1. The use of chemicals to destroy pests and weeds is becoming more
and more widespread, and as it increases, so too does the fear that these
chemical treatments may be masking a number of undesirable, or even
dangerous, side-effects. The risks run by the operatives who handle the
poisons were dealt with in our first report (January, 1951),* and our recom-
mendations on measures by which these dangers could be minimized have
now been incorporated in legislation. We have also dealt, in our report
to Ministers, of May, 1953,⁺ with the possible hazards run by the person
who eats food which, at an earlier stage in its production, was exposed to
some toxic chemical. Here, too, steps have been taken to implement our
main recommendations.

2. Our present report deals with a third set of problems — the dangers
which the toxic chemicals now used in agriculture bring to the wild life, plant
and animal. of our countryside. This comes last in the list of our enquiries,
not because we or the authorities by whom we were appointed do not share
the concern to which it gives rise, but because it was immediately apparent
that the subject constitutes an extremely difficult field of enquiry. There is
certainly no lack of statement about game birds and other animals dying
after having fed in fields treated with toxic chemicals, and about unwanted,
changes in verges and hedgerows following the use of toxic sprays. The
difficulty has all along been the impossibility of measuring the extent of
the danger. We wish to make this clear at the outset so as to obviate any
impression that the presentation of our Report implies that we have been
able to give an exact scientific answer to the questions put to us. But at
the same time we also wish to say that we believe that dangers do exist, and
that even if they cannot be defined precisely, measures should be taken to
minimize the possibility of the hew toxic chemicals used in agriculture spoiling
our countryside.

4. We took these terms of reference to mean that our task was to enquire
into a1l available information on the subject, and that we were to consider
not only the dangers to which wild animals and plants were being subjected
but also the accidental hazards which the new toxic chemicals used by farmers
introduce into the lives of ordinary farm animals. We have tried to deal
with insects generally. but have had to concentrate mostly on honey bees,
because it is only about these that information is adequate. Indeed. when
we discovered how limited was the information on which we could draw.
we obtained permission to conduct a few small field experiments, so as to
discover, first, whether useful information about the collateral effects of the
toxic chemicals used in agriculture could be obtained in this way; and
second, whether our experience justified recommending larger-scale enquiries.

* Toxic Chemicals In Agriculture, H.M. Stationery Office, 1951.
⁺ Toxic Chemicals ln Agriculture (Residues in Food), H.M. Stationery Office, 1953.

8. In approaching the problem, we were fully conscious of the fact that
there is no such thing as a fixed balance of nature, and equally that every
advance that has occurred in the evolution of plants and animals has meant
a change in this so-called balance. Man’s part in transforming the face of
the earth is only the most recent contribution to a process which is as ancient as
life itself. It is, however, probably a more rapidly-acting and more far-reaching
factor than any of those which have preceded it. As our species has multiplied,
and as we have spread over the globe, forests have been laid low, and many
animal species have disappeared, or their number been so considerably
reduced that they now constitute little more than fragments of a previous
picture of animal life. This process of change has intensified progressively
over the past four centuries, and the essential question that we have had to
ask ourselves is whether the chemicals now being used in farming constitute
a more serious hazard to the fauna and flora of our land than did earlier
innovations in farming and forestry practice. Even if the public interest in
these matters is now guarded by such official organisations as the Agri-
cultural Departments and the Nature Conservancy, we need to be assured
that these new chemical pesticides do more good than harm, not only
immediately, but also in the long term, before we accord them a general
blessing.

9. Chemicals such as nicotine, derris and compounds of copper have
been in use for many years. They are all toxic to some degree, but there is
already enough experience to show that, in the way they are used at present,
they do not constitute a menace to the wild life of the country. On the
other hand, arsenic, in the form of arsenates and arsenites, which has also
been used for a very long time, is still a source of danger, especially to bees
and to farm animals which gain access to sprayed potato crops. The main
danger, however, lies with the newer pesticides such as the organo-
phosphorus compounds. The aim of the research workers, both in industry
and at research stations, who devise these substances is, of course, that they
should be toxic only to specific pests. But absolute specificity is difficult
to obtain, because many of the chemical processes which govern the living
matter of plants and animals are the same. A chemical which will kill one
kind of animal is therefore likely to kill another. Hence, in spite of the
laboratory tests and field trials which precede the release of a new chemical
for general use, it is to be expected that unforeseen and undesirable collateral
effects may come to light only during a period of full scale introduction.
or after several seasons’ use. The problem of deciding whether there may
be undesirable long-term biological effects is, however, immensely difficult.
since these have to be disentangled from those due to a variety of other
factors which also influence the natural cycles of animal and plant
populations.

Acreages Sprayed
10. The land surface of the United Kingdom is about 59 million acres, of
which in 1952 about 18 millions were arable and I3 millions were permanent
grass. In 1953 about 2.6 million acres were treated with crop-protective
chemicals of which about 2 million were treated with hormone weedkillers.
It has been estimated that if these treatments were to be used to the full
in contributing to the expansion of agriculture, a further 2¹/₂ million acres of
farmland, including l.6 million acres of cereals, should be treated. This
would make about 5 million acres in all, or about 10 per cent of our land
surface.

11. An estimate of the types and acreages of the crops actually sprayed
with various chemicals in the United Kingdom in 1953 is given in
Appendix A.

Lethal Properties of Chemicals used
l2. Generally speaking, a pesticide which is toxic to one species of plant
or animal, will be found toxic, in greater or lesser degree, to many others.
But information on this point is very limited, particularly with regard to
wild plants and animals.

13. Poisons may also act in different ways on the animal body. It is.
therefore. difficult to compare their toxicities directly. Any comparison
must also take into account the fact that some poisons are rapidly and
completely eliminated from the body, in general without producing lasting
damage, whereas others may persist or accumulate and give rise to harmful
effects after a period. The toxicity of the acute poisons which belong to
the former class is sometimes expressed in terms of the single dose which
would be expected to cause death in 50 per cent of a number of animals
submitted to test. When little experimental data are available, as for
example for the larger animals, it is usual to state the toxicity in terms of
the minimum single dose which can be expected to kill any animal of the
species concerned. These doses, which are normally expressed in milli-
grams of the chemical per kilogram of body weight, and which are referred
to as the LD 50 dose and the lethal dose respectively, may vary widely
from one animal species to another. Appendix B contains a summary of
lethal doses for various chemicals which we have considered, and is quoted
only to give some idea of the orders of toxicity of the different chemicals
mentioned.

14. The risks from acute poisons arise mainly from the possibility of
animals taking large single doses, as when an animal has eaten a considerable
amount of a recently treated crop. lt is, of course, obvious that the amount
of poison which a wild animal might consume in one period of feeding
can hardly be controlled.

15. Chemicals which lead to poisoning through ingestion over a period
of time, or which accumulate in the body, are often referred to as chronic
poisons. As the amounts necessary to produce death from single doses
of such poisons are usually very high, the risks they entail arise less from
single accidental ingestions than from the continued consumption of
contaminated material. The possible persistence of such chemicals on crops
or other material is, therefore, important. This varies with different chemicals,
according to the weather which prevails after spraying, and with other factors.
Systemic insecticides are absorbed by the plant and are, therefore, not washed
off by rain, and may disappear more slowly than other insecticides.

16. Poisons also differ in the extent to which they produce symptoms.
If symptoms are produced rapidly and the dose necessary for their production
is much below the lethal dose, the risks of fatality from accidental ingestion
are relatively small. Unfortunately with certain of the poisons with which
we are concerned this margin is very narrow, e.g. 10 mg/kg. body weight
may have little effect, whereas 15 to 30 mg/kg. may well be lethal.

17. Bearing these generalisations in mind, the evidence which we have
collected suggest the following as the main factors which in theory govern
the incidence and the extent of casualties among wild birds and mammals: —

  1. Chemical used and its method of application

(i) Organo-Phosphorus Insecticides
    These are complex organic phosphorus compounds some of which are
systemic, that is to say they are absorbed by the plant to which they are applied
and translocated in the sap, which thus becomes toxic to sucking insects.
    Because of their greater efficiency, these insecticides are tending to replace
others that have been previously used. Unfortunately, the organo-phosphorus
compounds are in the main toxic to both birds and mammals. Furthermore,
they are dangerous not only when eaten, but when they are absorbed through

the skin or eyes. The majority of the compounds at present in use have
an oral LD 50 dose of l to 25 mg/kg., although ruminants appear to be
5 to 10 times more resistant than this. The LD 50 values of malathion
and similar new compounds, that have been developed for their decreased
toxicity, range between 500 and 800 mg/kg.
    As will be seen in Appendix A, these substances are used mostly on hops,
and also on sugar beet, brassicas and strawberries. Hop spraying twice a
season has become almost a routine practice. Spraying of brussels sprouts is
usually undertaken only when a aphid population is actually building up, and
in some years this crop may not be sprayed at all.

(ii) Arsenical Sprays
    For many years fruit trees have been treated with sprays containing lead
arsenate, and with an LD 50 dose of 5 to 50 mg/kg, these sprays are highly
toxic to many animals ; for example, they are very destructive to bees. There
is, however, no evidence that they have at any time killed the birds which
visit the orchards.
    In recent years, sprays containing sodium arsenite have sometimes been
used (in place of sulphuric acid or other chemicals) for the destruction of
potato haulm, and to facilitate mechanical harvesting. There have been a
few instances of fatalities to stock which have gained access to a sprayed
crop through open gates or broken fences.

(iii) Dinitro Weedkillers
    The dinitro compounds, dinitro-cresol (DNC or DNOC) and dinoseb
(DNBP) are extensively used as selective weedkillers on cereals and peas.
Their advantage is that they kill certain important weeds such as cleavers
and corn poppies, which are resistant to the hormone weedkillers. The
LD 50 dose of DNC for birds and rodents is about 25 to 35 mg/kg. These
chemicals raise the metabolic rate of the animal and, in consequence, internal
overheating may occur which, particularly in hot weather. can be fatal.

(iv) DDT
    DDT is used very largely as a general insecticide on brassicas and peas,
and on fruit of various kinds. When tested in the laboratory it is consider-
ably less toxic to mammals and birds than are certain of the organo-
phosphorus and dinitro compounds. The LD 50 dose varies from 150 to
1,000 mg/kg, depending on the animal species and the form of the DDT.

(v) BHC
    This chemical is used very extensively on brassicas and fruit, and is about
as toxic as DDT to mammals and birds. In general its effects are similar.

(vi) Copper Compounds
    Copper compounds are used as fungicides on hops, fruit trees and potatoes.
Their toxicity varies greatly for different animal species and some farm
animals, e.g. sheep, can be poisoned by relatively small quantities. It
follows that care must be exercised in allowing grazing animals access to
herbage in orchards following copper spraying operations. A sound pre-
caution is to await heavy rain after spraying before reintroducing the animals.

(vii) Hormone Weedkillers
    Hormone weedkillers are selective, and are used more than any other
chemical on cereals and grassland; They are only slightly toxic to birds

and mammals. The LD 50 dose of those compounds in most common
use is of the order of 500 mg/kg. While the weedkillers per se are not
dangerous, their initial growth-promoting effects may render some poisonous
plants (eg. ragwort) particularly attractive to animals for a short period.
and thus lead to poisoning.

20. For the past few years occasional deaths among game, vermin and
other wild life have been reported after cornfields and orchards have been
sprayed with the dinitrocresol selective weedkillers, and after carrots and
peas have been treated with DDT. Public concern over these casualties
came to a head with a spate of letters and articles in newspapers and
periodicals in the latter half of 1952, soon after a serious outbreak of
cabbage-aphid on brussels sprouts had been ended by treatment with the
organo-phosphorus compound, schradan. The largest number of fatalities
among birds and mammals that has been reported to us occurred during
this period. In one incident, 175 corpses were found, including 19 partridges,
10 pheasants, 129 other wild birds and 7 rabbits. These incidents occurred
just before the end of the harvest, and when sprouts were the only tall
standing crop which could provide cover for both game and other wild
mammals and birds.

21. A close review has been made of all the available information about
casualties that have been reported, and it may be summarised as follows:

23. Insect populations may be affected in two distinct ways, (a) directly
by the use of insecticides designed primarily to destroy pests but incidentally
also killing natural predators and parasites, as well as useful pollinating
insects of which there are many besides the domesticated honey bees; and
(b) indirectly by the use of herbicides. The latter may result in more
profound and permanent changes than the former.

24. The danger of disturbing the balance between pest insects and their
predators and parasites has long been recognised, and although the influence
of plant protective chemicals on certain aspects of this problem has been
the subject of intensive research, far too little attention has been given
to the problem as a whole, in spite of its economic importance to cultivators.

25. Another difficulty is caused by the appearance and spread of strains of
insect which are resistant to the insecticide being used. This possibility has
been realised for some time, but it has become prominent since the discovery
of DDT-resistant flies.

27. When the chemical attack upon a particular pest also results in the
indiscriminate killing of many other species occupying the treated habitat, but
not necessarily otherwise associated with the pest attacked (i.e. as predators

or parasites), conditions can easily be created in which the natural checks
upon the populations of other species are removed, lf such species happen
to be resistant to the direct and indirect effects of the chemical used. their
populations get out of control and may rapidly reach pest proportions. A
classic example of such a change is the serious increase in fruit tree red
spider mite [Metatetranychus ulmi (Koch)] following (i) the introduction and
general use of tar oil washes in the 1920s, (ii) the use of DNC sprays in the
1930s and early 1940s, and (iii) the introduction and widespread use of
DDT and BHC sprays in the years 1946—48. The tar oil washes destroyed the
mosses. lichens and algae in which many beneficial insects hibernated, thus
removing an essential feature of the habitat complex ; the use of DNC sprays
resulted in the almost complete elimination of the few important beneficial
insects which still over-wintered on the trees; and the DDT and BHC sprays
killed most of the remaining beneficial species. In all, some forty species
that were directly or indirectly associated with the mite have been reduced
or eliminated in heavily sprayed orchards. Certain of the organo-phosphorus
insecticides, which are very efficient against the mite itself but are even
more drastic in their action on insects generally than DDT and BHC, have
aggravated rather than alleviated the problem. Many predacious insects,
e.g. Mirids and Anthocorid bugs, are partially herbivorous. and are killed
by these sprays. Some degree of specificity may be obtained by the use of
systemic insecticides that enter the plant sap, but there is some evidence
that these too may effect beneficial species. The ideal insecticide would be
one that killed only the pest and none of its associates. Some of the most
recently developed acaricides (mite-killers) approach this ideal, and are
already being widely used in commercial orchards.

28. The problem in an orchard - a perennial crop - is not the same as that
in a field of annuals. The fruit trees and the plants beneath them form
group of permanent habitats for a number of species which, if their popula-
tions are totally destroyed, may not recolonize the treated habitat for some
time. This may allow the quick build-up of a pest such as the red spider
mite. But surrounding hedgerows provide what is probably a very important
natural reservoir of parasites and predators, and so long as there are sufficient
reservoirs, properly distributed, it would be difficult to cause permanent
damage to the population of beneficial insects on perennial crops by the use
of insecticides on the crops alone. lt would therefore appear very unwise
in the present state of our knowledge, to interfere with the hedgerows in the
vicinity of sprayed perennial crops.

29. With annual crops, grown under clean cultivation and probably form-
ing part of a rotation, no permanent habitat is formed, and nearly all the
insects found on crop plants in farms and market gardens, as well as orchards
have alternative wild host—plants. Since there is still a high proportion of the
land in this country under wild plants, and since these host—plants are mostly
well distributed over the countryside, spraying cultivated crops with chemicals
is not likely to result in any permanent widespread changes in the number of
species and in species distributions, although there may be local reduction
on the sprayed areas. But such treatment may interfere with pollinating
insects and cause harm by killing those that collect pollen and nectar for
example, pollen beetles may assume such pest proportions as to necessitate
spraying, yet they have on occasion been shown to improve pollination of
certain brassica seed crops. Hedgerows, again, may be an important reservoir
of insects-predatory, parasitic and pollinating-beneficial to annual crops
and the wholesale removal of these natural reservoirs may have far reaching
results.

30. Honey bees are dealt with separately below (para. 44 et seq).

32. lf certain plants are eliminated from a habitat, it will no longer be
able to support populations of any insects that are specifically dependent on
the presence of those plants; equally, an increase in the populations of other
plants (e.g. cereals and grasses) will favour population growth among insects
that are cereal and grass—dependent. It is also obvious. but less widely
recognised, that changes in the vegetation structure, and consequently in the
physical conditions, will bring about further considerable changes in the
fauna. What is not known is how great or important to cultivators these
changes may be.
    In addition, the problem must be visualized in terms of the nature of the
habitats treated. When used on farm or market garden crops, weedkillers will
reduce the food supply, and therefore the numbers, of pollen and nectar
feeding (pollinating) species visiting or living in the treated area; but prob-
ably no more so than any other method of clean cultivation. If, however,
similar treatment is extended to marginal habitats. such as hedgerows, much
of the value of these reservoirs is likely to be destroyed because of the
secondary changes brought about in the populations of predators, parasites.
and other beneficial species.

33. But insect numbers also fluctuate naturally, so that without greatly
increased knowledge and precise controls we cannot safely relate effect to
cause.

34. Observations on the use of DNC as a weedkiller on winter wheat.
made at our request, showed a slight decrease in the numbers of insects in
the air over the treated field on the day after spraying; but more elaborate
experiments of longer duration would be necessary in order to detect with
certainty possible differences between the populations in sprayed and un-
sprayed fields.

35. We have received evidence which suggests that after repeated applica-
tions, some of the more stable plant-protective chemicals may accumulate in
the soil. The resultant effect on the soil fauna is a subject for research, and
we are satisfied that the importance of the problem is recognised by
biologists.

36. Finally, it must once again be emphasised that with the exception
of work done on the fauna of fruit trees the amount of factual information
available is very meagre and that the prime necessity is for more work to
be done on all aspects of this subject. For example, when a new insecticide
or herbicide is introduced, or an existing one is used in a different way, it is
important to study carefully the effect not only on the noxious insects but
on all the predators, parasites, pollinators and neutral insects and other
animals involved, indeed to make an ecological survey of the " field
complex ".

C. FISH
37. Fish are very susceptible to contamination of the water in which they
live, and some work has been done to establish the lethal doses of many
commonly used chemicals. But although fish are so sensitive, the large
volume of water in which they live usually precludes the possibility of their
being poisoned by spray drift, or even by the accidental spraying of the
water with the concentrations normally used in agriculture. It is reported
from America, however, that fish mortality may be high when heavy rainfall
washes insecticides from a sprayed area into streams. It is doubtful if the
rainfall under British conditions would be sufficient to provide the run-off
necessary to produce a toxic effect.

38. We have received no evidence that plant protective chemicals, when
properly used, have affected fish either in streams or ponds. It is known.
however, that considerable damage has been caused when drums which had
contained concentrated chemicals had been washed out in a watercourse, or
when a large volume of spray not required for a field had been allowed to
run to waste.

39. We have examined the legislation pertaining to the prevention of
pollution of rivers, and it is clear to us that anyone emptying surplus poisonous
material or washing out containers of poisonous materials in a stream may
commit an offence under one or more Acts of Parliament, and may also render
himself liable to proceedings at common law.

Dangers to Domesticated Fauna

A. MAMMALS AND BIRDS

40. Some instances of losses of sheep and cattle have occurred. In one
incident which occurred in l953, 23 in-calf cows broke into a field of potatoes
which had been sprayed with sodium arsenite. Fourteen of the cows died,
mostly between 4 and 10 days after eating the sprayed haulm; 3 of them
cast their calves before they died. In another incident in 1952, 60 sheep
strayed into a field of sugar beet which had been sprayed with an organo-
phosphorus compound ; 49 of them died.

41. Since the movements and diet of farm animals are theoretically within
the control of the owner, the risk of casualty should be negligible, and human
negligence or inability to appreciate and to follow clear instruction and advice
has been the cause of those accidents which have come to our notice.
Negligence can consist of failure to dispose of partly emptied chemical
containers; to remove contaminated soil in farmyards; to refrain from
spraying when spray would drift into adjacent fields; to ensure that spraying
machines are filled in the field being treated, in order to avoid spillage of
spray liquid in other fields; and to maintain fences or close gates. But a
contributory cause may well be the natural curiosity of animals which, simply
for that reason, wander into a field where spraying is taking place, or has
just taken place.

42. While the chemicals used for crop protection are approximately as
toxic to domestic animals as they are to man, the method of poisoning is
different. For man. the greatest risks are from continued inhalation of air-
borne particles by the operator or from absorption through the skin; with
animals it is from consuming food which has been contaminated by spray
or spillage. Lethal doses of chemical may be ingested by farm stock from
several square yards of crop sprayed at normal concentration, especially
with the organo-phosphorus compounds. Thus a real responsibility rests
with the individuals supervising spraying, and with the managers of livestock

in the vicinity of spraying operations. Most spraying with toxic chemicals
is done on a contract basis by large companies whose employees are fully
aware of the hazards to which they and animals may be exposed. Since the
risks to animals arise from human negligence, it is important that all those
concerned should be obliged to take the necessary precautions, remembering
that farm animals may, apparently out of curiosity, wander into an area
which is being sprayed.

43. In our second report we drew attention to the potential hazards to
humans arising from the use of rodenticides in food stores. Apart from
the rodent control operations of local authorities, servicing companies and
County Agricultural Executive Committees, it is becoming a practice for
farmers to put down treated baits, particularly warfarin-treated baits, to
control rodents. All rodenticides are toxic, to a greater or lesser degree,
to mammals and birds, which means that care must be exercised when laying
baits to ensure that domestic animals will not be able to get at them. The
bodies of poisoned rats should also be removed as frequently as possible,
since animals eating the bodies of rats poisoned with certain rodenticides may
themselves be poisoned. While it is true that modern rodenticides such as
warfarin, which is an anti-coagulant, reduce the hazards to livestock as
compared with those arising from most of the older rat poisons, accidental
poisoning of cats, dogs and occasionally pigs, has occurred with warfarin,
despite the fact that several doses need to be consumed over a period before
toxic effects are produced. Larger farm animals and poultry are relatively
resistant to this type of chemical, and it is very unlikely that these species
could be poisoned under normal rat-baiting conditions. The advertising
and labelling of rodenticides should not convey the impression that rodenti-
cides of any kind are harmless to animals other than rats.

B. HONEY BEES

44. The average annual retail value of the honey produced by bees in
this country is Ł1·4 million. In addition, bees play an essential part in
agriculture and horticulture as pollinators.

45. The loss of honeybees because of the use of toxic chemicals in
agriculture is no new problem. But arsenical sprays used to be the only
lethal substances of any significance which the beekeeper had to worry
about. Now we understand that losses from chemical poisoning are more
than twice those occurring before the introduction of the newer crop protective
chemicals. The incidence of loss appears to fall most consistently and
heavily in such areas as the Holland division of Lincolnshire, where brassica
seed crops are grown extensively, and where chemical weed control of cereals
is commonly practised; and in parts of Cambridgeshire, where losses from
arsenical poisoning have been regularly confirmed. Even if the colonies
recover, sudden losses of the foraging bees in an apiary mean the loss of
the honey crop for the season, and a corresponding reduction in the number
of bees available as pollinators in the locality.

46. During 1948-50, thirty-two beekeepers in the Holland division of
Lincolnshire were interviewed, and details of damage, if any, from sprays
were noted. In that period chemical poisoning probably accounted for the
complete loss of 12 colonies of bees, and severe damage to 104. DDT and
BHC dusts were indicated as the cause of the majority of these losses. A
similar survey in 1951 recorded 50 colonies in one apiary affected by arsenical
poisoning ; 15 colonies in another apiary severely affected by BHC ; and less
severe losses in other instances where poisoning was indicated. but not traced
definitely to a specific spray or dust.

47. Sixty-three cases of suspected poisoning of bees reported to the
National Agricultural Advisory Service at Rothamsted Lodge in l948 were
investigated in detail. Only 30 could be confirmed on the evidence available.
These involved 189 colonies, of which 131 were damaged by arsenical
compounds. Comparable losses have been confirmed annually since l948.
arsenic being responsible for a high proportion.

48. The application of toxic chemicals to open fruit blossom has been
responsible for much loss, ranging from a high proportion of the workers
of a colony to the slow but complete extinction of the whole colony. The
spray falling on to flowering weeds under the orchard trees, and on to moist
patches which the bees use as watering places, has also contributed to the
loss.

49. The poisoning effects of chemicals on bees may be divided into:

51. Among weedkillers, concentrated sprays containing DNC are respon-
sible for most damage to bees. Although of relatively low toxicity as
compared with DNC, the hormone weedkillers, because of their widespread
use, have an important damaging effect by seriously reducing the succession
of weed flowers which provide nectar and pollen. DNC and DNBP, though
highly toxic to bees, do not affect them when used as winter washes in
orchards, for at that time the insects are not foraging outside the hive.

52. Among the chlorinated hydrocarbon insecticides, BHC is certainly
the most dangerous. Chlordane and DDT are also toxic, although the latter
has not been proved in practice to be the menace it was expected to be
when first introduced. Toxaphene, a member of the same group of chemicals.
is reported to be safer than either BHC or DDT.

53. Among the organo-phosphorus insecticides, parathion is the most
dangerous to bees; however, it is not very widely used as a spray, so that
casualties to bees from its use are not likely to be high. The systemic
insecticide schradan is much less dangerous, but more widely used. It is
too early yet to assess the extent of its danger to bees. It may appear
unchanged in the nectar of the flowers of sprayed plants, but so far it has
not been found in amounts lethal to the honey bee.

54. Lime sulphur and the copper and mercury fungicides are compara-
tively harmless to bees.

55. The same toxic material applied in different ways, or under different
conditions, may have very different effects. In general, sprays containing
oil are more dangerous to bees than those with little or no oil, while
emulsions are more persistent than sprays made with wettable powders.
These, in turn, are more persistent than dry dusts. High volume spraying
with DDT, but not with other chemicals, tends to be less dangerous than
low-volume spraying, probably because less oil stays on the bee. The danger
of aerial spraying of DDT is probably due to the high oil concentration
in the spray used.

56. Dusts may be very harmful quite apart from the nature of the active
ingredient involved, for some of the fillers may themselves be damaging.

57. Research on the effect of crop protective chemicals on pollinating
insects has been carried out at Seale Hayne Agricultural College, in collabora-
tion with Rothamsted Experimental Station, and is being continued. Studies
are being made on the effect of dusts on the lipoid waterproofing mechanism
of the Hymenoptera ; on the contamination of nectar by systemic insecticides ;
and on the use of repellent chemicals.

58. Bees are not the only insects which may be concerned in pollination.
The pollen beetle is a pest of brassica seed crops which must often be con-
trolled by spraying ; however, it has been found on occasion that too effective
a control (95 to l00 per cent) has led to reduced yields of seeds of these
crops.

59. The chemical analysis of the bodies of bees suspected of having been
poisoned by toxic sprays or dusts is very difficult except in the case of
arsenic. In addition, there is the physical difficulty of ensuring that the
dead bees are available quickly enough to the analysts, before the chemical
nature of the poison is transformed.

60. The Ministry of Agriculture, Fisheries and Food issues a Press Notice
annually, appealing to fruit growers to exercise care in the application of
orchard sprays, so as to protect bees. The text is reviewed each year to
keep it in line with current developments in the use of new insecticides. The
notice receives publicity in the farming, horticultural and beekeeping Press.
and through the B.B.C. The problem is also discussed at conferences and
meetings throughout the country. If we compare the losses of bees in this
country with those occurring on the Continent, it is reasonable to suppose
that this publicity has had a beneficial effect. It has, however, been suggested
to us that the general propaganda should be further reinforced by local
articles and appeals, since districts vary enormously in crops and conditions.

Dangers to Flora

61. Although much is known of the relative sensitivity of many plants to
the different chemicals now in use, and though it is possible to make an
estimate of the effect of the continued application of these chemicals, their
cumulative effects cannot yet be judged either from experimental results or from
ecological studies. With so little direct evidence to guide us, we have accord-
ingly sought the advice of a number of eminent botanists about the possible
effects of the increasing use of selective weed killers on crops and grassland.
But it must be emphasised here that the possible direct effects on plants
dealt with below cannot be isolated from the secondary changes these induce
in the populations of animals, more particularly insects, that are intimately
associated with the affected plant communities (see para. 22 et seq).

62. For a cultivated crop the problem appears comparatively simple. The
good farmer must get rid of weeds which reduce productivity, and the new
weedkillers are frequently cheaper and more effective than machine cultiva-
tions, hand hoeing, thistle cutting and the like. The steady reduction in
the weeds of cultivated crops, to which many writers have referred*, is as
much the result of this technical change as it is of improvements in grain
cleaning and seed dressing, which prevent these weeds from being regularly
resown with the grain. Even so, not all undesirable weeds are killed by
existing herbicides, and some species, such as wild oats, might be expected

* Downs and Dunes. SALISBURY, E. J. 1954, p. 123.

to increase, because of reduced competition with other weeds. Under any
conditions the crop " community" in an arable field is highly artificial, and
there are no grounds, historic or scientific, for its preservation as a plant
community.

63. The problem of the plant populations of hedges, borders of fields (head-
lands) and roadside verges is more complicated. Here we are dealing with
semi—natural communities which have been maintained in a more or less
stable condition for a long time. Any radical alteration in the composition
of these communities, such as might be induced by spraying with herbicides
(or even by reduction of the rabbit population), would lead to a new ecological
situation, and, if the treatment were maintained, to a new equilibrium. On
the slight evidence available, it is difficult to forecast the nature of the new
community. But a shift in the balance would be expected which would favour
the grasses and lead to a reduction or elimination of some dicotyledonous
species, including possibly some kinds of shrubs. These vary considerably
in their degree of sensitivity. For example, creeping buttercup is more
sensitive than bulbous buttercup to the action of 2-4D, and it is not outside
the bounds of possibility that communities which developed under a regime
of spraying might still control a fair proportion of resistant dicotyledonous
species, some of which would be classified as weeds. The process might
well result in the selection of resistant strains of species, the individuals of
which are at present mostly susceptible to control.

64. Thus, the net result of spraying might be to produce roadside and
hedgerow communities which were different from the old, but which still
contained species which were undesirable both from the point of view of
the agriculturist and of the road engineer, while removing many harmless
and even beneficial species, such as those which contribute to the food chain
of insects and birds useful to agriculture.

65. So far we have considered only the intentional spraying of fields
and verges. But there is another important factor, the possible effect of
the drift of spray on to uncultivated ground. Operators are generally very
careful to guard against spray drifting on to an adjoining crop or garden.
since they may become liable for any damage resulting from carelessness of
this kind. Precautions are perhaps not as stringent when spray drifts on
to common or down-land. Drift can however cause damage to susceptible
species at distances of up to a mile, so that, if spraying is repeated annually.
drift (e.g. on to down-land) might affect highly sensitive species and lead
to their ultimate elimination.

66. Such damage is unlikely to be economically serious, except indirectly
as it may affect food plants of beneficial species, since grasses are the only
usable important plants on uncultivated land, and, any change in the plant
community is likely to be to their advantage. Spray drift on to forests or
even hedgerow trees might cause damage, but is probably limited by the
fear of civil action referred to above.

67. The wild flowers of the countryside are an essential part of the land-
scape in which we live, and there would be general and understandable
resentment if they were reduced substantially by the cumulative effect of
the herbicides that are now applied to grass or scrub. We must guard
against any such eventuality, however unlikely it may be.

68. A number of reports have come to our notice about the effects of
spray drift on areas adjacent to those under treatment. The following
exceptional case will be of general interest. Tomatoes that were being grown

in glasshouses at an experimental station were all found to have been affected
at a time when a strong smell of hormone weedkiller had been noticed.
The two trusses which were setting at the time produced seedless hollow
fruit and there was a temporary loss of vigour in the plants and a dis-
tortion in growth. No spraying had taken place nearer than one mile.
While it should be stressed that tomatoes (and also rhubarb) are particularly
sensitive to hormone weedkillers, this incident emphasises the potency of
these chemicals, especially if applied through a low volume sprayer.

69. We were pleased to hear of the experiments carried out jointly by
the Nature Conservancy and Agricultural Research Council into the effect
of spraying roadside verges, and also about the long term experiment on
a N.A.A.S. Experimental Husbandry Farm, where the effect of using three
types of weedkillers in an arable rotation with a high percentage of cereal
crops is being studied. As we have tried to indicate, we are sailing in
unchartered waters, and we do not yet know precisely what shores we shall
reach. Experiments of this kind are therefore vital.

72. In both the DNC trials the only dead animals found were four
leverets and two skylarks. Wild life was much more sparse where the
schradan trial took place, and only the bodies of a yellow hammer and a
thrush were found.

73. A statistical examination of the population counts made in the second
DNC experiments showed that the two most common species of bird which
lived and fed in the fields, larks and partridges, decreased in numbers after
the area had been sprayed, but that only the decrease in the number of

larks exceeded that to be expected on the basis of normal fluctuations.
Although many hares were seen, their appearance was so sporadic that
statistical investigation was valueless.

74. It must not be concluded from this result that the effect of the DNC
was to kill the partridges which were missing, because we believe that, had
this been so, at least some of the bodies would have been picked up. It
seems more likely that the chemical rendered the wheat less attractive to the
birds, which, consequently moved to adjacent unsprayed areas. Similarly,
the adult hares may well have moved their feeding territories, whereas the
half-grown hares, which are much more likely to remain close to their form,
may have eaten enough of the sprayed wheat to succumb. (Suckling hares
are not likely to ingest sufficient of the chemical to be poisoned.)

75. It was intended to carry out both a field trial and field searches of
brassica crops after they had been sprayed with an organo-phosphorus
insecticide to kill aphids. There were, however, very few aphid attacks
on brassica during the summer of 1954, and it was impossible to do more
than search three fields, totalling 37 acres, which had been sprayed with
schradan. These fields were in market garden areas and not particularly
suitable for our purpose because the wild life population was low.

77. In the first instance it seems that the inhalation of lethal doses either
as dust or vapour is a remote risk to wild animals. The absorption through
the skin of lethal doses, even of the organo-phosphorus compounds, is also
considered likely to be negligible at the concentrations used in practice,
especially when the protective covering of hair or feathers is borne in mind.
It follows therefore that mammals and birds die because they ingest the
toxic chemicals on crops, or because they eat other creatures that have been
poisoned by these chemicals.

78. Field observations show that most casualties occur immediately after
spraying; and this can be related to the fact that there is often a marked
reduction after the first three or four days in the amount of the chemical
remaining on or in the vegetation. Although it is true that some arsenical
compounds and DDT persist and may have a cumulative effect, in practice
if a creature does not absorb a lethal dose within the first few days it is
unlikely to be subjected to a further dose of the same chemical in the same
season. This means that if some animal which has eaten contaminated
vegetation is not poisoned within a few days after spraying it is likely to
survive, although it may be weakly, and therefore more liable to suffer from
predators. The risk of poisoning clearly depends on the amount of chemical
ingested during those first few days while the chemical persists on the
vegetation.

79. It could, therefore, be argued that it should be possible to minimise
casualties by spraying only in certain weather conditions. As the only
object of the spraying is to kill specific pests, or unwanted plants in a

cultivated crop. spraying has, however, to be done when it will have the
best practical effect. Compromise measures are clearly called for. and to
get the best of both worlds the following general precautions should be
considered.

        (viii) evasive action by the beekeepers, i.e. moving colonies away from
    crops to be treated, or closing hives temporarily when crops are being
    sprayed.

    The above precautions relate specifically to domesticated honey bees. but
several of them might also reduce losses among valuable wild pollinating
insects.

    (i) The total number of casualties to wild birds and mammals that are
caused by spraying during an average season in Britain is not at present
high, and direct mortality from the use of toxic sprays is very low indeed
compared with other causes of death. This conclusion is supported by the
results of certain field trials which we initiated.

    (ii) The sprays most likely to be harmful to wild birds and mammals, in
order of danger are as follows: - organo-phosphorus insecticides applied to
brassicas in late summer ; arsenical compounds used for potato haulm destruc-
tion in September; dinitro weedkillers applied to corn and peas in Spring
and July; and DDT insecticides applied to orchards, carrots and peas.

    (iii) Incidents such as those reported in 1952, in which large numbers of
corpses of birds and mammals were found in fields after they had been
sprayed with organo-phosphorus insecticides, have not been; reported in
either 1953 or 1954, in spite of widely publicised requests for information.
In our view, the more numerous deaths in 1952 may well have resulted from
the intensity and extent of spraying which was necessary to control a severe
and late attack of aphids on brussels sprouts, when there was little alternative
cover for birds and mammals. The incidence of aphids on brassicas in 1953
was light, and in 1954 almost non-existent. It would therefore appear
reasonable to suppose that the recurrence of incidents such as those in 1952
will be limited to years in which aphid attack is heavy and late.

    (iv) Dinitro weedkillers, such as DNC and dinoseb, if used with proper
care and as early as possible in the season, will not cause heavy casualties
among birds and mammals.

    (v) The hormone weedkillers, which are of low toxicity, require no special
precautions to ensure the safety of birds. In certain circumstances, however,
farm livestock may be in danger since poisonous plants such as ragwort may,
for a period, become very appetising.

    (vi) The arsenical sprays, although very toxic to birds and mammals, do
not cause casualties when used on fruit trees ; when used to kill potato haulm,
special care should be taken to see that fences and gates are secure, so as
to prevent farm stock straying into treated fields, and to see that spray liquid
is not spilled in fields other than that being treated.

    (vii) DDT is of low toxicity, but may cause casualties among birds when
they feed persistently on a sprayed area, particularly in the absence of rain.

    (viii) It is important that it should be realised that domestic animals and
birds can be accidentally killed by spraying operations. and that both farmer
and contractor should co-operate fully in the necessary precautions.

    (ix) There should be suitable warnings on the containers of toxic materials
about the precautions necessary when using these materials.

    (x) In windy conditions, covers should be used over spray booms to
minimise spray drift. This applies particularly to the spraying of hormone
weedkillers and to spraying by contractors, who cannot wait for good weather
conditions to the same extent as can farmers.

    (xi) The danger to insects is probably much greater than that to birds and
mammals and at the same time more difficult to evaluate or even to discover.
We understand that the problem of the influence of plant protective chemicals
on the balance of insect populations is recognised by biologists, and that it is
being studied in more than one laboratory. We wish to underline the
importance of this field of research.

    (xii) Although the danger to honey bees is well appreciated, there has
been some increase in recent years in casualties to bees following the greater
use of the new crop protective chemicals. We consider, however, that improved
co—operation between beekeepers, growers, farmers and spraying contractors
should obviate the need for legislation to protect bees. General propaganda
should be reinforced by articles in local journals or papers, and by appeals
directed to local conditions.

    (xiii) Provided there is no carelessness in the disposal of containers, unused
spray liquids or machine washings, we are satisfied that the normal applica-
tion of toxic chemicals in agriculture will not result in damage to fisheries.
The existing legislation relating to the prevention of pollution of rivers should
be adequate to deal with all cases of carelessness.

    (xiv) The long term effect of weedkillers on the flora of the countryside
is a subject about which little is known. There is no doubt that by using
weedkillers the farmer is able to remove his weeds more effectively and
usually more cheaply than he could by machine cultivations. handhoeing.
cutting and the like, and that in both cases the resultant plant community
is quite artificial. The problem of changes in the plant population of hedge-
rows and roadside verges due to spraying is more complex, and is one which
is of interest to biologists and road engineers as well as to farmers. In our
view, long term observations of the changes are essential and we are pleased
that some experiments are already in progress.

    (xv) Care should be exercised by farmers in the use of rodenticides,
particularly when pigs, cats and dogs may gain access to baits.

    (xvi) Advertising and labelling of rodenticides should not convey the
impression that rodenticides are harmless to animals other than rats.

    (xvii) Finally, our enquiries have clearly shown how great are the gaps in
our knowledge of the effects which the toxic chemicals used in agriculture
may have on wild life, not to mention the possible consequential effects upon
successful crop cultivation; and our pilot observations have indicated how
well justified further field studies would be. There is a pressing need for
more fundamental research.

(i) That the terms of reference of the Interdepartmental Advisory Com-
mittee, set up in accordance with the recommendations of our second report,
should be widened to include risks to wild life arising from the use of toxic

chemicals on pests and weeds. In this connection. the committee should be
charged with the following further functions:

    (ii) That further research and investigation. referred to in (i) (b) above
might include: the effects of the main agricultural sprays on the wild life
in sprayed fields, including the effect on birds of the killing of large numbers
of insects; the effects of sprays on the parasites and predators of the insect
pests against which the sprays are applied; the extent to which the killing
of pollinating insects adversely affects seed-production in the sprayed crops;
the distances to which sprays of various types, such as high-volume and
low-volume, may drift, and the degree of injury they may cause to adjacent
or distant crops.

    (iii) That the importance of adopting the precautionary measures indicated
in our report be clearly made known to spraying contractors, farmers, growers
and beekeepers.

Estimate of areas sprayed with crop protection chemicals in U.K. during 1953

   2. The population of partridges and pheasants is kept high by artificial
methods, and is lowered by shooting. Any increase in the heavy normal
mortality of the breeding stock or of the hatched chicks due to other factors,
such as toxic sprays, is of considerable importance to the game preserver,
since it reduces still further the shootable surplus which is regarded as the
game " crop ". The game keeper’s job is to keep down rats, crows, magpies,
jays, etc., which destroy eggs and kill the young. The weather during the few
critical weeks after hatching is also a potent factor in reducing the potential
numbers of the game-bird population.
   3. Under average conditions the mortality of partridges may be something
as follows:—
   In April, starting with 100 nesting birds (50 pairs)

May ... ... ... ... ... ... ... ... .. May——June ... ... ... ... .. June——September ... ... September ... ... ... ... ... ... September ... ... ... ... ... ... October ... ... ... ... ... ... ... . September—April ... ... ... April ... ... ... ... ... ... ... ... ..

750 eggs laid 200 eggs fail to hatch (nest losses, etc.) 300 chicks die before maturity 250 young survive and 80 old birds 330 total before shooting 120 crop which can be shot 110 normal winter "wastage" (including emigration) 100 next season’s nesting stock

   4. The pheasant figures are similar. If the winter is prolonged or severe
then the number over—wintering may be reduced and if the hatching period is
wet and cold (as in l953) then few if any chicks are reared.

   5. An estimate has been made for us of the total numbers of each species
shot; this, which is based on the number of cartridges used, can only be
regarded as a guide, but the figures are the only ones available on which to
form any estimate of game population.

   6. The estimated number shot is:——