, and cages, as wood preservatives, as
slimicides  on masonry, as disinfectants,  and as biocides
for  cooling systems, power  station cooling towers,  pulp
and  paper mills, breweries, leather  processing, and tex-
tile mills.

    When introduced as antifouling paints, TBT paints were
of  the "free association" type, where the TBT is physi-
cally  incorporated into the paint matrix. In this form it
has  a high early release and very short life.  Co-polymer
paints were introduced later; in these the TBT  moiety  is
chemically  bonded  to  a polymer  backbone,  e.g.,  those
formed  from TBT acrylate  or methacrylate and  the corre-
sponding  acid.   The  biocide  is  released  by  chemical
hydrolysis of the organotin ester linkage.  Dissolution is
slow from ships' hulls and a low level of released TBT can
be  achieved over a  prolonged period. TBT  compounds have
also  been  impregnated  into neoprene  rubber  to produce
elastomeric  antifoulant coatings and slow-release mollus-
cicides.  In  this form,  much of the  TBT remains in  the
matrix of the rubber, though the effectiveness  lasts  for
several years.

    TBT compounds have not been suggested for use in agri-
culture because of their high phytotoxicity.

3.2.  Production

    The  world consumption of tin in 1976 was estimated to
be  200 x 103 tonnes,    of  which  28 x 103 tonnes    was
organotin.   Approximately 40% of  the total was  consumed

in  the USA (Zuckerman et  al., 1978). The United  Kingdom
Department  of  the  Environment (1986)  reported that the
worldwide  use of organotin  in 1980 was    30 x 103 tonnes.
This total was made up as follows:

*   PVC stabilizers (dibutyl), approximately 20 x 103 tonnes;
*   wood preservatives (tributyl), 3-4 x 103 tonnes;
*   antifouling paints (tributyl), 2-3 x 103 tonnes;
*   other uses of both di- and tri-butyltin, < 2 x 103 tonnes.

    The  annual world production of TBT compounds is esti-
mated  to be 4000 to  5000 tonnes (Organotin Environmental
Programme  Association (ORTEPA); personal communication to
IPCS, 1989).

    The  total  annual  use (production  and  imports)  of
organotin  compounds in Canada was reported by Thompson et
al.  (1985) to be in excess of 1 x 103 tonnes.   The total
annual  production  of TBTO  in  the Federal  Republic  of
Germany is reported to be 2 x 103 tonnes,   of  which  70%
is exported. National usage is as follows: 70% antifouling
paints;  20%  timber  protection; 10%  textile and leather
protection;  small amounts are also used as a preservative
in dispersion paints and as a disinfecting  agent.  Annual
tin  emissions are reported to  be less than 300 kg  (TWG,
1988a).  Annual TBT  use in  the Netherlands  in 1985  was
reported  to be 1.5 x 104 kg    for wood preservation  and
10 x 104 kg   for antifouling paints (TWG, 1988c). Organo-
tin  antifoulant use in Norway was 13.7 x 104 kg   in 1986
for  the treatment of nets  and sea pens at  approximately
600 fish  farms  (Linden,  1987).   In  Japan,  usage  was
estimated  at 1300 tonnes in 1987, of which two-thirds was
used for antifouling paints on vessels and  one-third  for
antifouling of nets in fish culture.

    A survey of total and retail sales  of  TBT-containing
paints  and antifouling preparations for  nets was carried
out  in Finland  in 1987.   Of a  total of  42 000 litres,
37 000 litres  were sold retail. The  concentration of TBT
in  the antifouling paints was 4-18%.  The previous use of
TBT as a slimicide or fungicide (estimated  at  2.1 tonnes
per  year during the  period 1968-1970) has  been  discon-
tinued.  The estimated sale of wood preservatives contain-
ing  TBT was 130 tonnes  in 1987; these  contained between
0.9  and 1.8% of  TBT. Champ &  Pugh (1987) reported  that
about  300 TBT antifouling paints  were registered in  the
USA in 1987, but only about 17 paints are  now  registered
for  use (US EPA;  personal communication to  IPCS, 1989).
MAFF/HSE  (1988)  listed  345 different wood  preservative
formulations,   24 surface  biocides  and  215 antifouling
paints  containing TBT with registration  approval for use
in  the  United Kingdom  under  the Control  of Pesticides
Regulations.   In 1989, the  number of antifouling  paints
containing  TBT registered for  use in the  United Kingdom
had fallen to 148, with the number of  wood  preservatives
and  surface biocides remaining about the same (337 and 26
registered products, respectively) (MAFF/HSE, 1989).

3.3.  Regulations

    In 1974, the USA set an occupational limit for organo-
tin  compounds  in  air of  0.1 mg tin/m3   (time-weighted
average). In 1979, the American Conference of Governmental
Industrial  Hygienists (ACGIH) recommended that  the occu-
pational  exposure standard for organotin compounds in air
should  be set at  a threshold limit  value (time-weighted
average) of 0.1 mg tin/m3   and a short-term TLV at 0.2 mg
tin/m3.   The Federal Republic of Germany was recommended,
in  1979, to adopt  an occupational exposure  standard for
organotin  compounds in air of  0.1 mg tin/m3,   specified
as  a  maximum  worksite concentration  (MAK).  The United
Kingdom  has also set a  recommended occupational exposure
limit of 0.1 mg tin/m3.

    A tentative acceptable daily intake (ADI) of 1.6 g/kg
per day has been adopted in Japan.

    In  December 1979, the Japanese  Government banned the
use  of  tributyltin  compounds in  certain  products  for
household  use, e.g., paint,  adhesive, wax, shoe  polish,
and textile products.

    Following the effects on the oyster industry in France
in the late 1970s, and the subsequent correlation  of  the
effects with TBT usage, the French government  banned  the
use  of TBT antifouling paints for an initial trial period
of three months, which was later extended. In 1982, paints
containing more than 3% TBT by weight were banned on boats
of < 25 m in length, although boats with  aluminium  hulls
were  excluded.  Initially the regulation only covered the
Atlantic  coast  (January  1982) but  was  later  extended
(September 1982) to the whole French coastline. All use of
organotin  compounds in antifouling paints, at any concen-
tration, is now banned in France.

    The  exception in the regulations  for TBT-based anti-
fouling  paints that many  countries have made  for  boats
with aluminium hulls is based on the fact that the copper-
based  alternative paints react  chemically with the  alu-

    In  January  1986,  the United  Kingdom enforced regu-
lations  that  prohibited the  retail  sale and  supply of
antifouling  paints with a total tin concentration greater
than  7.5% by weight in co-polymer paints (reduced to 5.5%
in  January 1987)  or 2.5%  in other  paints. These  regu-
lations  were meant to control  the use on small  pleasure
craft,  ban the sale  of "free association"  paints con-
taining high levels of organotin and set an upper limit on
organotin compounds in co-polymer paints. An ambient water
quality  target of 20 ng/litre was set. The United Kingdom
Department  of the Environment took steps to determine the
effectiveness  of the legislation by setting up a monitor-
ing  programme. Based on the results of this monitoring, a

total ban on the use of TBT paints on small boats (< 25 m)
and  fish farming equipment  was implemented in  July 1987
(Abel  et  al.,  1987). An  environmental quality standard
(EQS)  of  20 ng/litre  for  fresh  water  (covering  both
potable  water and protection of  sensitive aquatic biota)
and  2 ng/litre for sea water has been set (United Kingdom
Department of the Environment, 1989).

    The  paint industry of the Federal Republic of Germany
(FRG)  issued a renunciation in  1986 on the use  of mono-
meric  organotin  compounds  in antifouling  paints  and a
restriction  to 3.8% TBT  in co-polymeric paints.  The FRG
has  not, as yet,  issued any national  ban on TBT  marine
antifouling  paints and is  awaiting the outcome  of  dis-
cussions  on an EEC directive  (TWG, 1988b). Champ &  Pugh
(1987)  reported that both  Switzerland and the  FRG  have
banned all uses of TBT in antifouling paints in the fresh-
water environment.

    In 1987, the US EPA reviewed TBT usage, weighing risks
to the environment against benefits to users. In the mean-
time, some individual States have passed their  own  regu-
lations.  Both Virginia and  Washington State have  banned
the  use of TBT antifouling  paints on boats of  < 25 m in
length, excepting those with aluminium hulls.  Only paints
that  conform to a leaching  rate of 5 g/cm2     per  day
(steady state) can be used on boats longer than 25 m. Both
states  continued to permit  the use of  TBT paints,  with
acceptable  leach rates, in  16 oz (0.45 kg) aerosol  cans
for  use  on outboard  motors  and lower  units.  Maryland
instituted  similar restrictions but set  a lower permiss-
ible leaching rate of 1 g/cm2     per day (steady state).
Since  1985,  North  Carolina, Oregon,  and  Michigan have
instituted  restrictions  on TBT  use. California, Alaska,
New  York, and New Jersey  had TBT Bills pending  in their
respective  legislatures (Champ &  Pugh, 1987).  In  April
1988,  both the US House of Representatives and the Senate
passed  bills to restrict  the use of  TBT in  antifouling
paints.  The legislation was  signed by the  President  on
16th June 1988 and came into effect on 16th December 1988.
This  Act established an interim  release rate restriction
of 4.0 g/cm2     per day (steady state) and  a  provision
prohibiting  application of TBT antifouling paints to non-
aluminium vessels under 25 m length. Application to larger
vessels  was restricted to certified applicators only. The
outboard motor or lower drive unit of a vessel  less  than
25 m in length was exempted. A limit on  sales,  delivery,
purchase,  and receipt of TBT  paints was set in  December
1988 and a limit on use in June 1989 for  existing  stocks
of paint.

    A voluntary ban on the use of TBT compounds  for  nets
in  fish  culture  was imposed  in  1987  by the  National
Federation  of Fisheries Cooperative Association of Japan.
In 1988, the Japanese Ministry of Health and  Welfare  and
the  Japanese Ministry of International Trade and Industry

"designated" eight TBT compounds (and a further five TBT
compounds  in 1989) on  the basis of  persistence, accumu-
lation,  and  toxicity.  "Designated"  indicates that no
final decision on regulation has yet been taken  but  that
the  compounds have a  recognized hazard.  Following  this
action,  the Japan Paint Manufacturers  Association volun-
tarily  reduced the upper limit for TBT in paints to < 10%
wet weight for monomers and < 15% wet weight for polymers.
There  is  current action  to  monitor release  rates from
paint products as the next step in limiting human exposure.

    Maguire (1987) reported that tributyltin for the pres-
ervation of fish-farm nets is banned in Canada.  In  1987,
the  Canadian Department of Agriculture served notice that
antifouling  uses  of TBT  compounds  must conform  to the
following: a maximum short-term (first 14 days) cumulative
release-rate  from paint formulations of 168 g/cm2;     a
long-term  average  daily  release of  4 g/cm2;     and a
minimum hull length of 19.5 m for the use of TBT antifoul-
ing paints on non-aluminium vessels.

    In Australia, control measures on the use of TBT-based
paints  were introduced in the  States of New South  Wales
and Victoria.  TBT is prohibited for use on boats  with  a
hull  length of less than  25 m, while a leaching  rate of
5.0 g/cm2      per day was set for hulls of 25 m or more.
Aluminium vessels are not exempt from the ban.

    The  Republic of Ireland  instituted a by-law  banning
the  use of organotin compounds on boats and other aquatic
structures in April 1987 (Minchin et al., 1987).

    Norway has also prohibited use of TBT  in  antifouling
paints  except for boats longer  than 25 m and those  with
aluminium  hulls;  the  regulation became  effective  from
January 1989. There is also prohibition on the sale, manu-
facture,  and import of  paints containing TBT  without  a
specific  permit  from  the State  Pollution Control auth-
ority.  An agreement to prohibit use on nets of fish farms
has  been  concluded.  Under the  Helsinki Convention, the
Baltic States have formed an agreement on the  banning  of
TBT paints on small boats and have set up a joint monitor-
ing programme.

    The  Commission of the European Communities has made a
proposal  to the Council of  Ministers concerning restric-
tions  on  the  use  of  antifouling  paints  that mirrors
national  restrictions in member states (except that there
would  be no derogation  for boats with  aluminium hulls).
This  proposal  is  currently  being  considered  by   the
European Parliament and Council.



     Due to its physico-chemical properties, TBT introduced into
 natural waters will partly adsorb onto particles.  The  quanti-
 tative data show large variation due to differences  in  exper-
 imental  conditions  such  as salinity  and  concentration  and
 organic  content of particulate  matter. Once it  is  adsorbed,
 decrease  in  TBT concentration  takes  place mainly  by degra-
 dation.  It is known that TBT degradation rates in sediment are
 slower  than  in the  water  column, particularly  in anaerobic

     Although  abiotic  degradation occurs,  the process remains
 less important than biological action.

     Biodegradation  of TBTO in  soil and water  depends on  the
 environmental  conditions and the toxic effect of the available
 concentrations  to the organisms involved.  Hydroxylated inter-
 mediates  are formed during stepwise debutylation.  Aerobic and
 anaerobic organisms both cause biodegradation, but the relative
 efficiency  is not known conclusively. Illumination of the cul-
 tures  lowers  the  half-life, indicating  the  involvement  of
 photosynthetic organisms.

     The  lipophilic properties of TBTO contribute to bioaccumu-
 lation  in aquatic organisms, especially  molluscs.  Laboratory
 and  field studies corroborate this, although it is unclear how
 adsorption processes complicate the results. Bioaccumulation in
 all organisms studied is due, at least in part,  to  bioconcen-
 tration  from the water  phase.  Elimination takes  place  when
 organisms are no longer exposed to tributyltin compounds.

     Whether it is directly discharged into the  environment  or
 diffuses  progressively (at 1  to 10 g/cm2   per  day)  from
 coatings of the hulls of boats or nets, TBTO enters the aquatic
 environment  and  is  subject to  transformation resulting from
 physico-chemical  and biochemical processes. Speciation is out-
 lined in chapter 2.

4.1.  Adsorption onto and desorption from particles

    The effects of TBTO vary in relation to the  state  in
which the substance is present in the aquatic environment,
in  particular  whether it  is  available to  organisms in
estuaries  or sea shores. It  is important to have  infor-
mation  on its distribution  in natural waters  likely  to
have large amounts of suspended matter of  various  types.
Several  workers (Valkirs et  al., 1986; Maguire  et  al.,
1986; Randall et al., 1986; Harris & Cleary, 1987; Stang &
Seligman, 1987; Hinga et al., 1987) have conducted studies
on  adsorption and desorption of TBTO in laboratory exper-
iments,  observations in the  field, studies conducted  in
microcosms, and mathematical modelling.

    Mathematical  models  have been  developed to estimate
the  distribution of TBT in enclosed or semi-enclosed har-
bours  (Walton  et  al.,  1986)  and  estuaries  (Harris &
Cleary,  1987).  Good  agreement has  been  found  between
measured  and estimated concentrations of tin in San Diego
harbour, USA (Walton et al., 1986). The authors considered
the  results useful in  predicting levels in  ecologically
sensitive  areas of the  bay. The Harris  & Cleary  (1987)
model  was based  on the  estuary of  the River  Tamar  in
south-west  England.  This model, still under development,
aimed to reduce inputs in order to allow the model  to  be
used by non-experts and to be applicable to all estuaries.
Output  for the River Tamar  suggested that sediment-bound
tin would be distributed up the estuary by tidal influence
leading  to increased bound tin further from the open sea.
This effect would be most marked in the  summer.  Relative
to  soluble TBT, this  bound fraction does  not  currently
amount to a significant source of tin for  organisms.  The
authors  point out, however,  that this source  may become
increasingly  important as use  of TBT declines  and sedi-
ment-bound TBT represents the only available source of the

    The chemical properties of TBT, particularly its lipo-
philic character and poor water solubility, are such that,
when  TBTO  is  introduced into  water,  repartition  will
occur,  TBTO leaving the aqueous  phase and preferentially
adsorbing onto particles (Hinga et al., 1987).  Adsorption
and  desorption are dependant on  the nature of the  sedi-
ment.   Little  data  is  available  to  indicate  whether
adsorbed TBT is bioavailable.

    If this phenomenon is generally evident, its intensity
varies considerably as a function of the method  of  study
used and the measurements made.  Contradictory results are
apparent in the literature.

    Reports  from  different  authors using  various  con-
ditions  have estimated that between  10% and 95% of  TBTO
introduced  into water is adsorbed  onto particles.  There
is,  however, general agreement that  the compound remains
strongly  adsorbed.  It  has been  stated  that  sediments
remain  contaminated  for at  least 10 months; progressive
disappearance  of TBTO  is not  due to  desorption but  to

    In  an  in  situ study  of Pearl  Harbour sediment, the
rate of adsorption of tributyltin derivatives was found to
be  0.57 ng TBT/cm2   per  day (Stang &  Seligman,  1987).
There  was, apparently, no  desorption of TBTO  itself but
dibutyltin derivatives formed by degradation desorbed with
rates varying between 0.16 and 0.55 ng DBT/cm2 per day.

    Variability  in results, more evident in field studies
than  laboratory studies, is  explained by the  fact  that
adsorption  depends  on  many different  factors,  amongst
which are the following:

*   salinity;
*   nature and size of particles in suspension;
*   amount of suspended particles;
*   temperature;
*   presence of dissolved organic matter.

    Uncertainties are also evident in relation to the bio-
availability of TBT adsorbed onto sediment. Salazar et al.
(1987) considered that the effects of adsorbed  TBTO  were
partially  masked, i.e. that the  compound was unavailable
to  organisms.  This  conclusion  could  not  be  verified
regarding  effects  on  filtering or  burrowing  organisms
living in the sediment.

    It is generally agreed that part of the  TBTO  accumu-
lates  in the surface  monolayer of natural  waters.  This
TBTO will also be adsorbed onto organic matter  and  lipid
material present on the surface.

4.2.  Abiotic degradation

    A  number  of studies  have  shown that  a degradation
pathway  for tributyltin compounds exists  in the environ-
ment,  which  involves  progressive debutylation.   It  is
theoretically  completed with the liberation into water of
the tin oxide (SnO2).

    R3SnX -> R2SnX2 -> RSnX3 -> SnX4

    A number of studies have looked for evidence  of  such
degradation,  the cause and mechanisms, and an understand-
ing  of the kinetics in different environmental conditions
(Chapman & Price, 1972; Brinckman, 1981; Blunden  et  al.,
1984; Maguire & Tkacz, 1985; and Seligman et al., 1986a).

    Degradation  of  TBTO  proceeds via  splitting  of the
carbon-tin  bond, which can result from various mechanisms
occurring simultaneously in the environment. These include
physico-chemical  mechanisms  (hydrolysis and  photodegra-
dation) and biological mechanisms (degradation by microor-
ganisms  and metabolism by higher organisms). While degra-
dation  definitely occurs as  a result of  these different
mechanisms  in  laboratory  studies, it  is  necessary  to
assess the relative importance of these different pathways
to degradation of TBTO in the field.

4.2.1.  Hydrolytic cleavage of the tin-carbon bond

    Since  hydrolysis of the tin-carbon  bond of organotin
derivatives occurs only under conditions of extreme pH, it
is barely evident under normal environmental conditions.

    Studies  were carried out  in darkness and  a  sterile
medium  to  assess the  importance  of hydrolysis  in  the
degradation  of TBTO.  According to the work of Maguire et
al.  (1983) and of  Maguire & Tkacz  (1985), TBTO  remains
stable  for  11 months in  distilled  or natural  water at
20 C, in the dark, and in a sterile medium. Under various
conditions  of  pH, between  2.9  and 10.3,  these authors
found  no  change  in  TBTO  over  63 days.   According to
Seligman  et al. (1986a),  slight degradation of  TBTO was
apparent  after  94 days in  darkness  in the  presence of
formalin as a sterilizing agent.

    It  is, therefore, considered that  degradation occurs
either not at all or only very slowly in  normal  environ-
mental conditions of pH and temperature, when monitored in
the dark and in a sterile medium.

4.2.2.  Photodegradation

    Photodegradation  of  TBTO  by  ultraviolet  light  is
theoretically  possible. UV light with a wavelength longer
than 290 nm possesses an energy of 300 kJ/mol, whereas the
energy  required to break  the carbon-tin bond  is 190-220
kJ/mol. At the same time, TBTO absorbs in the UV region at
300 nm and, less strongly, at 350 nm.

    Field and laboratory measurements have shown that this
route of degradation can occur and that it  forms  deriva-
tives  of dibutyltin. These seem  to be resistant to  pho-
tolysis, since very little monobutyltin is formed (Blunden
& Chapman, 1986). While the phenomenon clearly exists, its
importance  varies considerably with different environmen-
tal conditions.  Conditions of illumination, conditions of
transmission  of  light,  and the  presence of photosensi-
tizing  substances (acetone, humic  acids, etc.) can  con-
siderably accelerate the process.

    Results  of  laboratory studies  vary considerably de-
pending on whether experiments are conducted under natural
sunlight  or UV light  of known wavelength.   According to
Slesinger & Dresser (1978), the half-life of TBTO  in  sea
water subjected to ultraviolet light is 18.5 days.  In the
presence of a photosensitizing substance, such as acetone,
the  half-life  is  3.5 days.   Seligman  et  al.  (1986a)
suggested that, under natural conditions, photodegradation
is  less important than biological action, the development
of phytoplankton leading to a partial degradation of TBTO.
Their  measurements were made  at relatively high  concen-
trations of TBTO (744 g/litre).   Under these conditions,
light  caused no degradation over  144 days.  According to
Lee  et al. (1987),  degradation of low  concentrations of
TBTO  (less than 5 ng/litre) in estuary water is increased
when  the assay is  conducted in light.  The half-life  is
between  6 and 12 days,  and the presence  of  significant
concentrations  of  phytoplankton  increases the  speed of

degradation.  According to Maguire et al. (1983), photoly-
sis under natural light conditions in distilled or natural
water is limited, leading to a TBTO half-life in excess of
89 days. Under experimental conditions of strong UV light,
degradation is apparent. At 300 nm the half-life  of  TBTO
is  1.1 days, whereas at 350 nm  it is more than  18 days.

In these assays, it is possible to demonstrate the role of
humic  acids, particularly fulvic acid, which considerably
augment  the speed of photolysis.   Under such conditions,
the half-life of TBTO falls to 0.6 days at 300 nm  and  to
6 days  at 350 nm. Under natural conditions in the port of
Toronto,  Canada, the degradation after  89 days, remained
less than 50%.

4.3.  Biodegradation

    A number of studies have been conducted to verify that
microorganisms, notably bacteria, are capable of degrading
TBTO.  In  practice, physico-chemical  mechanisms and bio-
logical  mechanisms  of degradation  overlap. Evidence for
biodegradation  constitutes  an  important element  in the
assessment of risk. Published studies of observations made
in  the field or the  laboratory have shown definite  evi-
dence  of  biological degradation  of TBTO. Biodegradation
kinetics  depend on environmental conditions  such as tem-
perature,  oxygenation, pH, the level of mineral elements,
the  presence of easily biodegradable  organic substances,
and  the nature of the  microflora and the possibility  of
their  adaptation. Biodegradation also depends on the con-
centration  of TBTO being lower than the lethal or inhibi-
tory threshold for the bacteria.

    Biodegradation is based on the formation of intermedi-
ate   hydroxylated   derivatives,  progressive   oxidative
debutylization  following the splitting of  the carbon-tin
bond.   Dibutyl derivatives are formed, which appear to be
degraded  more rapidly than  tributyl derivatives to  give
monobutyl  derivatives; these, conversely, are mineralized
slowly.   The end product may be butene. The quantities of
carbon dioxide formed remain small.  The biodegradation of
organotin compounds does not seem to involve the formation
of methyl derivatives of tin. Such methyl derivatives have
been  measured in some  studies (Braman &  Tompkins, 1979;
Guard et al., 1981; Hallas et al., 1982; Brinckman et al.,
1983),  but have been shown to be the result of the trans-
methylation  of inorganic tin  by certain marine  bacteria
( Pseudomonas ) frequently found in estuaries.

    Sheldon  (1975)  proposed  the  following  scheme  for
degradation involving microorganisms:

R3SnX -----> (R3Sn)2O -----> (R3Sn)2CO3
                                  |  UV or microorganisms
                                  |  UV or microorganisms
                                  |  UV or microorganisms

    A   mechanism  of  biodegradation  also  exists  under
anaerobic  conditions (Maguire & Tkacz,  1985).  Anaerobic
degradation  is considered to be very slow by some workers
and more rapid than aerobic degradation by others.

    Slesinger  &  Dresser  (1978) conducted  studies  in a
Warburg  respirometer under aerobic conditions  and showed
that  microflora  derived  from activated  sludge and soil
were  capable of partially degrading  TBTO.  The half-life
was  70 days, whereas  under anaerobic conditions  it  was
200 days.

    Henshaw  et al. (1978)  showed that pure  cultures  of
certain  wood-degrading fungi, such  as  Coniophora puteana
and  Coriolus   polystictus, were capable of  slowly biode-
grading  TBTO and transforming it to dibutyl and monobutyl

    Barug & Vonk (1980) studied the degradation of TBTO in
soil  but could show no  clear evidence for the  action of
microorganisms.  Under  their experimental  conditions, in
sterile  or  non-sterile  medium, the  half-life  of  TBTO
varied between 15 and 20 weeks depending on the soil type.
Barug  (1981) was not  able to isolate,  from sediment  or
soils,  microorganisms capable of utilizing TBTO as a sole
carbon  source.  By contrast,  in the presence  of  easily
biodegradable  organic  matter, biodegradation  of TBTO is
apparent  with the production of monobutyl derivatives and
smaller  quantities  of  dibutyl derivatives.  A number of
species were found to be capable of conducting such degra-
dation  aerobically  (bacteria:  Pseudomonas aeruginosa and
 Alcaligenes   faecalis ;  wood-degrading fungi:  Coniophora
 puteana,  Trametes  versicolor, and  Chaetomium  globosum ).
Under   these conditions, they observed 70% degradation in
3 weeks.  However, the breakdown  of TBTO is  not  clearly
proved since the authors showed that TBTO  accumulates  in
the cell walls of bacteria and fungi.

    Using  water  containing  natural microflora,  Olson &
Brinckman (1986) found no degradation of TBTO at a concen-
tration of 100 g/litre  and a temperature of 5 C but did
record  degradation at 28 C. Their work also confirmed an
acceleration of degradation when the incubations were con-
ducted  under light; the  authors explained this  acceler-
ation by invoking the role played by photosynthetic micro-

    Seligman  et al. (1986a) also showed evidence for bio-
degradation; in medium polluted by  TBTO at  0.5 g/litre,
the  TBTO half-life was 7 days  in the dark and  6 days in
the  light. In water containing  0.03 g TBTO/litre,   the
half-life was 19 days in the dark and 9 days in the light.
In  all cases, dibutyl derivatives  were formed and, to  a
lesser  extent,  monobutyl  derivatives. In  studies  with
14C-labelled  TBTO,  the  measurement  of 14CO2 production
suggested a half-life of between 50 and 75 days.

    Stein  & Kuster (1982) demonstrated that TBTO is elim-
inated  from waste water passing  through sewage treatment
plants  by  adsorption  onto sludge  and biodegradation by
sludge  organisms,  provided  that concentrations  of TBTO
remain less than 5 mg/litre (see also section 5.3).

    According  to Maguire et  al. (1984), the  green  alga
 Ankistrodesmus   falcatus was  capable of  bioaccumulating
TBTO  (with bioconcentration factors of 3 x 104)   when it
was  cultured in the presence  of 20 g TBTO/litre.   When
the   cultures  were  transferred  to  a  non-contaminated
medium, 50% of the TBTO was transformed to dibutyl deriva-
tives or monobutyl derivatives and even to  inorganic  tin
over the course of 4 weeks. The assays were  conducted  on
axenic cultures of algae. It may be supposed that  a  bio-
logical effect was superimposed on physico-chemical degra-
dation mechanisms.

    Maguire  & Tkacz (1985)  have shown that  in sediments
there  are oligochaetes that  are also capable  of  metab-
olizing  TBTO after it has been accumulated.  However, the
simultaneous  presence  of  bacteria in  the  test systems
means that a clear conclusion could not be reached.

    According to Maguire et al. (1986), degradation can be
characterized as follows:

*   Loss  of TBTO by volatilization is very limited with a
    half-life of more than 11 months.
*   Hydrolysis of TBTO is equally slow with a half-life of
    11 months.
*   Photodegradation  of TBTO plays a  more important role
    but  the half-life of photodegradation  is longer than
    3 months.  This  route theoretically  takes place but,
    under  natural conditions of illumination and the poor
    penetration of UV light into turbid or coloured water,
    it is inefficient.

*   Aerobic biodegradation plays a role in water and sedi-
    ment.   The half-life varies considerably according to
    conditions but is in the region of 4 to 5 months.
*   Anaerobic  degradation plays a role in water and sedi-
    ment.  The half-life varies considerably but is around
    1.5 months.

    The  kinetics of degradation of  dibutyl and monobutyl
tins  are less well  known. However, the  degradation pro-
cesses of TBTO always results in the formation  of  metab-
olites less toxic than the parent compound.

    Hinga  et  al. (1987)  indicated  a TBTO  half-life of
between  5 and 19 days  at 22-24 C in  model  ecosystems.
Thain  et  al. (1987)  suggested  half-lives of  6 days in
fresh  water and 60-90 days at 5 C in sea water. In water
and  sediment of the port of Toronto, the half-life varied
between 4 and 5 months (Maguire & Tkacz, 1985). In estuar-
ine  waters of San  Diego Bay, USA,  the half-life  varied
between  7 and 11 days  at 12 C, while  in waters of  the
Skidaway  Estuary, it varied between 5 and 9 days at 28 C
(Seligman et al., 1986a).  Stang & Seligman  (1986)  using
contaminated  sediment from San  Diego Bay found  that TBT
was degraded to monobutyltin.  The degradation kinetic was
lower  than  in  water, the  half-life being approximately
162 days.  In studies  carried out  by J.E.  Thain &  M.J.
Waldock  (Personal communication to IPCS, 1989), naturally
contaminated  sediments were maintained in the laboratory,
under  flow-through conditions, at 12 C.   Degradation of
sediment-bound  TBT was  found to  be a  slow process.  In
aerobic  layers the half-life of  TBT was between 4  and 5
months, but in deeper anaerobic layers a  half-life  value
was not obtained within 500 days.

4.4.  Bioaccumulation and elimination

    The  lipophilic properties of TBTO  and its moderately
high  octanol-water  partition coefficient  (log Pow >  3)
contribute to bioaccumulation in living organisms.

    Evidence  for  such  mechanisms and  an  evaluation of
their importance is highly relevant for hazard assessment,
both for the environment and for humans, since some of the
organisms  exposed  to TBTO  are  human food  items, e.g.,
bivalve  molluscs, crustaceans, and  fish.  Alzieu et  al.
(1980) showed that in contaminated areas tin levels in the
flesh of oysters were 100 times higher than concentrations
in the water.

    Laboratory  experiments have been conducted under dif-
ferent conditions to demonstrate such bioaccumulation, and
have shown that bioconcentration factors vary considerably
between species.

    In  estuarine bacteria, Blair et al. (1982) found bio-
concentration  factors varying between  100 and 30 000  in
species  resistant to concentrations of  20 mg TBTO/litre.
As  was  indicated  earlier, such  bioconcentration  might
result either from adsorption to the surface of the organ-
isms  or  from true  bioaccumulation  into the  cells.  In
phytoplankton, Maguire et al. (1984) reported a bioconcen-
tration  factor of 30 000 in the green alga  Ankistrodesmus
 falcatus exposed  for  1 week  to concentrations  of 20 g
TBTO/litre.  In the diatom  Isochrysis galbano, Laughlin et
al. (1986b) reported a bioconcentration factor of 5500.

    Studies on the possibility of bioaccumulation and bio-
magnification  in molluscs, particularly bivalve molluscs,
are  prominent in the literature because of human consump-
tion of oysters and mussels. Alzieu et al.  (1982)  showed
that TBTO accumulated in oysters, maintained in tanks with
panels  of antifouling paint based  on TBTO, to levels  of
25 mg/kg  (dry weight) of tissue and that this resulted in
problems  of  cavitation of  the  shell.  Waldock  et  al.
(1983), in studies of the Pacific oyster  Crassostrea gigas
exposed  for 22 days to TBTO concentrations of 0.15 g/litre
and  1.25 g/litre,   reported bioconcentration factors of
6000  and 2000, respectively.  In European oysters  (Ostrea
 edulis) exposed  to  the  same concentrations,  they found
concentration  factors of 1500 and 1000, respectively.  In
both cases, after transfer of the oysters to  clean  water
there  was a 50% fall in TBTO levels due to loss or degra-
dation.  Laughlin et al. (1986b) reported bioconcentration
factors between 1000 and 7000 for mussels  (Mytilus edulis)
exposed   for between 3 and 7 weeks to TBTO concentrations
of  23, 45, 63, 141, and 670 ng/litre. For the higher con-
centrations,  a  plateau  in  uptake  was  reached  within
2 weeks,  but  for  lower concentrations,  no  plateau was
reached  within  the  7-week experiment.  The authors con-
sidered that the mussel would be a good indicator organism
for  monitoring  marine  pollution. Cheng  & Jensen (1989)
transferred  mussels ( Mytilus  edulis ) from an unpolluted
area  into net bags suspended in a marina in Denmark. They
monitored  tin uptake and water concentrations of tin over
a period of 51 days.  Accumulation was found  to  increase
exponentially  with time for  both total tin  and  organic
tin.   Bioconcentration  factors  of 5000  to 60 000, much
higher   than  those  from  laboratory  experiments,  were
reported.  Transfer of the mussels to the laboratory after
exposure resulted in a half-time for loss of  organic  and
total tin of 40 and 25 days, respectively. Laughlin et al.
(1986b)   showed that bioaccumulation  of TBTO by  mussels
was  not significantly affected  by the presence  of humic
acids or kaolin but that the presence of  mucins  secreted
by bacteria did limit bioaccumulation. It was  also  shown
that  bioaccumulation by mussels was greater if the phyto-
plankton  used as a food organism  (Isochrysis galbana) was
also  contaminated with TBT. Contamination via food organ-
isms was more important than via the water.

    When  feeding  crabs  with the  brine  shrimp  (Artemia
 salina) containing  concentrations  of TBTO  of   6200 g/kg
wet  weight, Evans & Laughlin (1984) found a concentration
factor  of 4400. Allen et al. (1980) reported limited bio-
accumulation  (< 50)  in  a 1-week  study using freshwater
gastropods  (Biomphalaria  glabrata).  In crustaceans, par-
ticularly  the  crab  Rhithropanopeus harisii, accumulation
of TBTO from a water concentration of 0.28 g/litre   pro-
duced a moderate bioconcentration factor of 60 over 4 days.

    Bioaccumulation  of TBTO is  equally evident in  fish.
After   exposure  of  the   sheepshead  minnow  (Cyprinodon
 variegatus) for  58 days to concentrations of TBTO varying
between  0.96  and  2.07 g/litre,   Ward  et  al.  (1981)
reported a whole body concentration factor of 2600.  After
returning  the fish to clean water, loss of TBTO was rapid
over  the  first 7 days  then  slower. After  20 days, the
authors  reported a loss  of 74% from  the muscle and  80%
from  the viscera. Detection of  dibutyltin, monobutyltin,
and inorganic tin suggested possible metabolism. Bressa et
al.  (1984) exposed the mullet  Liza aurata for 2 months to
concentrations  of 5 g   TBTO/litre and  reported biocon-
centration  factors of 20 to  30 in the liver  and kidneys
but  no residues in  the muscle. After  transfer to  clean
water,  concentrations of tin fell in all organs.  Short &
Thrower   (1986)   studied   bioaccumulation   in   salmon
 (Oncorhynchus   tshawytscha)  exposed for 96 h  to concen-
trations  of  1.49 g/litre    and obtained  concentration
factors of 4300 in the liver, 1300 in the brain,  and  200
in  muscle.  Tsuda  et al.  (1987) showed  that  TBTO  was
accumulated  by carp  (Cyprinus carpio) exposed for 14 days
to  concentrations  varying between  1.8 and 2.4 ng/litre.
Over  10 days they found a plateau in uptake and a concen-
tration  factor of 1000; metabolism was evident.  Tsuda et
al.  (1986) reported concentration factors ranging between
360  and 3400 for round  crucian carp  (Carassius carassius
 grandoculis) tissues  exposed to tributyltin  chloride for
7 days.



     Levels  of TBT in water,  sediment, and biota are  elevated
 within  the proximity of marinas,  commercial harbours, cooling
 systems, and fish nets and cages treated with  TBT-based  anti-
 foulant paints.

     TBT  levels have been found to reach 1.58 g/litre   in sea
 water  and estuaries, 7.1 g/litre in fresh water, 26 300 g/kg
 in  coastal sediments, 3700 g/kg    in fresh water  sediments,
 6.39 mg/kg  in bivalves, 1.92 mg/kg in gastropods, and 11 mg/kg
 in fish.  However, these maximum concentrations of  TBT  should
 not be taken as representative, because a number of factors may
 give  rise to anomalously high values (e.g., paint particles in
 water and sediment samples).

     It has been found that measured TBT concentrations  in  the
 surface  microlayer of both sea water and fresh water are up to
 two  orders of magnitude  above those measured  just below  the
 surface. However, it should be noted that the  recorded  levels
 of  TBT in surface  microlayers may be  highly affected by  the
 method of sampling.

     Older data may not be comparable to newer data  because  of
 improvements  in the analytical methods available for measuring
 TBT in water, sediment, and tissue.

5.1.  Sea water and marine sediment

    The  concentrations of TBT  in sea water  and sediment
are  shown in Tables 3  and 5, respectively.   Many papers
have  reported an association between  increased levels of
TBT  in  water,  sediment,  and  biota  and  proximity  to
pleasure boating activity (especially marinas) and the use
of antifouling paints on fish nets and cages.  The  degree
of  tidal flushing and  turbidity of water  also influence
TBT concentrations in particular locations.

Table 3.  Concentrations of tributyltin in estuarine and sea water
                         Sample    Concentration        Detection
Location          Year   deptha    (g/litre)    Formb   limit    Reference
                        (metres)                        (g/litre)
Coastal waters    1986   0.1-0.2   <0.04        tin    0.04      Jensen & Cheng (1987)
Marinas           1986             <0.04-1.05   tin    0.04      Jensen & Cheng (1987)
Harbour areas                      0.63-2.64     OTo              ICES (1987)

Harbours          1988   0.2       0.02-0.2      TBT    0.01      Yla-Mononen (1988)

Bay of Arcachon   1982             0.1-0.3       OT               Alzieu & Heral (1984)
                  1984             0.7-1.2       tin    0.15      Alzieu et al. (1986)
                                   <0.15-0.5    OT     0.1       Alzieu et al. (1986)
                  1985             0.3-1.0       tin    0.15      Alzieu et al. (1986)
                                   <0.15        OT     0.1       Alzieu et al. (1986)
Anse de Camaret,
 Brest            1987   (1)       <0.002-0.004 TBT    0.002     Alzieu et al. (1989)
Auray river       1986-
 estuary          1987   (1)       0.009-0.069   TBT    0.002     Alzieu et al. (1989)
La Rochelle       1986-
                  1987   (1)       0.02-0.119    TBT    0.002     Alzieu et al. (1989)
Oleron Island     1986-
                  1987   (1)       0.039-1.5     TBT    0.002     Alzieu et al. (1989)
Arcachon Bay      1986-
                  1987   (1)       <0.002-0.089 TBT    0.002     Alzieu et al. (1989)

Oslo fjord                         <0.01        TBTt   0.01      NIVA (1986)

Coastal waters                     ND-0.04       TBTt             Bjorklund (1987b)

United Kingdom
Essex coast       1982   0.1-0.2   <0.03-0.9    TBTt   0.03      Waldock & Miller (1983)
South-west coast  1984             <0.04-0.35   OT     0.04      Cleary & Stebbing (1985)
South-west coast  1986   surface   0.12-5.34     OT     0.04      Cleary & Stebbing (1987)
South-west coast  1986   0.5       <0.04-1.44   OT     0.04      Cleary & Stebbing (1987)
South-west coast  1986   bottom    <0.04-2.6    OT     0.04      Cleary & Stebbing (1987)
South-west coast  1985             <0.02-0.68   TBT    0.02      Ebdon et al. (1988)
Poole harbour     1986             0.002-0.646   TBTt             Langston et al. (1987)
Essex coast       1986   0.1       <0.001-0.831 TBT    0.001     Waldock et al. (1987b)
South coast       1986   0.1       <0.001-1.52  TBT    0.001     Waldock et al. (1987b)
South-west coast  1986   0.1       <0.001-1.27  TBT    0.001     Waldock et al. (1987b)
South Wales coast 1986   0.1       <0.001-0.29  TBT    0.001     Waldock et al. (1987b)
North Wales coast 1986   0.1       <0.001-0.012 TBT    0.001     Waldock et al. (1987b)

Table 3.  (contd.)
                         Sample    Concentration        Detection
Location          Year   deptha    (g/litre)    Formb   limit    Reference
                         (metres)                       (g/litre)
Chesapeake Bay    1985   surface   ND-1.171      TBT    0.008-    Hall et al. (1986)
                         microlayer                     0.01
Chesapeake Bay
 (South)          1986   0.15      ND-0.1        TBT    0.001     Huggett et al. (1986)
San Diego Bay     1986   >0.5     0.005-0.235   TBT    0.005     Seligman et al. (1986b)
Californian coast 1986             <0.002-0.6   TBT    0.001-    Stallard et al. (1987)
San Diego Bay     1983-
                  1985   0.3-0.6   <0.01-0.93   TBT    0.01      Valkirs et al. (1986)
San Diego Bay     1983-
                  1985   (0.1)     <0.01-0.55   TBT    0.01      Valkirs et al. (1986)
USA harbours &
 estuaries               (0.5)     <0.005-0.35  TBT    0.005     Grovhoug et al. (1986)
Coos Bay, Oregon         surface   0.007-0.014   TBT              Wolniakowski et al.
                         water                                    (1987)
a   Figures in parentheses indicate distance from water bottom.
b   TBT = sample analysed for TBT and expressed as TBT.
    TBTt = sample analysed for TBT and expressed as tin.
    tin = total tin expressed as tin.
    OT = total organic tin expressed as tin.
    OTo = total organic tin expressed as TBTO.
    Alzieu  et al. (1986) monitored tin and organotin con-
centrations  in both water and oyster tissue from Arcachon
Bay, France, between 1982 and 1985. They found that levels
in  oyster tissue  decreased by  5 to  10 times over  this
sampling  period following French  Government restrictions
on  the use of TBT  in antifouling paints.  Alzieu  et al.
(1989)  monitored TBT water levels at various locations on
the  French Atlantic coast in 1986 and 1987 (Table 3), and
found that concentrations generally ranged between < 0.002
and 0.1 g   TBT/litre with the exception of a  marina  on
Oleron  Island, which had  levels of up  to  1.5 g/litre.
Levels  were highest both  in marinas and  in the  autumn,
presumably when boats were being hosed off ready  for  the
winter.  The authors concluded that levels of TBT had gen-
erally decreased since the restrictions on TBT antifouling
paints,  but in certain marinas  levels were significantly
higher,  suggesting continued use of TBT paints in contra-
vention of restrictions.

    Waldock & Miller (1983) measured TBT levels  in  water
samples collected monthly during 1982 at Burnham-on-Crouch
on the east coast of the United Kingdom. They found a rise
in  TBT  levels in  May, at a  time when boats  were being
freshly  painted with TBT antifouling paints.  There was a
second  rise in TBT water  concentrations in August, at  a
time when boats were repainted for the major sailing event

of the year.  Analysis of water samples from several areas
on the Essex coast showed that the highest levels  (up  to
2.25 g TBTO/litre)    were  associated  with the  highest
density  of pleasure craft. The authors also reported that
a  site used  by a  large number  of boats  (on the  south
coast  of  the  United Kingdom  but  situated  on an  open
coastal  site and with  less turbid water)  had relatively
low TBT levels in the sea water (< 0.08 g TBTO/litre   in
early August).

    Waldock  et  al.  (1987b) analysed  water samples from
nine  sites around the  United Kingdom coast  during  1986
following  restrictions placed on the tin content of anti-
fouling  paints  containing  TBT in  January  1986.   They
sampled  from an enclosed bay,  an open coastal site,  and
seven   estuarine  sites.  Within  these   general  areas,
locations  were found which  reflected the incoming  water
from  a river, an area fished for shellfish, and a harbour
or  marina. Half of the 250 samples taken during 1986 were
found  to equal or to be above the United Kingdom environ-
mental  quality  target level  (EQT; 20 ng/litre).  Levels
were  barely  above  the  detection  limits  at  the sites
upstream of boats. Harbours and marinas showed the highest
levels  with tidal flushing  being an important  factor in
determining amounts of TBT detected. A marina in Plymouth,
which  has poor flushing, had  TBT concentrations consist-
ently  greater  than  1 g/litre   from  May to September,
whereas  a marina in the  estuary of the River  Dart, with
good flushing, had levels of less than 0.2 g/litre.   Six
of  the nine sites exceeded the EQT by 3 to 4 times; these
were  all sites used regularly by yachts.  The other three
sites  not used by yachts all showed low but often detect-
able levels with just one sample exceeding the  EQT.   The
authors  also found increased levels of TBT close to areas
where boats were hosed down.  Other reports confirmed that
the distribution of TBT in water was associated  with  the
proximity  to intense boating activity (Cleary & Stebbing,
1985; Ebdon et al., 1988). Langston et al. (1987) reported
that  sediments,  likewise,  contained  more  TBT  (up  to
520 g    tin/kg) near marinas  than at the  harbour mouth
(20 g    tin/kg) in Poole harbour, United Kingdom.  There
was poor flushing in the harbour and sediment was not dis-
tributed;  this was reflected  in the water  levels, which
were  0.002 to 0.139 g   tin/litre in the general harbour
area and 0.234 to 0.646 g/litre in the marina.

    Cleary  & Stebbing (1987) surveyed vertical water pro-
files  in south-west England at sites already investigated
two  years before.  They did not find a systematic decline
in  concentrations  between  the two  surveys. The concen-
trations  in the surface microlayer were 1.9 to 26.9 times
higher than those at 0.5 m below the surface (see Table 3).

    Waldock et al. (1988) analysed water samples collected
in  1987 from commercial  harbours and anchorages  in  the
United  Kingdom.   Significant concentrations  were found;

several samples taken in the immediate vicinity  of  ships
had  levels  exceeding 0.05 g    TBT/litre.  However, the
highest  concentrations  were  found near  to  centres  of
yachting activity, with over 0.6 g/litre   being found at
one site. The highest concentration found close to commer-
cial vessels in harbours was 0.078 g/litre,  but this was
within 2 m of an oil tanker. A concentration of 0.25 g per
litre was recorded outside a shipyard where  a  3000-tonne
vessel was being hosed down on the foreshore, and  a  con-
centration  of  0.137 g/litre    was measured  in surface
water  close to a  vessel at anchor  in the River  Fal. In
general,  however, few samples taken in close proximity to
commercial ships exceeded 0.02 g/litre.

    Bacci & Gaggi (1989) monitored TBT and its degradation
products in harbours, marinas, and the open sea  from  the
northern  Tyrrhenian Sea, Italy.  Concentrations of up  to
3.93 g    TBT/litre were measured in the various harbours
and marinas, but no organotin compounds were  detected  in
samples from the open sea. However, considering the detec-
tion limits of the analytical technique used (0.02 g  per
litre for both TBT and DBT), levels higher than  the  NOEL
(i.e.  0.01 g/litre,    UNEP,  1989) cannot  be excluded.
From  these  preliminary  results, it  appears that, under
unfavourable  meteorological  conditions  (e.g.,  moderate
southerly   winds),  significant  quantities  of  TBT  and
related compounds could contaminate open sea sites  for  a
few days per year.

    The highest levels of TBT around the coasts of the USA
and  Denmark were also associated with marinas or harbours
used  by small pleasure  craft, with TBT  levels generally
showing  a  falling  trend from  the  inner  part  to  the
entrance (Grovhoug et al., 1986; Seligman et  al.,  1986b;
Jensen  & Cheng, 1987).   Stallard et al.  (1987) analysed
both  water  and  sediment  from  the  Californian  coast.
Highest TBT levels, up to 0.6 g/litre  water and 23  g/kg
sediment, were found near marinas.  Levels were  lower  in
other coastal areas and were lowest out in the  open  sea.
Valkirs  et al. (1986) measured TBT in surface water (at a
depth  of 0.3 to  0.6 m) and found  that, over the  period
1983-1985, TBT levels had increased in San Diego Bay, USA.
Seligman  et al.  (1989) measured  TBT in  the  waters  of
several harbours in the USA. Of the samples collected, 75%
contained TBT levels below the detection limit (< 5 ng per
litre).   The highest concentrations  were found in  yacht
harbours and near to vessel repair facilities,  with  sig-
nificant  levels being found  near dry docks.  The authors
also  found a high  degree of variability  in TBT  concen-
trations  depending on the tidal movement, the season, and
intermittent point source discharges.

    Hall  et al. (1988a)  measured TBT biweekly  for a  4-
month  period (June-September 1986) in  the Port Annapolis
marina,  Mears marina, Back  Creek, and the  Severn  River
area  of  northern  Chesapeake Bay,  USA.  Maximum concen-
trations  of  TBT were  reported  at both  Port  Annapolis

marina (1.8 g   tin/litre) and Mears marina (1.17 g  tin
per  litre)  during  early June,  followed  by significant
reductions during late summer and early autumn. The day of
the  week  (Thursday-Monday)  on which  samples were taken
during  the daily experiments  was not found  to  signifi-
cantly affect TBT concentrations. Peak concentrations were
found to occur during a rising tide.

    Balls  (1987) reported that  TBT levels in  water were
initially  (immediately after fish cages were treated with
antifoulants)  1 g/litre    (as  tin) within  fish cages,
falling to 0.1 g/litre   after 2 weeks and 0.005 g   per
litre  after 5 months. Initial concentrations  were 0.1 g
tin/litre  at a  distance of  20 m from  the  cages,  with
concentrations  in the  main body  of the  sea loch  being
< 0.028 g/litre.

5.2.  Fresh water and sediment

    Analysis  for  TBT compounds  in  the Great  Lakes, N.
America, has revealed levels often comparable, and in many
cases  higher (200 times higher in one sample), than those
measured  in  estuaries  (Maguire et  al.,  1982; Maguire,
1984;  Maguire et al., 1985; Maguire et al., 1986; Maguire
& Tkacz, 1987).  Levels of TBT in water were found  to  be
greater in the surface microlayer than in  the  subsurface
samples. For example, water samples from Ontario lakes and
rivers showed surface levels of 0.15 to 60.7 g  tin/litre
compared to subsurface levels of between 0.01 and   2.91 g
per  litre (Maguire et  al., 1982). TBT  was found in  the
Great  Lakes and in rivers  at levels up to  those causing
effects on trout in the laboratory; Maguire & Tkacz (1987)
reported  a level of  66.8 g   tin/litre in  the  surface
microlayer.  In the United Kingdom, samples of fresh water
from  near  boatyards  contained up  to 3.2 g   TBT/litre
(Waldock 1989).  In Lake Zurich and Swiss  rivers,  levels
were  found to be  much lower, i.e.  up to    0.015 g/litre
(Muller,  1987b).   Kalbfus (1988)  analysed water samples
from  marinas on Lake Constance in 1987 and 1988 and found
that TBT levels rose to a peak in May  which  corresponded
to  the boating  activity on  the lake.   For example,  at
Goren,  TBT levels rose  from 0.13 g/litre   in  April to
0.58 g/litre    in May, but by July the levels had fallen
again to 0.028 g/litre.    At the same time TBT levels in
sediment rose from 830 g/kg   in May to  2700 g/kg    in
June  and then to  3700 g/kg   in July.   Similarly, when
samples were taken on Wannsee in Berlin, levels were found
to  be 0