academisch proefschrift ter verkrijging
van den graad van doctor in de wis- en
natuurkunde aan de rijksuniversiteit te
utrecht, op gezag van den rector magni-
ficus dr th. m. van leeuwen, hoogleeraar
in de faculteit der geneeskunde, volgens
besluit van den senaat der.universiteit
in het openbaar te verdedigen op maan-
dag 5 juni 1939, des namiddags te 3 uur

-____* ^

De beschreven onderzoekingen werden verricht ten be-
hoeve van de N.V. De Bataafsche Petroleum Maatschappij,
Laboratorium Amsterdam, dat mij tevens de noodige chemische
gegevens verstrekte.

Bij de genomen proeven werd in ruinie mate medewerking
verleend door Mejuffrouw G. F. E. M. Dierick en de Heeren
F. E. Loosjes en Ph. Gerolt.

Carbon disulphide — Hydrogen cyanide — Chloropierin — Hydrogen
phosphide — Areginal — Ethene oxide — Methyl bromide.

Examination of the gassed insects — Procedure of the experiments and
recording of the results obtained — Experiments with methallyl chloride in
gassing boxes — Experiments with eggs, larvae and pupae.

IV.nbsp;Determination of the effect of the presence of the products
which are to be fumigated on the dose required and of the
effect of the gas on the quality of the products ...31
The influence of the quot;fillingquot; on possible change in quality.

Fumigation time of 24 hours and longer — Fumigation time of 8 hours
or less — Gram-hour value.

11. Proportion of concentration to time. Gram-hour value .nbsp;67
Haber's law. Calculation of practice dose.

INSTRUCTIONS FOR THE USE OF METHALLYL CHLO-
RIDE IN ACTUAL PRACTICE.......^ • • • 93

For those who are interested in the control of noxious insects
but have only very little or no knowledge of entomology a
general introduction is given, which probably contains very little
news for entomologists.

An insect is primarily caused by natural equilibrium being
disturbed, as a result of which the conditions for insect life
become disproportionately favourable. In many cases the
equilibrium is disturbed by man who lays out plantations, stores
goods and carries insects to areas where they did not exist
previously.

By improving agricultural and horticultural plants, varieties
are obtained which are even more susceptible to insect attack.
By crowding a small area with plants conditions are made more
favourable for insects.

Insects are very often carried from one area to another where
the conditions are more favourable, the more so, if, as is very
often the case, the enemies of these insects do not occur in the
new area.

Large quantities of goods are accumulated only by man. Very
often the storage conditions of the goods are very auspicious
to the development of an insect fauna.

The aim of control of insect pests should be in the first
place to restore the disturbed equilibrium, hence to make the
conditions which have become too favourable, unfavourable
again. The cause of the insect pest is then removed; so this
method of control may be called a causal one. If this method is
applied the pest should first be analysed. A close study must be

made in every case of tiae direct causes of the pest. Sometimes
it will be possible to eliminate these. If, for instance, a variety
of insects has been carried from one area to another, but its
enemies have not been brought along with it, they may be
introduced later, thus eliminating the cause of the pest. How-
ever, the cure is not always as simple and straight-forward as
this. It is possible that the insect itself thrives in the new area,
but that its enemy cannot live there. The case then becomes a
little more complicated and one way to meet the difficulty would
be to find a new enemy that can live in this area.

In many cases there are even more complications. Methods
for growing must be changed so that they will not longer suit
the life circle of the insects; varieties of plants with a high resis-
tance should be grown, and so on. Generally, these methods
consist in either adding a factor or withdrawing one from the con-
ditions for life of the insect, as a result of which these conditions
become unfavourable. This may be called natural or hiological
control, and may be considered as work belonging to the field
of biology. This control method is characterized by the fact
that the effect is permanent, since the pest will generally not
return as a result of the same cause.

In many cases it is not possible, however, to affect the con-
ditions for life of the insect by natural means. In that case means
should be found to kill the insects and to limit the pest as much
as possible, though not removing the cause of the trouble.
It must be remembered, however, that there remains a chance
of the pest returning.

This kind of control will often be effected with the aid of
chemicals. This may therefore be referred to as chemical or
technical control. It should also be preceded by accurate
biological analysis. For the actual control, however, the biologist
will need the help of the chemist and the physicist and generally
also of the technician.

In agriculture and horticulture biological control will in many
cases afford relief. This is not the case with insect pests in

stored goods. Storing of goods is unavoidable and only in very
few cases can the method of storage be changed. Here the
conditions created by man are the cause of the trouble. Methods
such as those worked out by Rank (i), in which a continuous
current of air is passed through the goods, would in the long
run be too expensive and can not be applied in the case of many
products, such as tobacco. In by far the most cases chemicals
must therefore be used.

The extent of this kind of pests is still too little known. Just
as the presence of bed-bugs in houses is kept secret by the
occupants, the owner of warehouses endeavours to hide the
fact that insect pests occur in them. The owners of the goods
would object — and rightly — and in some cases might even
claim damages. On the other hand the consumers would be
greatly displeased to know that their food is, or has been,
infested with „verminquot;. For these reasons the control of the
pest is often carried out in secrecy. Yet the pest is controlled
although by no means in a sufficient degree. As a result of this
many warehouses still swarm with millions of moths and as
many weevils as well as larvae of both.

It is the task of the biologist to find and to try out the means
of putting an end to these pests in the simplest and cheapest
manner.

Particulars concerning the insects used during this investigation
and concerning the fumigants which are already known for controlling them

The investigation described below is meant to be a con-
tribution towards the control of insects in stored goods with
the aid of a gas poisonous to these insects. It was carried out
mainly on species of Calandra occurring in stored grain. The
gas may also be used, however, for the control of other kinds
of insects in other products. These applications are also men-
tioned in my investigations. In all cases, however, Calandra
granaria L. were used for the experiments.

The test object I mainly used was Calandra granaria L. and
Calandra oryzae L. from my own, well-kept, cultures. During
the summer the insects were bred at room temperature
(i8—22° C) and during the winter part of them in an incubator
at 20° C and the rest in an incubator at 27—28° C. The latter
part was kept at 20° C for at least two or three days before
fumigation. When adults were used, young, strong specimens
about four weeks old were taken.

The maintenance of the cultures offers few difficulties. It is
important that the degree of humidity should be sufficiently
high, since otherwise the weevils will not lay any eggs.

According to Schulze (2) Calandra granaria does not oviposit
at a relative humidity below 30—40 %. Moreover, the lifetime
of the adults is shortened if the humidity is too low.

I am able to confirm the fact that neither of the Calandra
species multiplies in a very dry atmosphere. C. oryzae is not
so sensitive as C. granaria in this respect. I observed several
cultures of Calandra on wheat containing both species. As these
cultures were kept a little too dry C. granaria died off altogether
while C. oryzae thrived.

There is no difference in opinion as regards the systematic
position of Calandra granaria L. and Calandra oryzae L. In
American literature, however, these insects are generally menti-
oned by the names Sitophilus granarius L. and Sitophilus
oryzae L. Sometimes in literature a certain species, referred to
as Calandra zeamais Mötsch., may be encountered. Opinions
differ however, as to whether this is a separate species or a larger
variety of C. oryzae. I found large-sized Calandra in a batch of
maize which were all determined as C. oryzae. I bred normal-
sized C. oryzae obtained from rice for several generations on
maize and observed a considerable increase in size. According
to Cotton and Good (3), C. oryzae and C. zeamais are identical.
Zacher (4), Weidner (5) and Andersen (6), on the other hand,
consider these insects to be different species.

An excellent description of the anatomy and biology of
C. granaria is given by Müller (7). The information given there
also applies to C. oryzae.

The publication by Teichmann and Andres (8) is somewhat
older and contains beautiful, coloured illustrations.

A very good monograph was recently published by Andersen
(6). The following facts are important for the control of these
insects. The entire development of the Calandra takes place

inside the grain of corn. The full-grown weevil bores a hole
in a grain of corn with its mandibles and the egg is deposited
in it. According to Zacher and Andersen the opening is closed
by means of a secreted substance. Müller, however, is of the
opinion that although a glass-like, transparent lump is attached
to the egg, the opening is not entirely blocked by it. The fact
that the eggs are within easy reach of fumigants pleads for
Müller's opinion. It is generally agreed that only one larva
can develop in one grain. Although Zacher states that more than
one egg is deposited in one grain, Müller maintains that these
originate from different females and that only one larva develops
completely. Only in maize it should be possible that more than
one larva may develop in one grain. Normally, larvae and
pupae remain in the grain during the whole of their existence.
It is impossible therefore to tell from the appearance of a batch
of grain whether it is infested. This can-only be stated with
certainty by opening a great number of grains or by placing
a sample of the shipment in an incubator at about 28° C for
six weeks. If during that period no weevils appear the shipment
may be considered to be not infested.

The duration of the total development from egg to adult is
very much dependent on the prevailing temperature.. For
Germany in the summer Zacher gives a period of roughly
two months. In my cultures at 27—28° C the duration of the
development amounted to about six weeks.

Opinions differ as regards the number of progeny of one
female Calandra. According to Andersen this number is about
160. According to my own observations this number depends
very much on the humidity of the wheat. At a relatively high
humidity the number will certainly be much higher than 160
and will probably amount to many hundreds.

In a seriously infested lot of grain or in an overcrowded
culture it may occur that older larvae which during their first
stages have lived inside parts of a grain of corn later on may
be observed to lie exposed. They can then develop normally

in the flour formed by the infestation. Eggs and young larvae
cannot develop outside the grain of corn.

According to Zacher C. granaria may develop in wheat,
rye, maize, barley, malt, rice, millet, buckwheat, sweet chestnuts,
acorns, maccaroni and noodles, while C. oryzae can also develop
in cotton seed.

C. granaria is an insect which only occurs in stored goods and
cannot fly. C. oryzae can fly, however, and therefore would be
able to leave the sheds and infest the crops in the fields.

Schulze (2) states that C. oryzae makes very little use of its
wings. I often observed that weevils which had been killed
by fumigants lie with their wings and elytra spread, while I
sometimes noticed them flying around in the rooms where they
were being bred.

Statements on the damage which may be caused by Calandra
vary greatly. In any case the damage is considerable. It does
not only include the grain which has been devoured, but the
presence of many Calandra increases the humidity of the grain,
as a result of which overheating may occur. Furthermore, the
germinating power of seed-corn is considerably impaired.

In order to obtain an idea of the damage caused by the grain
being devoured, I put 1000 adults of C. granaria into 4 kg of
wheat in the incubator at 27—28° C. After a period of seven
weeks the wheat was reduced to 3.67 kg which means a decrease
of 8.25 % and the number of weevils had risen to many tens
of thousands.

Corn weevils have always been very difficult to control, one
of the reasons for this being that the larvae and pupae pass their
life-cycle concealed in the grain. One of the oldest methods for

control is to submit the grain to repeated motion. The weevils
are disturbed and emerge from the grain. This may be demon-
strated in the laboratory by shaking a culture of Calandra. This
method, however, has no effect on the younger stages.

A primitive repellent is fresh hay or straw. In Holland I
noticed some farms in the province of Groningen where this
method is still applied. According to Zacher the active con-
stituent is coumarine. In such a small dose, however, it has
no appreciable effect.

A more modern method is coating the walls of the store-
rooms with chemicals, e.g. aniline. Vide Zacher (9).

From the bionomy of the insects described above it is clear
that these and similar methods cannot ensure efficient control.
I should like to define efficient control as: quot;one operation or one
series of operations by which all insects present in all stages of
development are killedquot;. The effect should be such that the
insects cannot occur again in rooms or goods which have been
treated, unless they are introduced from some other place.

Since species of Calandra and the majority of insects occurring
in other products are spread throughout these materials only
those means of control can be effectively applied in which the
substance used penetrates all through the products. As such
may be considered temporary increase or decrease of temper-
ature. This method can, however, only be applied in exceptional
cases, while furthermore the cost is too high.

Further, fumigation comes into consideration. This method
consists in the introduction of a quantity of gas poisonous to
the insects into a properly closed storeroom containing the
products. After a certain period the gas is removed by means
of ventilation. A list of the gases which are generally used for
this purpose is given by Frickhinger (10). A great deal of in-
formation may also be found in circulars No. 369 and 1313 of
the U.S. Department of Agriculture (11). I will briefly mention
here those fumigants which in actual practice are applied in
considerable quantities, though this list does not pretend to be

exhaustive. All these products are used both for the control
of Calandra species and other insects in stored goods.

This is one of the oldest fumigants. It is mainly used as a
soil insecticide. Vide Trappmann (i 2). He advises 200—300 g per
cub. m for fumigating stocks of grain (not seed-corn) for 3 2 hours.
This material was further used for the control of Lasioderma
in tobacco, as was communicated by De Bussy (13).

One of the greatest drawbacks to the use of carbon disulphide
is its inflammability. Hinds (14), for instance, states: quot;it is
hardly safe to have steampipes very hot or to turn on or off
an electric light or fan. Even the heavy striking of a nail with
a hammer might cause an explosionquot;.

Scherpe (15) also draws attention to the high inflammability
ot CSg. In some countries fire insurance is said to be suspended
during fumigation, while in U.S.A. the fumigation of railway
carriages with CSg is prohibited. Scherpe also describes diffi-
culties owing to the grain absorbing the gas too strongly.

The control of Calandra by means of hydrogen cyanide
is described by Teichmann and Andres. They state that
Calandra granaria is highly resistant to HCN. For efficient control
about 3 % by vol. is necessary, corresponding to 30 g of HCN
per cub. m.

Another great disadvantage of this gas is its extreme toxicity
to human beings. In many countries there are legal measures
restricting the use of this fumigant. See, inter alia, Lentz und
Gassner (16). These measures, however, have not been able to
prevent fatal accidents. For reference see, inter alia, Frick-
hinger.

be used for constructive purposes. These gases were too
poisonous to human beings, however. Only chloropicrin can
be used as a fumigant against insects. Experiments with this
gas against Calandra granaria are described by Wille (17) among
others. A complete bibliography of chloropicrin is given by
the U.S. Department of Agriculture (18).

Its disadvantages are that it badly irritates the eyes, nose,
throat and all other respiratory organs and that it adversely
affects the germinating power of the seeds. As a dosage
for actual practice Wille advises 40 ml per cub. m for 24
hours.

This gas is also used against Calandra in various ways.
Fluty (19), however, states that inhaling this gas by man may
cause death within two or three days. This is sufficient reason
to bar its use as a fumigant.

The composition of this material is not known with certainty.
Probably it consists of a mixture of ethyl- and methyl formate.
The insecticidal effect of this gas is roughly equal to that of
carbon disulphide, while its mixtures with air are said not to be
explosive. Frickhinger states that areginal is not dangerous
to man, which, however, is contradicted by the results of my
experiments on mice. Frickhinger also states that areginal does
not affect the germinating power of seeds.

Kleine (20), however, concludes from his experiments that
areginal is highly inflammable and should be handled with the
utmost care, and besides that it strongly affects the germinating
power of seeds.

This material was first described as an insecticide by Cotton
and Roark(2i)in 1928. In the subsequent ten years an extensive

Pure ethene oxide is highly inflammable and very easily gives
rise to explosions. Cf. Lentz and Gassner. In order to minimize
this danger of explosion it is mixed in various proportions with
carbon dioxide, which according to the investigations of
HazelhofF (22) also stimulates the respiratory centres of the
insects.

Besides its great inflammability ethene oxide has the draw-
back that it strongly affects the germinating power of seeds and
that it is very poisonous to man. Investigations in this field
were carried out by Flury (39) who came to the conclusion that
especially the after effect of ethene oxide is very dangerous to
the human organism. The symptoms of poisoning occur quite
a long time after the gas has been inhaled, so that it may by
that time be too late to afford assistance.

Owing to its low boiling point this material and its mixtures
can only be stored and transported in steel cylinders, which
increases the cost of the fumigation considerably.

This material was first used as an insecticide in California in
1935. Its activity is described in recent publications by Mackie
and by Fisk and Shepard (23). According to these publications
methyl bromide has attractive properties as an insecticide. A
great disadvantage was pointed out by Lepigre(33), viz. too
great a quantity of HBr remains behind in the fumigated food-
stuffs, which might endanger the health of the consumers.

Apart from their favourable properties, the fumigants which
are in use at present have still so many disadvantages that
efforts should be made to invent other materials which either
do not have these disadvantages at all or only to a smaller
extent. It is of the greatest importance to find a material which
is about just as toxic to insects as the fumigants already known,
but which is less toxic to man.

For this reason I carried out investigations with methallyl
chloride. This material had only been described as an insecticide
in the patent literature (24). I published a short paper on the
investigations in Nature (25), while further information was
communicated by me at the 7th International Congress for
Entomology at Berlin (26).

Methylallyl chloride or methallyl chloride, which will further
be referred to as M, is a colourless liquid which has the following
structural formula:

The specific gravity at 20° C is 0.925 and the boiling point
72° C. The liquid itself is inflammable. The explosion range of
mixtures of M vapour with air lies between 93 gr/cub. m
and 375 gr/cub.m.

The liquid evaporates readily in the open air, its latent heat
of evaporation at 20° C is 89 cal./kg.

The odour of the liquid is strong, but not unpleasant.
While investigating this fumigant I endeavoured to answer
the following questions:

3.nbsp;What is the penetrating power of the gas into the goods
which are to be fumigated?

4.nbsp;What influence has the presence of the goods which are to
be fumigated on the concentration required?

7.nbsp;What is the proportion of the concentration required to the
time of exposure?

8.nbsp;What substances may be added to the liquid to make the
mixture non-inflammable without adversely affecting its
toxicity?

These questions were in the first place applied to Calandra
species in stocks of grain, but the results obtained were also
tested upon other insects and other products.

In these experiments I used closed wooden gassing boxes
made of three-ply wood 0.8 cm thick. The dimensions are:
height 50 cm, length and breadth 40 cm, while the capacity
is 80 litres. The boxes are provided with two sliding panels,
one on the top and one in the front. A glass window is provided
at one of the sides. The insects were put in tins, 5 H cm high
and 14 cm in diameter. The bottom and the lid are provided
with gauze. Vide photograph No. i.

In all cases some wheat was added for the insects to feed
upon. The tins were placed in the gassing box on small legs

about 2 cm high. The two sUding panels were secured by strips
of surgical plaster pasted round them before each fumigation.
The liquid which was to be evaporated was introduced through
an opening in the top of the box by means of a 5 ml pipette
graduated to 0.05 ml. Under this opening a glass disk was placed
with filter paper in it. The opening was closed by means of a
cork which was sealed with cellon lacquer.

The boxes were placed in a room facing North, where the
temperature was 18° C and fairly constant.

The tins with the insects were taken out of the boxes 24 hours
after the liquid had been introduced. The two sliding panels
were opened and the boxes were degassed in a fume-cupboard,
to which a powerful exhauster had been connected.

This system was worked out by P. Korringa, who, before
me, carried out a number of preliminary experiments with
methallyl chloride, the results of which were not published.
I examined the insects in some cases immediately after they
had been taken out (dir.), and in all cases 24 hours and one
week after they had been taken out (24 h., i w.).

50 insects were used for each fumigation. In cases where
this number was deviated from, this is mentioned.

Opinions differ about the manner of evaluating the results
obtained in experiments for insect control. Not all insects are
killed immediately, some of them are definitely affected, but
still move their legs or antennae. The question is whether these
insects should be mentioned in the results as quot;affectedquot;, quot;half
deadquot; or otherwise. Sometimes two quot;half deadquot; insects are
computed as quot;one deadquot; andquot; one alivequot;, but I wish to emphasize
my absolute disapproval of this procedure.

The object of the fumigation is to kill the insects. An insect
is either dead or alive. If it is merely affected it can either recover
or succumb. This can only be ascertained by keeping the animals
for a certain period and examining them from time to time.

I evaluated the results as follows: the Insects which in any
way were still moving were considered to be alive. Only those
insects which showed no movement whatever were considered
to be dead. Applying this procedure the difficulty remains,
however, that Calandra can simulate death for quite a long time.
To the trained investigator, however, the position of the legs
and antennae easily reveals whether the insect is really dead or
is only simulating death. Besides, all the motionless insects were
closely examined, by means of a heated glass rod. This should
be done with great care so that the insects which may possibly
be still alive are not burnt. In this manner the insects were
sometimes checked immediately after gassing and invariably
24 hours and one week after gassing. The latter is essential since
it is possible that:

a. the insects appear to be dead immediately after gassing, and
remain motionless for a few days, but recover again after-
wards;

h. the insects immediately after treatment appear to be affected
only very slightly, but succumb in a few days.

Procedure of the experiments and recording of the results obtained.
During the first series of my experiments in the gassing boxes
described above the duration of the fumigation and the temper-
ature were kept constant. No special measures were taken to
keep the relative humidity constant. This is usually fairly high
in the Dutch climate. With these constant or nearly constant
factors the dosage of gas applied was varied.

A similar method was also followed by Strand (27). Strand
introduced about 30 Tribolium confusum Duv. into an
Erlenmeyer flask with a capacity of 6.4 litres and observed the
kill for various fumigants with increasing doses.

The results are laid down in graphs in which the percentage
of killed insects are plotted on the ordinate and the concen-
trations of the fumigants in mg per litre on the abscissa. With

the aid of the points thus obtained he draws curves which
always have a sigmoid shape.

To have a criterion for comparing different fumigants this
author wishes to introduce the quot;median lethal dosequot;. This
suggestion was based on his observation that in repeated series
of experiments the results fitted best at the point of 50 % kill.
For this reason he wishes to determine the dosage required to
kill 50 % of the test objects. Following Trevan (28) he calls
this the median lethal dose and the rate of this dose is supposed
to give the relative value of the fumigant.

I wish to discuss Strand's experiments here because both
his sigmoid curves and his suggestion of a quot;median lethal
dosequot; have been adopted by many authors, while I do not agree
with this.

I must raise the following objections against the experiments
of Strand. He speaks of concentration while really he is only
informed about the dose. These two conceptions are not kept
sufficiently distinct by him and many others.

Concentration may be expressed as the number of mg of
gas which are really present in a litre of gas-air mixture.

Dose may be expressed as the number of mg of gas which
are applied per litre of space.

The concentration should be determined by taking a sample
from the space filled with gas and analysing this.

The dose is always known. It is incorrect and also confusing
to speak of a quot;theoretical concentrationquot;.

The concentration of the gas will never be the same all over
the space containing the gas, unless the gas mixture is continu-
ally kept in motion, which is not done by Strand. He compares
HCN and CSj, a very light and a very heavy gas, while during
the experiments the insects were kept at the same level in the
space.

I wish to observe the following with respect to the recording
of the results in sigmoid curves. In recording the results of an
investigation only those data should be reported, which have

really been observed as well as the conditions under which the
observations were made. Facts therefore. Strand and others
find a number of — sometimes very few — figures. These
figures are plotted in a graph. There is nothing against that.
But then they construct a curve along the few points obtained,
which in my opinion is insufficiently justified. I admit that such
a curve may be obtained with a very great number of test insects
under ideal, standardized conditions. In most of the experiments
with insects the conditions cannot be controlled so easily and the
number of insects is not so great as to justify constructing such
a curve with the values obtained. Evidently the reasoning is as
follows: it is known that under ideal but practically unattainable
conditions a curve of a certain shape may be obtained. A few
points are obtained which only very roughly give the course of
the curve. Then the presumed curve is drawn, which, however,
in most cases does not even intersect the points found in the
experiment.

In my opinion it is more correct to record only that which has
really been observed. So a graph is obtained which is based on
facts and not on ideal conditions which were never attained.

Like Shepard and his collaborators (29) and Peters (30)
I reject Strand's suggestion to judge different fumigants with
respect to each other with the aid of the quot;median lethal dosequot;.
The two authors mentioned, like myself, are of the opinion
that fumigants which must serve in actual practice to kill all
the insects, may only be evaluated in the laboratory by the
dose which effects a kill of 100 %.

Strand and many others forget that insects are living creatures
which must be regarded from a biological point of view. In
each experiment he fumigated 30 Tribolium. These insects
form a population amongst which are insects with a low, medium
or high resistance to the gas. If 50 % of them are killed by an
insecticide, these consist of all the weak ones and half of those
with a medium resistance to the gas. If this were to be expressed

in quot;life unitsquot;, it might mean a kill of some 30 % with the median
lethal dose. Besides, it is not impossible that all the male insects
would be killed and the — generally fertilized — females would
survive.

It should be demanded of an insecticide that, at least under
ideal laboratory conditions, it kills all the insects, including the
strongest. If the latter survive, the treatment has an opposite
effect, viz. a selection takes place as a result of the survival of
the fittest, a danger which always exists in the control of insects
and to which too little attention has been paid up to now.

For this reason I reject both the recording of the results in
sigmoid curves and comparison by means of the quot;median
lethal dosequot;. Instead of this I give graphs representing exclu-
sively the facts that have been observed and a comparison by
determination of the dosage at which a kill of 100 % is obtained.

In my graphs the doses applied have been plotted horizont-
ally. On this line vertical columns have been drawn representing
the corresponding kill for each dose. Side by side are given
the series of observations directly after gassing (dir.), 24 hours
later (24 h.) and one week later (i w.). In this manner three
series of columns were obtained, which only represent what
has really been observed and yet give an excellent idea of the
effectiveness of the gas in question.

EXPERIMENTS
with methallylchloride in gassing boxes with a capacity of 80 1.
Fumigation time 24 hours

1.nbsp;To affect Calandra granaria L. to such a degree that all
insects are killed one week after fumigation a dose of about
18.75 g of M per cub. m is required under the conditions
mentioned above.

2.nbsp;Under the same conditions about 12.50 g per cub. m is
required for Calandra oryzae L.

3.nbsp;C. oryzae is therefore much more susceptible to the gas than
C. granaria.

4.nbsp;M has a very strong after effect. In the case of C. granaria
about 80 % were still alive immediately after fumigation with
a dosage of 18.75 g; 24 hours later about 15 % were left and
a week after fumigation all the insects were dead.

5.nbsp;The results of fumigation with one fixed dose vary consider-
ably. This variability always occurs in this kind of experi-
ments. In order to obtain a clear idea of the dose required it
is essential to carry out more than one series of experi-
ments.

The values obtained were checked with the aid of larger,
firmly constructed gassingboxes. These boxes have a capacity
of 345 L and are made of three-ply wood 2 cm thick. The dimen-
sions are 80 x 70 X 65 cm. The boxes contain a number of
wooden grids. The loose door measures 60 X 70 cm and is
screwed on the box by means of ten large winged nuts, while
soft rubber is fitted round the edges to seal the box effectively.
The boxes are provided with a wide exhaust tube, which reaches
to the bottom of the box and is connected to a powerful
exhauster (vide photograph No. 2).

There are two openings at the top closed by means of screw
caps, through which fresh air enters when the exhauster is in
operation. The tins contaiaing the insects were placed on a
wooden grid halfway up the boxes.

EXPERIMENTS
with methaUylchlofide in gassingboxes with a capacity of 345 L.
Fumigation time 24 hours

The results are somewhat more favourable, which may
probably be ascribed to relatively more gas having been ad-
sorbed to the walls of the small gassing boxes, while it is also
possible that small quantities of gas diffused through the walls
of the small gassing boxes.

The question now is whether the other stages of the two
species of Calandra can also be killed by fumigation with M.

As has already been mentioned these stages live inside the
grains of corn and therefore are more difficult to reach.

In order to solve this question I carried out experiments in
accordance with the following principle.

A great number of adults were introduced into a quantity
of wheat. These were kept there for from six weeks to two
months. It is then fairly certain that all stages are present in the
wheat. If this is not the case it will become apparent from the
experiments.

7\11 the adults were then screened oS and the wheat was
divided into two equal parts. One batch was gassed, while the
other was not. Then the two batches were screened daily
whenever possible. The difference in numbers of the adults
found was taken to represent the effect of the fumigation.

On June 7th 100 adults of Calandra granaria were introduced
into wheat and were screened off again one week later on June
14th. The wheat therefore could contain at the most eggs and
very young larvae. Half of the wheat treated in this manner
was fumigated with a dose of 13 g of methallyl chloride per cub.
m for 24 hours.

The adults which hatched later were screened off and the
results are given in the table below:

Wheat which had contained about 500 adults for a period of
I % months was screened, divided into two equal batches of
which one was fumigated with a dose of 14.5 g of M per cub.
m for 24 hours at an average temperature of 17° C.

775 g of wheat containing eggs, larvae and pupae of C. gra-
naria and C. oryzae and a similar batch of 900 g were each gassed
with a dose of 17.4 g of M per cub. m.

Third series non-fumigated total 3291 adults, fumigated i.
Fourth series non-fumigated total 5162 adults, fumigated i.

1.nbsp;Fumigation with a dose of 14.5 g of M per cub. m for 24
hours is sufficient to kill practically all the eggs, larvae and
pupae of C. granaria and C. oryzae.

3.nbsp;Fumigation with 18.5 g of M per cub. m for 24 hours at
a temperature of about 18° C kills all stages of C. granaria
and C. oryzae completely.

The vapours of M are three times as heavy as air and there-
fore the question arises how does it distribute itself in a space
where no special measures have been taken (for instance
mixing by means of a fan) to make the gas-air mixture homo-
geneous ?

In order to ascertain this fourteen gauze bags each containing
50 C. oryzae were placed in a gassing box with a capacity of
345 L. Five of them were placed on the bottom of the box
(one in each corner and one in the middle); 5 halfway up the
box and four against the roof. A dose of 18.85 g per cub. m
was introduced at the top of the box on a large sheet of filter
paper. The temperature was 20° C and the duration of the
fumigation six hours. The insects were examined 24 hours and
one week after fumigation.
See table pages 23

The gas in the box is distributed practically homogeneously.
In the first series of experiments the effect is somewhat less

Second series. In these experiments the liquid was not evaporated on
a large sheet of filter paper, but in a small petri dish with a small
piece of filter paper.

pronounced, because the sheet of filter paper had adsorbed a
good deal of gas. In fumigating trials in actual practice I had
the opportunity to investigate the distribution of the gas in
much larger spaces (page 60).

III. INVESTIGATION INTO THE PENETRATION OF THE GAS
IN WHEAT. FUMIGATION OF SILO'S

In order to ascertain whether the gas penetrates into the
wheat sufficiently to kill the insects in the middle of the bags
as well, a wooden frame measuring 45 x 45 X 50 cm was
constructed and covered with jute sacking. This square bag was
filled with about 80 kg of wheat in each experiment and it was
ascertained what dose would be sufficient to kill Calandra
granaria contained in a bag in the middle of the wheat. To this
end the whole bag was placed in a 345 L. gassing box and was
fumigated with different doses, at a fumigation period of 24
hours.

dir.

Ï3

96
100

I week

97
100
100
100

The same experiment was repeated with the other stages of
the Calandra species. To this end a small gauze bag of wheat
containing all stages of the insect was placed in the middle of
the large bag of wheat. A parallel sample remained unfumigated.

The dosage was 24.7 g per cub. m, duration of fumigation
24 hours, temperature 20° C. After fumigation the two samples
containing the eggs, larvae and pupae were again screened
practically every day.

In order to obtain better idea still of the distribution of the
gas in a large quantity of wheat, one of the 80 L gassing boxes
already described was entirely filled with wheat, and five small
gauze bags each containing 50 Calandra granaria were placed on
the bottom of the box, five halfway up the box and five against
the roof.

The gas was introduced in one of the top corners, immediately
above bag No. 11. Bags No. 6 and i are situated immediately
below. The dosage for this gassing box was 3 g. Duration of
fumigation 24 hours, temperature 20° C.

The insects were examined immediately after fumigation,
24 hours later and a week later. The figures represent the per-
centage of insects killed.

The results of this experiment have been conveniently
arranged in figure i. The figures in this drawing represent the
percentage of kill one week after fumigation. The positions of
these figures correspond to the positions of the bags containing
Calandra. This illustrates very clearly that the gas immediately
descended to the bottom and spread there; also that the con-
centration must have been weaker halfway up and reached a
minimum at the top.

For reasons which are described further on experiments were
also carried out to examine the penetration during a shorter
period of fumigation. The same square bag was used for this
purpose filled with 83 kg of wheat. Five small gauze bags each
containing 25 Calandra granaria were placed on the bottom,
5 halfway up and 5 on top of the wheat. The positions of the
bags were:

Middle
9 8
lO

6 7

Bottom
14 1}
ij

IInbsp;12

Top

4 3
5

Inbsp;2

The square bag was covered with jute sacking and placed
in a gassing box with a capacity of 345 L and fumigated with
a dose of 72 g per cub. m for 6 hours at a temperature of 20° C.

Six hours after the liquid had been introduced the gas was
withdrawn by means of the exhauster and the insects were
removed from the wheat and examined 24 hours later.

It is clear that the effect was somewhat stronger near the
bottom and it is important to note that the effect on bag No. 10,
which was situated in the middle, was much less.

A week after fumigation this difference existed no longer,
since all insects were dead.

During this experiment I noticed that gas was still present
in the middle of the wheat after the gassing box had been
exhausted. The procedure was therefore as follows: The gas
penetrated into the quantity of wheat very slowly reaching the
middle bag after some time. Fumigation was stopped 6 hours
after the liquid had been introduced. At this moment the bags
containing the insects were removed from the wheat.

In actual practice the procedure is different. If a storeroom
containing bags of wheat or other products is fumigated, the
gas penetrates to the middle of the bag very slowly in the same

way as in the experiment. When subsequently the room is
ventilated, the gas which is present in the bags diffuses very
slowly from them. Inside the bags fumigation therefore con-
tinues even after ventilation has been started.

In order to imitate these conditions in the laboratory, the
experiment was repeated and modified as follows. The square
bag of wheat was removed from the box at the end of 6 hours,
but the little bags containing the insects were left in the wheat
for another 24 hours. Furthermore, a smaller dose was applied,
so that any differences would not be imperceptible owing to
all the insects succumbing. In this case the dose was 40 g per
cub. m. Temperature 20° C. The insects were examined immedi-
ately after they had been removed from the wheat, 72 hours
later and one week later. The following results were obtained:

Now it may be seen, indeed, that the effect on the middle
bag No. 10 does not differ from that on the other bags.

This experiment shows that the problem of the penetrating
capacity of a gas is not of such importance as is generally believed.
If a gas takes a long time to penetrate, it also takes a long time
to escape. The final result is by no means below that of a quickly
penetrating gas.

Of course, the gas should penetrate quickly enough to reach
the middle of the bags within the fumigation time.

In any case the penetration of M into bags of wheat is
sufficient. This is not the case, however, for the fumigation of

silos. These may be up to 30 m high, so that the gas must
penetrate through a very thick layer of wheat to reach all the
insects.

Experiments were carried out in a glass tube 3 m long and
7.5 cm in diameter. The total capacity is about 11.5 L.

This tube was filled with wheat. Small gauze bags each
containing 25 C. granaria were placed at distances of 50 cm.
Altogether there were 6 bags, so the top bag was situated 50 cm
below the surface of the wheat. It was found that even a dose
of 260 g of M per cub. m (calculated on the total capacity of
the tube of 11.5 L) was insufficient to reach the second bag
from the top. The insects in the first bag, situated 50 cm below
the surface, were already killed by a dose of 43.5 g per cub. m.

A duplicate experiment was carried out in a tube 1.50 m
long, the diameter being the same as that of the other tube,
which was filled with glass beads. It was found that 260 g per
cub. m was only just sufficient to kill all the insects in the bags
Although in the case of glass beads only adsorption comes into
play the required dose is already very high.

These experiments show that for silos up to 30 m high
effective fumigation with M is impossible by merely introducing
the gas above the surface of the wheat.

The only solution in such cases is to pump the gas through
the wheat, as is done, inter alia, with areginal.

In order to try this out the glass tube was prepared once
more with six bags of wheat each containing 25 C. granaria,
and a pump with a capacity of 2 L per minute was connected
to it. A Wulff bottle which served as evaporation space was
also inserted in the system. The liquid was introduced into
the botde. The current of gas passed from the bottom of the
tube to the top. The dose was fixed at 120 g per cub. m,
duration of fumigation 6 hours, temperature 20° C. In this
experiment all the insects were killed in the column of wheat.
The dose of 120 g per cub. m should not be regarded as a
minimum value.

This experiment shows that circulation of the gas by means
of a pump is the right solution. Vide Kunike (31).

Thus the penetrating capacity of M is very good, except
when very thick layers are concerned in which case special
technical provisions should be made.

IV. DETERMINATION OF THE EFFECT OF THE PRESENCE OF
THE PRODUCTS WHICH ARE TO BE FUMIGATED ON THE
DOSE REQUIRED AND OF THE EFFECT OF THE GAS ON THE
QUALITY OF THE PRODUCTS

The first part of this investigation provided an answer to the
question as to in how far M is able to kill insects in empty
spaces, the dose being determined at which the kill was 100 %.

The presence of goods, in this case stores of grain, alters the
prevailing conditions, however, and we shall have to see what
influence this has on the effect of the gas with respect to the
insects. It is clear that the presence of goods reduces the quan-
tity of air in the space. Page and Lubatti (3 2) found that the
quantity of air in one cubic metre of wheat is about 400 litres.
So the intergranular space is 400 L. If the same quantity of gas
were applied in i cub. m of empty space and in i cub. m. of
wheat the concentration would therefore become 2 % times as
high in the latter case. For this reason Lepigre (33) regards a
chamber containing wheat as being composed of one part
filled with material and an empty part; he calculates the dose
required for each part separately. Thus for ethene oxide he
recommends quot;23 grams per cub. m. of wheat and 50 grams
per cub. m. of actual spacequot;. So he specifies a smaller dose
for I cub. m. of wheat than for i cub m. of empty space.

The second influence, i.e. of the presence of the product on
the effect of the gas, will, in this case, not be evident. For,
the product deactivates part of the gas owing to adsorption and
as a result of the gas dissolving in constituents of the product

or chemically combining with it. Therefore, on the one hand
the presence of the product in the space causes a rise in the
concentration, because the actual space becomes smaller, on the
other, a reduction owing to absorption etc.

1.nbsp;The average concentration in the filled space remains higher
than in an empty one (a gt; b);

2.nbsp;The average concentration of gas in the filled space remains
lower than in an empty one (a lt; b);

3.nbsp;The average concentration remains the same (a = b).
This refers to the influence of the presence of goods on the

The effect of the gas on the quality of the goods is of far
greater importance. There is little information to be found on
this point in the literature on the subject. Generally the materials
for combating insects are only tested for their insecticidal effect
in fumigation chambers containing nothing but the insects. In
many cases experiments are also carried out to determine the
effect of the gas on the germinative power of seeds, but it
was only in exceptional cases that its effect on the odour and
taste of the products fumigated was investigated.

Roark and Cotton (21) carried out laboratory experiments
with fumigants in spaces partially filled with products. They
used 500 ml. Erlenmeyers. They placed ten insects on the
bottom and poured 250 ml. or about 200 g of wheat over them.

In U.S. Dept. Agric. Bull. No. 1313 experiments are referred
to with fumigants in the presence of a certain quantity of wheat.
In the same Bulletin a number of milling and baking tests are
reported on. It is said that flour often retains the odour of
carbon disulphide and that the flour ground from fumigated
wheat has less good baking qualities. Ethyl acetate in a

Flury (39) states that ethene oxide has an adverse effect upon
the taste of some products.

We therefore see that information on this point in the liter-
ature is rather sparse and in those cases where an investigation
was carried out it was rather superficial. On the other hand, it is
a much discussed point in actual practice. When fumigation
is discussed with managers of warehouses, their first enquiry
is inevitably about the effect of fumigation on the quality of
their goods.

I therefore decided to carry out an accurate and thorough
investigation into the influence of fumigation on the odour and
taste of the products.

In order to find out whether M can affect the odour and taste
under very unfavourable conditions, I carried out the following
experiment:

Two kg of wheat were fumigated in a 345 L gassing box
with a dose corresponding to 72.5 g per cub. m for 24 hours
and at a temperature averaging 18° C. After fumigation the
wheat was aired for 24 hours. At the end of that period there
was nothing abnormal about it. The wheat was then ground
to flour. This flour had a slightly musty smell. The bread baked
from the flour looked perfectly normal, but the mouldy smell
persisted and the taste of the bread was also affected.

Under certain conditions it is therefore possible that M has
a detrimental effect on the odour and taste, which property it
has in common with many other fumigants. The problem is
therefore, what procedure must be followed to avoid this with
absolute certainty.

To gain some information on the problem just mentioned,
the influence of the quot;fillingquot; was first investigated. By quot;fillingquot;
I understand the number of kilograms of product in one cubic
metre of space.

It is by no means the same thing whether loo kg of wheat
are fumigated in one cub. m or in ten, under the same conditions
as regards the dose applied, fumigation time and temperature.
In the latter case the absolute quantity of gas is much larger,
the ratio between the quantity of gas and the quantity of product
more unfavourable. It is therefore possible that in this case
more gas is absorbed by the product than when the latter is
in a smaller space. To make this clear I carried out a series of
experiments.

Glass jars of 3.1 L capacity were used, with metal lids fitting
with a bayonet catch. A series of these jars was filled with 0.50,
0.75, i.oo, 1.25, 1.50, 1.75 and 2 kg of wheat, respectively. On
the bottom of each jar, underneath the wheat were placed gauze
bags each containing 50 Calandra granaria.

It was then determined how much gas was needed in each
case to kill the insects. A small aperture was pierced through
the lids to admit the liquid. Underneath this aperture a glass
dish with filter paper was placed. After the liquid had been
introduced, the opening was sealed with surgical plaster which
was then coated with cellon lacquer. The edge of the lid was
also secured with surgical plaster. The jars stood in a room with
a temperature of 18—19° C. When recording the results the
dose applied is given in ml per kg of wheat and ml per jar.
The liquid was dosed by means of a i ml pipette graduated
in o.oi ml.

This series displays a few irregularities, presumably due to
the difficulty of dosing such small quantities. It is shown,
however, that with a filling of 2 kg about 0.080 ml of M per
jar, or 0.040 ml per kg is required, as against 0.045 inl per
jar, or 0.090 ml per kg with a filling of 0.50 kg. The other
values lie between these figures. The dose per jar, i.e. per unit
of fumigation space therefore decreases in proportion as there
is less wheat present, whereas the dose calculated per kg of
wheat rises considerably if the filling is decreased.

At a filling of 0.50 kg per jar the most gas is needed relativ-
ely. Let us see whether in this case the odour and taste are
affected.

It was found that, with a dose of 0.100 ml per jar, the taste
of the bread baked from this wheat was slighdy affected. The
dose required to kill the insects was 0.045 ml; the dose that
brought about a noticeable change in the taste was 0.100 ml
Therefore,

It now remains to be seen whether this margin becomes
wider if the filling of the fumigation chamber is increased. This
I investigated by fumigating equal quantities of wheat and
other products under the same conditions as regards the dose,
time and temperature, in a 3.1 litres jar (large filling) and a
gassing box of 345 litres (small filling). It was not possible to
observe any difference with any measure of certainty, however;
besides, taste is such a subjective matter. At any rate there was
not a marked difference. Some other means ought therefore to
be found for increasing the margin.

As the influence on the quality was not sufficiendy removed
by the above means, the influence of the fumigation time was

examined. For this purpose experiments were made in a series
of 3.1 Utres jars, each containing z kg of wheat, in the manner
described above. But in these tests the fumigation time was
three hours, the temperature averaging 18° C. The dose is again
given in ml per kg of wheat and in ml per jar. The insects
subjected to the test were again Calandra granaria, 50 at a time.

It is evident from these series that a dose of a little over
0.3 ml per 3.1 litres jar containing 2 kg of wheat is sufficient to

kill the Calandra. The tests with a gassing time of 24 hours
showed that under the same conditions 0.080 ml was required.
Therefore:

This reveals the surprising fact that when the fumigation
time is reduced to one eighth, only four times as large a dose
is required to produce the same effect.

According to Peters (30) the effect of a gas may be expressed
as the product of the concentration applied and the time of
exposure. According to Haber this product in different cases
amounts to a constant value if related to the same mortality.
(Haber's law: c.t = K). So, if the time t is reduced, the concen-
tration c must be increased accordingly. The results of my
investigations show that this law does not hold good in the
case of M. In how far this applies will be dealt with in a sub-
sequent chapter.

The experiments with wheat in jars were continued upon a
series in which the filling was varied; the fumigation time
invariably being three hours.

This series again proves very clearly that the required dose,
calculated per kg of wheat, increases in proportion as the
filling is smaller and that this dose, calculated per jar, decreases
in proportion as the filling is reduced.

It is now essential to know what happens to the taste and
odour of the goods in a fumigation time of three hours. It had
already been found that wheat fumigated in a jar with a filling
of kg and with 0.200 ml of M per kg, fumigation time 24
hours, produces bread with a faintly different taste. In this
connection I made an experiment in which some jars containing
14 kg of wheat were dosed with 0.8 ml per jar or 1.6 ml per kg;
fumigation time three hours. The bread baked from this was
quite normal. Now the evaporation time of 0.8 ml of M is
comparatively long in relation to the total fumigation time.
This was checked in a test allowing a fumigation time of 3 ^
hours, after it had been ascertained that the liquid evaporates
within half an hour. The bread was again quite normal. We
therefore now have the following data:

For fumigation of % kg of wheat in a 3.1 litres jar the dose
required is with a Jumgation time of 24 hours'.

the minimum dose (insects) is 0.090 ml per kg of wheat
„ maximum „ (taste) „ 0.200 „ „ „ „
with a fumigation time of 3 hours:

the minimum dose (insects) is 0.36 ml per kg of wheat
„ maximum „ (taste) gt; 1.60 „ „ „ „ „

With a fumigation time of 24 hours the taste is impaired
under these conditions if the dose is slightly more than doubled.
With a fumigation time of three hours the taste is not affected
if slightly more than four times the dose is applied. This proves,
therefore, that a short fumigation time offers considerable
advantages as compared with a longer time. This does not
only save time, but the risk of the goods being affected is
practically reduced to nil. Products of a susceptible quality can
therefore also be fumigated if a shorter fumigation time is taken.

It is certainly also important to study the eflfect of the gas
upon the eggs, larvae and pupae when a fumigation time of three
hours is allowed. For this purpose 3.1 litres were again used,
containing 2 kg of wheat infested by eggs, larvae and pupae.
To be able to make a comparison a test was also made with
areginal, while an equal quantity was left unfumigated. The
dose of M was 0.16 ml per kg of wheat, or 0.32 ml per jar and
of areginal 0.425 ml per kg or 0.85 ml per jar.

After fumigation the three samples were screened daily as
often as possible and the adults developed were counted. The
results were as follows:

Blank: 1136 adults, areginal 73 adults, M o adults. The dose
of areginal was therefore insufficient, whereas M had killed all
the stages.

During the fumigation gauze bags containing 50 adults of
Calandra granaria were laid on the bottom of the jars of
infested wheat.

Result with M: 28 % dead immediately after gassing, 100 %
dead two days later. With areginal: all the weevils seemed to
be dead immediately after fumigation, two days later 80 %
seemed to be dead and a week after the gassing 68 %. The effect
of areginal is therefore the reverse of M: immediately after
fumigation all the insects seem to be dead, but many of them
revive afterwards.

M does not kill the insects at once, but has a very strong
after-effect-, areginal causes a serious knock-down, but the insects
revive. In this respect ethene oxide resembles M, whereas
carbon disulphide has a similar effect to Areginal.

In the above test series it is shown that the dose per jar drops ac-
cording as it contains less wheat and that the dose calculated per
kg of wheat rises according as less product is present in it.

This would lead to the conclusion in practice, too, the dose
should be adjusted to the filling of the space. However, I was not
quite satisfied with the method with the glass jars, because thelayer
thickness of the wheat differs. I therefore cast about for an essent-
ially similar method with the same layer thickness in all cases.

To this end I used tins differing in capacity but containing the
same quantity of wheat, thus varying the fumigation space
instead of the quantity of wheat.

Seven tin cans, 15 cm in diameter and having a capacity of
10, 5, 3.3, 2.5, 2.0, 1.7 and 1.4 litres (see photograph 4) were
filled with I kg of wheat each. Calculated per cubic metre, this
means a filling of the various tins as tabulated below.

I used these tins in the same way as described above for the
jars; fumigation time 6 hours, temperature 20° C. The results
are given below. The mortality was determined 24 hours after
gassing; the doses are stated in ml per tin.

70
70

86
96

79
100
98
100
100

0.50
0.17

O.IO

0.07

95
100

0.50
0.17

O.IO

o.o-

0.50
0.25
0.17
0.13

O.IO
0.08
0.07

0.52
0.26

0.17
0.13

O.IO

0.08
0.07

0.42
0.26
O.I7

0.13

O.IO

0.08
0.07

0.28
0.22
0.18
0.15

O.I}

0.12
O.IO

0.42
0.28
0.18
0.14

O.II

0.09
0.08

0.35
0.26
0.18
0.14

O.II

0.09
0.08

12
85

98
100
100
100
100

10
5

3-5
2.5

1-7

1.4

As is usual in tests on insects, the results show rather con-
siderable spreading. This is the least in the last three series,
which were carried out with the most scrupulous accuracy.

The conclusion is that, to get the same mortality, the approx-
imate requirements are:

In other words: with a different filling there is no appreciable
difference in the dose required. This amounts to about 50 ml
in all cases.

The difference found in the tests with glass jars of the same
capacity and differing in filling was, therefore, actually caused
by the differences in thickness of the wheat layers; this was
eliminated in the tins.

Dug out Delphiniums, three days after fumigation. Fumigated
with M and with T gas (90 % ethene oxide 10 ^^ ^
24 hours, dose 19 g per cub. m

The method with the glass jars provides an excellent means
of comparing the rate of penetration of two different gases.

In the case of wheat the filling has no influence on the dose,
so that the absorption is equal to the quantity of gas displaced

The dose required for the fumigation of wheat is therefore
50 ml per cub. m, i.e. 46 g per cub. m at a fumigation time of
6 hours and a temperature of 20° C.

These facts were tested in two series of tests with tins, with
a dose of 40 ml per cub. m, in which the insects were examined
three times:

has hardly any influence on the dose required. This is even
more clearly evident in comparison with wheat flour.

Each tin contained kg of wheat flour. The bags of Calandra
were laid under the flour on the bottom of the tin. Some

tentative tests were first carried out with tins ofthej-a/^^ capacity.

Subsequently, three tins of different capacity were used, the
dose being 100 ml per cub. m.

In this test it is shown that, in the case of flout, the filling
did influence the results. Calculating the filling per cub. m and
taking the mortality after two days, we find:

With a larger filling the effect is therefore smaller. This
result was tested on the whole series of tins, dose icq ml per
cub. m.

The result obviously decreases with increasing filling (with
the same doses). With even the smallest filling (50 kg per cub. m)
a considerably higher dose is required than for wheat. The flour
evidently absorbs a great deal of gas. This is a case of a lt; b
(page 32).

The same was tried on peas. Each tin contained i kg; the
dose was 30 ml per cub. m.

The tendency here is obviously in the opposite direction: the
larger the filling, the greater the effect, a gt; b.
Summing up we find:

1.nbsp;^ith.peas the average concentration is higher in a filled than
in an empty space (a gt; b);

2.nbsp;With n^heat the average concentration in a filled space is
equal to that in an empty one (a = b);

3.nbsp;With flour the average concentration is lower in a filled than
in an empty space (a lt; b).

It was moteover found that flour is the most sensitive as to
odour and taste, wheat much less so and peas practically not
at all. The test method with the series of tins is therefore also
highly suitable to find out in how far odour and taste are
likely to be affected. It is also more exact than the direct method
of fumigation, working up of the product, and tasting it. The
subjective element — tasting — has a great influence in the
latter method.

It should, however, be taken into accoimt how the products
are to be used afterwards. Peas, wheat and flour are either
cooked or baked. With cocoa beans it is quite different. These
are probably slighdy more susceptible than wheat, so the effect
of fumigation is slightly lower when the filling is larger. All
the same, cocoa beans can be fumigated to any desired extent,
for they are roasted, in which process any influence of the gas
is eliminated.

It seemed desirable to test the above method on a quite
different object. For this purpose dried apricots were used.
Being slightly more sensitive to gassing than wheat, they were
expected to show a somewhat smaller effect when fumigated
with the same doses in a larger filling.

The tins were filled with kg apricots and gassed for six
hours at 20° C. See table on page 47.

Although this series shows some irregularities (tin 2.5 litres
of the first series may have been wrongly dosed), there is a
pronounced tendency towards decrease in effect with increasing
filling, but the decrease is by no means so great as with flour.
This perfectly tallies with the fact that a change in taste is
slighdy sooner noticeable than with wheat, but by no means

Dried fruit in general is one of the products that can very
well resist a six hours fumigation. The dose should however be a
litde higher than for wheat, and slightly increase with the filling.

Finally, some trials were made on wheat in the laboratory
on a larger scale to test results found above.

1.nbsp;A bag containing 75 kg of wheat was fumigated with a
dose of 54 g per cub. m in a 345 litres gassing box for six
hours at 20° C. This dose is only 20 % higher than was
found to be necessary by means of the tins. A gauze bag
containing 50 Calandra granaria was placed in the centre
of the wheat, and removed 24 hours after fumigation.
68 % were dead. A week later all the insects were dead.

2.nbsp;Two bags containing 78 and 87.5 kg of wheat were placed
together in a 345 litre gassing box and fumigated with a
dose of 100 g per cub. m under the same conditions as in
the preceding test. All the insects had died 24 hours after
fumigation.

Samples of the wheat used in these tests were ground and
baked. The bread was quite normal.

To ascertain this, a large number of flour samples were
fumigated in 345 litre boxes at 20° C. Time and dose were
varied as follows:

After the fumigation the flour was aired for 24 hours by
spreading it on a piece of paper on a table in a well ventilated
room. It was then submitted with untreated samples to the
Chemische Fabriek quot;Chefaroquot; at Rotterdam. The baking proper-
ties were found to have improved rather than deteriorated by
the fumigation. This continued to be the case when the flour
had been stored for some months after the fumigation.

Besides cereals and pulse intended for consumption, sowing-
seeds also frequently have to be freed from insects. It is, of
course, imperative that the fumigation should not deteriorate
their germinating power. Both carbon disulphide and ethene
oxide are known to have a highly detrimental effect on this.

I made extensive experiments in this direction with methallyl
chloride. To judge the germinating power, the fumigated
seeds were sown in boxes of earth, which were embedded in
the soil of a glasshouse. This gives a better judgment than with
Petri dishes in the laboratory. Seeds that seem to germinate well

in Petri dishes may quite possibly lack the strength to force
the germ through the earth. This would lead to an altogether

The seeds were fumigated in 80 litre boxes with a dose
corresponding to 125 g per cub. m, for 24 hours at i8° C.
The results were as follows:

Species

Wheat. . .
Rye . . .

Oats . . .
Buckwheat .
Caraway . .
Cole seed .
Vetch. . .
Mawseed. .
Mustard seed
Lettuce seed.
Grass seed .
Canary seed.
Sunflower seed
Peas . . •
Grey peas .
Yellow peas.
White beans.
Broad beans.

These figures show that the influence of the gas is different
on diff'erent seeds. It is absent, or at most slight, in the majority
of cases. Pulse is slighdy stimulated. With wheat, mustard and
canary seed there is a deterioration, but this is not serious,
especially when the excessive dose is considered. Yet damage
is possible when very damp seeds are fumigated. Peas stored
for 24 hours in an atmosphere saturated with water vapour
were found to have seriously deteriorated in germinating power.

Bulbs of narcissus and tubers of dahlia and begonia were
treated with a dose of 19 g per cub. m for 24 hours. They were
not damaged.

Dug-out plants of delphinium (photograph 5) and chrysan-
themum, subjected to the same treatment; the latter were slightly
damaged.

Flowers of double and single roses and of marigolds were
fumigated with a dose of 45 g per cub. m for 3 hours. The
single roses were damaged.

Aphids, ants, larvae of hover-flies (Syrphides) and a spider
were killed by this fumigation.

In general it may be said that inert parts of the plant are
not damaged, and that growing parts are damaged in some cases.

It is conceivable that, though the quality of a product is not
influenced by fumigation in an unfavourable sense, its market-
ability might be diminished. The colour might become different,
it might acquire some offensive odour, or a similar blemish. So
far, nothing of the kind has been noticed after fumigation
with M, with the exception that shelled walnuts had become
a little darker. No other deviation of any nature has ever been
found.

The general laboratory tests were terminated herewith, and
fumigation trials in practice were started.

After it had been found in the laboratory tests described above
that methallylchloride possesses satisfactory properties as
gassing insecticide, it was decided to test this material in practice.
In order to get a reliable insight into its effect, I chose test
objects varying widely in nature.

The fumigation trials were not restricted to cereals and
Calandra species, though standardized Calandra granaria from
laboratory cultures were in most cases utilized for judging the
effect.

This room is specially equipped for fumigations with HCN.
It contained the following goods:

L. and Calandra oryzae L.
2 „ „ cocoa beans, about loo kg, affected by Sitodrepa
quot; panicea L. (drug store beetie) and by Ephestia
elutella Hb. (cocoa moth).
2 „ „ peas, about 200 kg, affected by Endrosis lacteella

2 „ „ apricot stones and two cases of currants, affected
by Paralispa gularis Rag.

For checking purposes I put down a number of gauze boxes
containing wheat infested with Calandra granaria. One of these
boxes was placed in each of the four corners on the floor, one
in the middle of the floor and one in the middle half way up.
One box was fixed to the ceiling, but dropped during the
fumigation.

The liquid was poured out into four shallow zinc trays of
about 1.25 X 0.50 m. The temperature was registered with a
thermograph. It averaged 20° C. Twenty four hours after the
liquid had been poured out, the exhauster was started and
allowed to run for 48 hours. Any smell was absent by the end
of this time. All the insects were found to be dead, and no new
ones could be bred from the samples taken. The goods had
evidently been cleared of all stages.

The wheat and the apricot stones had acquired a slightly
musty smell (this fumigation was carried out before the investig-
ation into the influence on smell and teste had been terminated).

The cocoa beans were sent to a chocolate works, where they
were manufactured to chocolate. The quality had in no way
been impaired.

This loft contained thousands of larvae of the cocoa moth
(Ephestia elutella Hb.). Most of them had crept into the crevices
of the ceiling, where they abounded in layers. The loft had two
windows and three doors. There were only two bags of cocoa
beans and about 400 empty gunny bags. The space had a
capacity of about 150 cub. m.

All the openings were sealed off with gummed paper. In
different places gauze boxes containing adults of Calandra
granaria were put down. The liquid was poured into the same
trays as used in the previous test. Three were placed half way
up, one on the floor.

The total quantity of liquid poured out was lo kg, which
is almost 67 g per cub. m. The fumigation was started on a
Saturday morning at eleven and terminated on Monday morning
at nine o'clock. Scarcely any smell of gas had been noticed in
the adjoining rooms during that time. To degas the loft, doors
and windows were opened, so that almost all the gas had
disappeared within twenty minutes on account of the strong
draught. The floor was thickly covered with dead larvae, with
a few convulsive specimens among them. The Calandra in the
boxes — about 5000 in all — were all dead.

Two days afterwards I inspected the loft once more. All the
larvae were dead now, with the exception of some live ones in
the crevices of the windows. They had been covered by the
gummed paper, so that the gas had not reached them or only
insufficiently.

The gassing temperature was 14—15° C, hence comparatively
low. This test shows that a space not specially adapted for
gassing can quite well be fumigated with M without any trouble
from gas being experienced in the adjacent rooms.

It is also clear from this test that larvae lodged in the crevices
of the ceiling — hence high up in the room — are killed.

The complex consisted of eleven rooms in two buildings.
They were seriously infested with Ephestia elutella.

The gassing was carried out in two operations:
a. Building with ten rooms, total capacity 2801 cub. m.
The rooms were distributed as follows:

The building had three storeys; on the ground floor there
were offices, a passage and some unused rooms. The capacity
of the rooms was as follows:

331 y2
175 y2
331 y2

420

32
17
25
13
25

32
4

The dose was therefore about 75 g per cub. m. Only in the
crushing room, which contained much woodwork, a slightly
higher dose was applied. In most of the rooms cocoa beans
were stored.

The building was very old, and had a great many windows
and much old woodwork. It was difficult to seal off the rooms

The M was poured out into sixty trays totalling 200 sq.m.,
so that the liquid in each tray was about i cm high.

In many places gauze boxes were put down, each containing
50 Calandra granaria. The doors between the three rooms on
each floor remained open during the fumigation. The floors
were, however, separated from each other.

The fumigation was started between 10 and 11 on a Saturday
morning and stopped at 9 on Monday morning, thus lasting
for about 46 hours. The temperature was 16—17° C. Before
degassing, some samples were taken by means of evacuated
tubes.

An inspection after the fumigation showed that the floor was
covered with thousands of dead moths. Not a single moth had
survived.

Not all the liquid had evaporated, about 20 kg being left,
so that the total quantity vaporized was 197 kg or more than
70 g per cub. m.

The actual concentration did not, of course, at any moment
reach this calculated value of 70 g per cub. m. During the
fumigation process liquid evaporated regularly, but gas was
continuously lost, too. Owing to the slow evaporation a certain
concentration was regularly maintained.

The degree of concentration can approximately be deduced
from the mortality of the Calandra granaria put down. This
was found to be as follows on page 56.

It may be concluded from the mortality of Calandra that the
average concentration on the two top floors was about 15
g per cub. m and on the first floor 11 to 12 g per cub. m.

The risks of losses were much greater on the first floor,
where paper was stored and which contained much wood-
work, both of which absorb much gas. However, as Calandra
granaria is much more resistant than Ephestia elutella, it is
improbable that any of the Ephestia survived the fumigation.

Height
of this
place in
relation
to total
height

100

100
100

98
84

100
100
100
100

100

100
100
100
100

88
95

46
23
4

100
100

92
98
75

86
88

100

O
%

A remarkable result was obtained on the paper loft, where
gauze boxes were placed at three different heights.

It would seem from these figures that the concentration was
lower at the top of the loft, so that the gas must have sunk
down to some extent.

b. Building with a capacity of 1300 cub. m, consisting of one
room for crushing beans. There was a lot of woodwork, so
that the sealing was very difficult. A driving shaft ran through
a wall to another room. The hole around this shaft was closed
by stuffing it with jute sacking. In addition, boards, putty and
gummed paper were used for sealing off.

There was much cocoa powder on the walls of the conveyors
in the room, which also contained a large number of bags of
cocoa beans. For checking purposes eight boxes, each with 50
Calandra granaria, were placed on various spots.

By way of trial woollen flannel cloths were hung over two
troughs, so that they hung down one metre on all sides. Evapor-
ation was greatly accelerated by this.

The fumigation was started between twelve and one o'clock
on Saturday and finished at 8.15 on Tuesday, so that it lasted
about 67 hours in all. The liquid had then completely evapor-
ated. Fumigation temperature 18—19° C. The floor was strewn
with dead moths, none had survived. All the Calandra were
also dead. A quantity of cocoa powder was taken from the
conveyors and placed in the laboratory incubator at 28° C.

These two fumigations were carried out by the end of May.
In the period from May to November moths were found only
sporadically in the works; a pest as in previous years was out
of the question. This proves that this factory can be kept free
from moths by gassing once a year or perhaps once every other
year.

The building was very old and seemed at first impossible to
seal off. Yet a fumigation was decided upon, which meant a
very severe trial to the fumigant.

In the warehouse base materials for a confectionery were
stored, which were affected by Ephestia elutella and Paralispa
gularis. It had four storeys, the first of which was divided into
smaller rooms (offices etc.).

In order to ascertain whether the quality of the stored
products could be influenced by the gas, I fumigated samples
in the laboratory. These samples consisted of the following
products, 2 kg of each: almonds, apricot stones, shelled hazel-
nuts, grated cocoanut, shelled walnuts, sultanas, American
raisins, muscatels, currants, candied peel, moist sugar, corn-
flour, wheat-flour, corn flakes and ginger.

All these samples were spread out on wooden grids with
paper in a 345 litre gassing box. Ephestia larvae and adults of
Tribolium confusum were also placed there for checking
purposes.

The fumigation took place with a dose of 100 g per cub. m
for 6 hours at 20° C.

The insects were immediately killed. The products were
worked up to cake and pastry in the usual manner. They proved
to be unaffected, the only exception being that the wheat-flour

had become slightly musty in taste and that the walnuts had
become a little darker.

The fumigation could, therefore, be started after the removal
of the stored wheat-flour.

The building was closed off as well as possible, but this could
not be done completely. The storeys could not be closed
separately, so that they remained in connection with each
other.

The furniture in the offices: chairs, writing tables, carpets,
heavy curtains, etc., were left in the building.

As the fumigation was not allowed to last longer than six
hours, the method of slow evaporation could not be applied.
Instead, the liquid was atomized by means of fine atomizers and
cylinders of compressed air.

first „

2nd

3rd

loft. . .

For checking purposes, gauze boxes were put down here
and there, containing adults of Calandra granaria and larvae
of Ephestia kuhniella and Borkhausenia pseudospretella.

The liquid was atomized very rapidly, the whole building
being filled gas within an hour. Six hours later doors and
windows were opened. The top floors were free from gas within
a few minutes, the lower took much longer. The gas had
evidently sunk down.

moths were found, which were also dead after 36 hours. All
the flies and spiders had been killed as well; moreover, I found
a number of Anobium species, apparently from the old beams.

A number of dead moths (Tineola biselliella Hum.) were
found near the office curtains.

No damage had been done to the furniture.
The insects in the gauze boxes were inspected 40 hours after
furnigation, the result being:

This clearly proves that the gas had sunk down.
When some floors in a building have to be gassed simul-
taneously all communication between them should therefore
be prevented. In very high spaces the gas should be kept
moving by fans.

By the special kindness of Mr. Barges at Le Havre I had an
opportunity of making a series of tests in the vacuum installation
of the quot;Station de désinfection des produits végétauxquot;.

This installation works as follows: the goods are passed into
a vacuum tank, in which pressure is reduced by about 650 mm.
The liquid is drawn in by suction, passing through a coil in a
hot water bath and consequently quickly evaporating. The

admittance of the gas reduces the vacuum to about 200 mm,
upon which the air is allowed to rush in so as to get an mtimate
air gas mixture. The gas consequently penetrates very rapidly

Fumigation is terminated by twice applying vacuum and
admitting air, owing to which the goods are practically freed
from gas.

I will only mention here a large-scale test in a 29 cub. m tank.
It contained 5000 kg of maize, seriously infested by Calandra
oryzae and Tribolium spec, and a bag of wheat, seriously
affected by Calandra granaria. Dose 74 g per cub. m, for 6

After the fumigation some insects only showed slight convul-
sions, the majority was dead.

The batch of maize did not remain at our disposal, but we
kept the wheat. Three months later not a single msect was to
be found in it. A small quantity of potatoes with three larvae
of the Colorado beetie (Leptinotarsa decemlineata) were gassed
at the same time. The insects were killed, but the potatoes were
damaged, as they could not develop to good plants.

VI. LARGER QUANTITIES OF SOME PRODUCTS
FUMIGATED IN 345 LITRE LABORATORY GASSING BOXES

The boxes were crammed with skins. These skins contamed
a good deal of grease, and a M dissolves in this, much loss

70 g per cub. m, applied for 24 hours, amply sufficed to kill
all the insects. The fur remained absolutely unaltered. This
result induced a fur dealer to construct a fumigation chamber,
in which many batches of skins are now being successfully
treated.

b. Three bales of Taschowa tobacco, each weighing about 20 kg,
were obtained from a tobacco company.
One of these bales was gassed with a dose of 120 g per cub. m
for 3 hours. The bales were then stored for four months.
It was impossible for experts to distinguish the gassed bale
from the other two, so that the quality had evidently remained
absolutely unaltered.

Dose 70 g per cub. m, time 24 hours, temp. 20° C. All the
insects dead after fumigation. The dose might have been
lower.

Dose 67 g per cub. m, time 46 hours, temp. 14—15° C. All
the insects dead after fumigation. Dose amply sufficient.

Dose 75 g per cub. m, time 46 hours, temp. 16—17° C.
After fumigation: moths dead, not all the Calandra.
Dose exactly sufficient.
b. 1300 cub. m.

Dose 90 g per cub. m, time 67 hours, temp. 18—19° C.
All the insects dead after fumigation. Dose sufficient.

Dose 70 g per cub. m, time 6 hours, temp, about 20° C.
Gas had sunk; the dose was presumably sufficient.

Dose 74 g per cub. m, time 6 hours, temp. 19° C. All the
insects dead after fumigation; dose amply sufficient.

per cub. m for 24 hours.
b. The quality of the tobacco is presumably not altered by
the fumigation.

In all the tests gas masks were used provided with eyeglasses
and with canister A of the Degea. The masks permit of a
prolonged stay in the rooms where the liquid is poured out,
because the concentration never becomes high.

Great caution is, however, imperative when the liquid is
atomized (test IV), because in this case a high concentration
is rapidly attained. Some mist is then formed as well, so that
the gas mask should be fitted with a mist filter. In this case
Otter and Groenendijk's canister, of Belgian origin, may for
instance, be used.

Anyhow it is desirable to leave the space immediately after
opening the nozzles. A stay of longer than 30 minutes in the
gassing space with the same canister is impossible.

All fumigations, with M as well as with other gassing insect-
icides, in which the gassing space has to be entered, should be
carried out by more than one person: preferably two entering
the room and one staying outside.

The toxicity of M for higher animals in comparison with
that of other gases is discussed in Chapter IV.

It has been stated before that M has a very strong after-
effect. Immediately after the gassing not all the insects are dead;
they die successively, however, in the hours and days following
the fumigation. For this reason the insects were observed for
a week after each gassing. Even after this the after-effect is
presumably not yet spent.

M has, however, another highly important property, which
was noticed in the tests.

It is known that insects shield themselves from some in-
secticides by so-called protective stupefaction. They are rigid

and hardly move or breathe. The effect of the gassing is, of
course, greatly lessened by this. HCN has this influence and
for this reason frequently substances are added which irritate
the insects, such as various methyl esters. The drawback is that
the irritating substances do not penetrate sufficiently, so that
the insects are not reached. In the present experiments I found
that M itself possesses strongly irritating properties. Even a
low concentration causes the insects to become greatly disturbed
and to emerge from their hiding places.

In trial II, on the cocoa loft, thousands of larvae were lodged
in the crevices of the ceiling. They had already started spinning
their cocoons, one clinging to the other, thus forming a cluster
of many layers. At the end of the fumigation it was found that
nearly all of them had come out and dropped to the floor.

In fumigating the cocoa works it was observed through the
windows that the moths, which had at first kept quiet in their
hiding places, began to flutter about wildly as soon as the gas
was introduced.

Upon gassing of some pieces of furniture numerous larvae
and adults of Anobium species appeared. Larvae ot Sitodrepa
panicea emerged when cattle cakes were treated.

M consequentiy proves at the same time to possess insecticidal
and irritating properties.

This may be of great importance for instance in the case of
dwelling fumigation. The bugs, which mostly hide in inaccessible
places, will also be roused up and thus are sure to be reached
by the gas.

This can be applied in buildings that are fumigated in the
weekend and in sheds where goods are stored whose smell

or taste is unaffected by the gas, such as cocoa beans and goods
that are not consumed, such as skins.

With a fumigation time of 24 hours the actual concentration
must be about 20 g per cub. m; with a time of 40 hours and
more, about 12 to 15 g per cub. m.

The dose entirely depends on the circumstances. In specially
constructed gassing chambers and vacuum installations a dose
of 30—40 g per cub. m will in most cases be found to suffice.

For products such as skins, the hairs of which absorb a large
quantity of gas which, moreover, dissolves in the grease, a
substantially higher dose — about 70 g per cub. m—is required.

In buildings containing a great deal of woodwork or which
cannot be completely sealed off the dose should also be higher.
It should be adopted to the prevailing conditions.

When there is a great risk of losses, it is advisable to evaporate
the liquid slowly by pouring it out to a height of i cm in flat
trays. This provides a constant supply of gas; the concentration
consequently is not very high, the losses are less great and are
constantly replaced.

The troughs should be placed as high up as possible. If
necessary, evaporation can be accelerated by hanging cloths of
woollen flannel in the trays and over the edges, which should
then not be higher than 4 cm.

Before the fumigation the space should be closed off as well
as possible. Gummed paper, thin wood or putty can be used
for this purpose. Large holes can be stuffed with moist gunny
bags.

When fumigation has been started, thorough inspection
should be made on the outside to ascertain whether any gas
escapes, which is at once noticeable from the smell.

The person entering the room to degas it should wear a
gas mask and open all the doors and windows.

For many products it is advisable to choose a short fumig-
ation time — say of 8 hours — so as to prevent any disagreeable
effect on odour or taste. A dose of about 65 g per cub. m suffices
when the fumigation space is thoroughly sealed off. In cases
where losses are to be feared the dose should be increased
accordingly.

The best way of supplying the gas is by atomizing it finely;
m vacuum installations the liquid can be conducted through
a heated coil. For the atomization low-pressure steel cylinders
can be used, filled with liquid and air to about 20 atm. (photo-
graph 7).

In calculating the dose use can be made of the gram-hour
value, i.e. the number of grams applied per cub. m. multiplied
by the time (see also the next chapter). Under practical con-
ditions this is about 500—1000 for M. With a fumigation time
of six hours in a well-closed installation the dose is about
500:6 = approximately 80 g per cub. m; with a fumigation time
of eight hours 500 : 8 = approximately 60 g per cub. m. The
gram-hour value rises greatly when much absorbent material
or many other factors causing loss are present and when the time
of fumigation is much longer. For a fumigation time of 24 hours,
20 g per cub. m is not sufficient, but should be about 40 g per
cub. m, which corresponds to 960 gram-hours. This has been
worked out in Chapter IV.

Both in the general laboratory investigation and in the
practical experiments it was found that methallyl chloride is very
valuable for the control of noxious insects.

During the investigation, however, a number of questions
partially of practical and partially of theoretical importance
remained open.

At this moment, however, it is not possible to furnish a
complete explanation for all questions which arise during the
use of M. As far as possible I will give a number of preliminary
conclusions, while further information will be supplied in
following publications.

The first question which arises is: quot;Does methallyl chloride
provide any advantages over other known insecticides and if
so, why?quot;

Peters (34) denies these advantages in an paper which may
be regarded as a reply to my short note in Nature (25).

He apparently judges the value of a gas insecticide entirely
by figures obtained in the laboratory. It is regrettable that he
omitted to mention in what manner and in how many exper-
iments these figures were obtained. For that reason his con-

elusion that methallyl chloride compares very unfavourably with
ethene oxide and even with carbon disulphide is by no means
convincing.

Peters regards the „Grammstundeneinheitquot; (gram-hour unit)
as the most important value. It is clear that if the fumigation
period is shortened, the concentration should be increased. Accord-
ing to many authors this is directly proportional, c X t therefore
is a constant value (Haber's law).

Already in 1936 Peters (30) suggested expressing the value
of a gas in quot;gram-hourquot; units. The smaller the activity of the
gas, the more quot;gram-hourquot; units will be required to obtain
a good effect. Later Peters stated (35) that this only applies to
HCN and not to other gases.

To this end fumigations were first carried out in gassing
boxes with a capacity of 80 litres. Fifty Tribolium confusum
Duv. were placed in each gassing box and fumigated for
different periods.

This shows that for a fumigation time of 24 hours about i. 5 ml
is required and for 3 hours about 7 ml. It should be taken into
account that the time required to evaporate 7 ml of M is fairly
long as compared to the entire fumigation time of 3 hours.

I attempted to obviate this difficulty by the following pro-
cedure. A fan was placed in a 345 litres gassing box described
elsewhere. A small pump with a capacity of about 4 litres per
min. was connected to the gassing box. This pump removed
the gas from the box, caused it to pass through three wash
bottles (without liquid) and back to the gassing box.

At the beginning of the experiment the tube leading to the
pump was closed. The gassing box was also closed and the
required quantity of liquid was introduced through the aperture
at the top. Then the fan was switched on and kept rutming
during the experiment. After the liquid had evaporated com-
pletely, a gauze bag containing 50 Tribolium was placed in each
of the three wash bottles.

Then the tube leading to the pump was opened and the latter
set going. At this moment the fumigation actually started.
Fumigation was carried out for three hours and for six hours.

Using the value for a fumigation period of 24 hours obtained
in the previous series, the following rough values are obtained:

A remarkable thing is that with higher concentrations and
therefore with a shorter period of activity the gram-hour values
are lower.

Here the values for 24 hours, 6 hours and 3 hours are related
in a proportion to each other of about 2.15 : 1.17: i.

The following values (for Calandra granaria) were found on
fumigation of 2 kg of wheat in glass jars of 3.1 litres:

Here the gram-hour values are higher, because the gas had
to penetrate through a layer of wheat. The proportion, however,
is practically equal to that in the gassing boxes.

There is an indication here that the gram-hour value indeed
decreases as the concentration becomes higher. In that case
the gas has a better effect in higher concentrations and with
shorter fumigation periods than in lower concentrations.

However, these are only rough experiments; in order to
obtain more accurate results they should be worked out in detail.
The most important objections which may be raised are:

1.nbsp;If a liquid is evaporated in a space where atmospheric pres-
sure prevails, a slight excess of pressure is the result. In the
wooden boxes this will cause a loss of gas owing to the
mixture partly escaping.

2.nbsp;The liquid takes sometime to evaporate, as a result of which
the time during which the gas is effective cannot be deter-
mined accurately.

In view of this criticism apparatus had to be constructed
meeting the following requirements.

1.nbsp;The evaporation of the liquid may not cause excess of
pressure. This may be obtained by evaporating it in a space
where part of the air has previously been removed (vacuum).

2.nbsp;The insects may only be placed in the apparatus after all
the liquid has evaporated.

only very little gas.
4. The gas-air mixtures should be continuously kept in motion.

C and d are glass botdes each with a capacity of 10 litres.
These bottles are connected at the top by means of a glass tube.
A wash botde h is connected to botde r, b is connected to the
pump ƒ by means of a tube, the other end of which is connected
to a series of wash botdes i to 5 and also to the glass tube e.
The other end of this conglomerate is connected to the bottom
of bottle d.

Bottles I to 5 are glass wash bottles with a capacity of 20 ml,
provided with a double-pierced, ground-in glass cover.

Pump ƒ is a tube-pump, which consists of a length of rubber
tubing tightly stretched round three cams fixed to a rotating
disk. The capacity of this pump is about 4 litres per minute.

A duplicate of the apparatus is built symmetrically to the
described one. The four bottles are placed in a water bath. The
two pumps are driven by the same motor. The whole apparatus
is placed in a room with a constant temperature of 20° C.

The wash bottle b is removed and the resulting break in the
apparatus is connected up. The three-way cocks I to VI are
adjusted so that bottles i to j are turned off and tube e is open.

A vacuum pump is then connected to the three-way cock a.
This pump reduces the pressure in the bottles c and d to about
10 mm Hg. The liquid required is introduced into the wash
bottle b, which is removed. As soon as the required vacuum
is obtained this wash bottle is again inserted in the system. Air
is then allowed to enter through the cock a, via the wash
bottle. The liquid evaporates and this vapour is forced into
the bottles by the current of air. Care must be taken that all
liquid is evaporated before atmospheric pressure in the bottle is
restored. Then cock a is closed and the pump is started. The

Meanwhile 50 insects have been placed in each of the wash
bottles. After the gas has been circulated long enough these
bottles are inserted in the system and cocks I to VI adjusted
so that the current of gas does not pass through tube e but
through the botries containing the insects. It is now possible to

2 have been removed, while the three-way cocks I and II have
been adjusted so that the gas reaches botde 5 via part of tube e.

With this apparatus it is made possible to apply the same
concentration for different periods. So it is possible to determine
accurately the period required to obtain 100 % kill at the
concentration applied.

For the following experiments I always used adults of Calandra
granaria, one month old. The kill was determined one week after
fumigation. Each botde contained 5 o insects for each experiment.

The fumigation time was varied for the same dose. The
temperature was always 20° C.

Each column represents one series of experiments. Total
number of insects used was 3000.

It will be seen that the average kill for 3 hours and 20 minutes
is 99-5 %gt; while this is 100 % in 11 out of the 12 series.

The quot;gram-hoursquot; value must therefore be somewhat, but
only a litde, higher than 3% x 60.1 = 200. This experiment
was repeated with a somewhat higher dose.

In this case a kill of 99.4 % was obtained with a fumigation
time of 3 hours and 10 minutes, while with a fumigation time
of 3 hours and 20 minutes all the insects were killed.

The gram-hour value therefore lies between 205 and 217.
This result corresponds very well with that of the previous
series of experiments.

I also carried out a number of other experiments with
different fumigation times.

A kill of 99.7 % was therefore obtained after a fumigation
period of i hour and 10 minutes. The gram-hour value therefore
lies somewhat higher than i^/e X 165 = 193.

Total number of insects used was about 2250. A kill of 99.6 %
was obtained with a fumigation period of 2 hours and 10
minutes and a kill of 100 % after 2 hours and 20 minutes.
The gram-hour value therefore lies between 200 and 216.

Total number of insects used was about 3000. The kill is
99.8 % with a fumigation time of 4 hours, so that the gram-hour
value lies somewhat higher than 210.

The gram-hour value of methallyl chloride, therefore, is
about 200, while moreover it is possible that this value becomes
slightly higher if the fumigation time is longer.

Let us realize here exactly what is represented by the gram-
hour value. This is the figure obtained by multiplying the
number of grams of fumigant which are supplied per cub. m.
by the number of hours for which the insects must be exposed
to obtain 100 % kill.

It will be evident that this g-am-hour value is quite different
from the Haber value, which is obtained by multiplying the
number of milligrams per litre of actual concentration by the
number of minutes of the fumigation period. One is only

allowed to compute the Haber value by simple multiplication
if the concentration remains constant during the entire fumig-
ation period. The Haber value may also be expressed in grams
per cub. m. and hours. I should like to refer to this figure as
the c.t value. This c.t value will always be lower than the gram-
hour value, because of the latter including the over dose needed
to compensate for losses by sorption and leakage.

According to Haber this c.t figure amounts to a constant
value for a certain percentage of kill (c X t = K).

The gram-hour value, on the contrary, depends upon the
circumstances under which the fumigant is applied. The more
favourable these circumstances, i.e. the fewer the losses, the
more the gram-hour value will approximate the c.t value.

In order to obtain some idea of the extent of the losses in
my apparatus (by sorption to the walls of the containers), I took
samples of the gas after the apparatus had been filled with
70 ml of M per cub. m or 64.7 g. An average of 51 g per cub. m
was determined in these samples of gas. The loss therefore
was 13.7 g or about 21 % of the dose.

A number of measurements were furthermore carried out
with a similar apparatus, dosed with 100 g per cub. m. In this
case the partial pressure of methallyl chloride is equal to 27 cm
of water pressure. With the aid of a water manometer it was
observed that after 2 hours the pressure drop was about 6.5 cm
and after three hours about 7.5 cm. This corresponds to a loss
in concentration of M of 24 % to 28 %.

These two determinations show that the loss in concentration
is roughly 20 to 25 %, so that the actual concentration is about
75 to 80 % of the dose.

The quot;approximated c.t valuequot; of M for Calandra granaria is
therefore 75 to 80 % of 200 or 150 to 160, for a kill of 100 %.

I also carried out a number of tentative determinations with
carbon disulphide in the same apparatus. The dose was 100
g per cub. m.

Time

3 hours — min.

3nbsp;» 3° »

4nbsp;» — jj

4nbsp;„ 3° »

5nbsp;53

This shows that the kill is satisfactory only after a fumigation
period of four hours, whereas even after five hours the kill
obtained is not definitely loo % yet. The gram-hour value of
carbon disulphide for Calandra granaria, determined in this
apparatus therefore lies in the order of magnitude of 400.

For methallyl chloride Peters (34) gives about 275 and for
carbon disulphide about 195, as may be read from his graphs.
It is not evident, however, whether these figures represent
gram-hour values or approximated c.t values. In any case the
proportion of M to CSg as 275 : 195 is decidedly incorrect.
I feel almost sure of the cause of this incorrectness.The after
effect of M is very strong, while carbon disulphide, on the other
hand, shows a strong knock-down effect, but the insects revive
again. Probably Peters did not keep the insects under observation
long enough.

It is striking how very littie information regarding the
insects used is supplied by non-biologists carrying out experi-
ments upon insects. What age were the insects used? Were
they standardized? At what temperature were they kept?

In a paper by Peters and Ganter (36) they state that, for their
experiments, they used Calandra granaria quot;from the same
culture, of the same size, of the same colour (tiefschwarz), of
the same feeding condition and of the same temper (measured
by their intensity of motion)quot;! — In my opinion these are not
the essential facts for the choice of one's tests insects. Besides,
the number of insects which these authors use in one trial
_ mostly ten — must be considered insufficient.

The value of many, otherwise excellent, publications is
decreased by the fact that in many cases incorrect criteria are
taken into account as regards the test objects! —

We now have the disposal of the approximated c x t value
of M, which is about i6o.

For ethene oxide Peters gives about 60. But may this be
regarded as proof that the latter is a much stronger gaseous
insecticide than the former?

Here now the difference between gram-hour value and c.t
value becomes evident. The latter value is an entirely theoretical
one, which only can be determined approximately in laboratory
trials. An insecticide, however, is intended to be used m practice.

The dose of ethene oxide which is used in actual practice is
50 g for 20 to 24 hours, which corresponds to 1000 to 1200
gram-hours. This figure is out of all proportion to the value
of 60 gram-hours found in the laboratory. The losses caused
by adsorption and other factors are about 20 times as much
as this value. The practical value can only be determined by
using the material under varying conditions in actual practice
or under conditions which are nearest to those prevailing in
actual practice. This leads us to a second shortcoming of many
publications on gaseous insecticides, viz. their lack of practical
examples. The final evaluation is only possible in actual practice,
laboratory experiments can only be regarded as tentative. Now
we shall compare trials in actual practice with M with the
prescribed practical dose of ethene oxide. In experiment V
(page 60) carried out in the vacuum apparatus the dose was
74 g per cub. m for six hours, which corresponds to 444
gram-hours.

In the very unfavourable case of trial IV (p. 58) the dose
was 100 g per cub. m and the fumigation time 6 hours. This
makes 600 gram-hours.

In experiment I in the gassing chamber the dose applied was
70 g per cub. m and the fumigation period 24 hours. This is

equal to 1680 gram-hours. In this case all the insects were dead
immediately after fumigation. This dose was therefore much
higher than necessary.

The other experiments, which were carried out during the
week-end, lasted longer than was really necessary, so it is not
possible to judge by their results.

The bags of wheat in the gassing boxes in the laboratory
were fumigated for six hours with doses of 54 g and 100 g per
cub. m, corresponding to 324 and 600 gram-hours. These
practical experiments show that the order of magnitude of the
gram-hours of M required in actual practice must be below looo.

How is this possible ? Probably because ethene oxide dissolves
readily in water and M does not. As a result of this, considerable
losses may occur which must be compensated by increasing
the dose.

In his publication Peters (34) gives a practical dose of6o to 70
g per cub. m for M. He does not mention in what manner these
figures were obtained.

According to Peters the gram-hour values of ethene oxide
and M determined in the laboratory are 60 and 275, respectively.
For actual practice he gives 1200, and 1440—1680 (50 g per
cub. m and 60—70 g per cub. m for 24 hours).

His laboratory value for carbon disulphide is 195 gram-hours
and for actual practice he advises a dose of 200—500 g per
cub. m or 4800—12000 gram-hours. According to the in-
vestigations carried out in his laboratory the proportion of
ethene oxide to methallyl chloride to carbon disulphide is
60 : 275 : 195. In actual practice this proportion is 1200 : 1440 :
4800 or 60 : 72 : 240.

The above — with the aid of data supplied by Peters him-
self — proves that the gram-hour value determined in the
laboratory under ideal conditions cannot be regarded as a
measure for the practical value of a gaseous insecticide. These
kinds of determination are at the most of theoretical importance.

Since it is necessary to have some preliminary information
about the relative value of a new insecticide before testing it in
actual practice, one must have a laboratory method at one's
disposal. As such may be considered the experiments with
gassing boxes as discussed in Chapter II. These experiments
were carried out imder quot;reducedquot; practical conditions, as a
result of which a much better idea of the practical value of the
gas is obtained than in experiments in apparatus which may
give very accurate results, but which do not provide a true
picture of what really happens in actual practice.

We now come to the question as to whether Haber's law may
be applied to M. In the experiments with the accurate laboratory
apparatus (page 72 ff) the differences in the results obtained
with fumigation periods of 1—4 hours are only very small.
For this range of time Haber's law therefore applies. It was
not ascertained whether this also applies to a wider range, but
this is of very little practical importance.

In the experiments with glass jars filled with wheat (p. 38)
it was found that the law did not apply. The figures obtained
here, however, were gram-hour values. Now Haber's law does
not apply to gram-hour values. So the above statement might
be expected. The conclusion that may be drawn from this is
that the gram-hour values increase with increasing fumigation
times.

Considerable deviations were also noticed in the experiments
the aid of gassing boxes described in the beginning of this
chapter. In both cases the gram-hour values for fumigation
periods of 24 hours and 6 hours bear a proportion to each
other of roughly 2:1.

It may therefore be expected that also in actual practice the
gram-hour value will increase as the fumigation period is longer.

the experiments. It would be very convenient, however, if they
were also known for the intermediate periods. These may be
deduced from the values already known. The figures obtained
in the experiments with gassing boxes were chosen for this
purpose (page 70). These are admittedly only very rough
figures, but for practical purposes they are sufficiently accurate.
The proportion for 24 hours, 6 hours and 3 hours was 2.15 :
1.17 : I. If, in view of the practical experiments carried out,
we take a gram-hour value of 480 for six hours, the three values
are 881, 480 and 410. If these figures are plotted in a graph a
straight line may be drawn through these points. The other
gram-hour values may then be read off (graph page 96), for
instance, for a fumigation period of 12 hours the gram-hour
value is 610. The required dose is then 51 g pet cub. m.

The figures apply to quot;averagequot; cases, for instance for fumi-
gation of not highly absorbent products in properly closed spaces.
In unfavourable cases they will be higher, in favourable cases
lower. For such cases lines have been drawn in the graph,
starting from 50 g per cub. m and 100 g per cub. m applied for
six hours. All gram-hour values which may be considered for
fumigation with M may be found between these two lines.
With the aid of the series of tins (page 41) it may be found out
very easily whether a product should be classed among the
favourable or the less favourable cases. As far as the fumigation
spaces are concerned, vacuum installations and gas boxes
should be classed among the favourable cases, old ware-houses
among the unfavourable. The lowest dose will be required for
fumigation of pulse in a vacuum installation, a much higher one
for furs in an old ware-house.

Methallyl chloride is an inflammable liquid. Its vapour is
explosive if mixed with air in certain proportions.

Its explosion range was determined in a tubular vessel (tube
test) and was found to be from 105 g per cub. m to 339 g per
cub. m.

When determined in a globular vessel (globe test) this range
was found to be from 93 g per cub. m to 375 g per cub. m.

So in the most unfavourable case the gas is explosive if the
concentration of 93 g per cub. m is exceeded. Below that
concentration it is quite safe.

This concentration however, is never obtained in actual
practice. Here again a distinction should be made between dose
and concentration. Insects are killed by a concentration of 25 g, for
instance, during six hours (c X t = 150). In order to obtain
such a concentration, however, a much higher dose should be
applied. If a dose of 100 g per cub. m is applied in an old ware-
house this does not mean that the work is being carried out
within the explosion range, since the concentration is much lower
than 100 g per cub. m.

It is possible, however, that owing to the gas settling down
because of its weight a higher concentration is produced in
some places.

For that reason care should be taken that the gas does not
come into contact with fire; this applies to other fumigants.

Methallyl chloride may be made non-inflammable by adding
carbon tetrachloride to it. Experiments are being carried out
to find better materials for this purpose.

The influence of the temperature on the effect of the fumi-
gation was not studied separately. The temperature in practice
experiment No. II was 14—15° C. The effect of the gas at that
temperature was very good. Insects are not likely to occur in a
harmful degree at temperatures under 14° C, so that in my
opinion it was of very little use to carry out experiments at
lower temperatures.

Some kinds of insects are easier to kill with M than others.
To obtain some idea about the extent of these differences
I made experiments with several species of insects in gassing
boxes of 80 litres. These insects all originated from well kept
laboratory cultures, though they were not all of the same age.
Fumigation time 24 hours, temperature 20° C.

II.6

11.6
12.5

14.5
14.5

14.5
16.9
18.0
18.5
18.75

20.0
20.2

21.0/ perhaps no

21.01nbsp;minimum
26.0 ( value
26.0

Hofmannophila pseudospretella Stt. All stages
Ephestia kuehniella Zell. All stages
Calandra granaria L. All stages ....
Cimex lectularius L. All stages ....
Lasioderma serricorne F. Adult ....
Sitodrepa panicea L. All stages ....

The list shows that the doses required vary from 11.6 to
21 g per cub. m, except for Trogoderma species, which,
apparently, are an extreme case.

If one were to take all these differences into account, it would
involve a procedure too intricate for actual practice. For my
experiments I, therefore, chose one object, viz. Calandra gra-

naria L. which, as is shown in the above list, is one of the
highly resistant species, needing 18.75 g pet cub. m. Applying
the dose calculated for Calandra granaria, all the less resistant
species will be killed as well. As for the more resistant species,
they show but a slight deviation from Calandra granaria. Since
in practice a considerably higher dose must always be applied
to compensate for loss oftoncentration by sorption and leakage,
the slight differences in sensibility are, at the same time, made up
for.

Only if Trogoderma is concerned it might be necessary to
apply a higher dose, whereas in cases such as Araecerus a
somewhat lower dose would be sufficient.

As a last point of investigation I will report here some
observations about the toxicity of methallyl chloride to mam-
mals, in comparison to the toxicity of ethene oxide and carbon
disulphide.

A gas which is poisonous to insects is also poisonous to
higher animals and it will never be possible to find a gas which
will kill insects, but which may be inhaled by higher animals
or human beings without harm. In this connection I may
refer to Peters (35) who remarks: „We shall never see but can
only dream of the man who can gas his dwelling for moth and
bed bugs, but, at the same time, remain in the rooms that are
being fumigated.quot;

It would be foolish to reject these means of control for that
reason. A great number of poisons are used by man for all
purposes and to the great benefit of society. Poisonous gases
used for the control of insects also belong to this category and
if the necessary precautions are taken, there is no danger
whatever in the application of these products.

In this respect I quite agree with what C. L. Metcalf says
in his paper: quot;Bugaboo or Backbonequot; (37).

We entomologists may not and cannot tolerate that our best
weapons for the fight against insects are knocked out of our
hands on loose grounds. It should be considered that insect
control is of vast importance. It has not yet been properly realized
that a gigantic battle is being waged between two large groups
of the animal world: man versus insect, and it is not at all
certain that man will conquer in the long run. Metcalf rightly
remarks that it is mainly the fear for the unknown that makes
people hesitate to use poisonous gases. There is a pipe-system
in every house for coal-gas, which is one of the most poisonous
gases....!

Our aim should be to find insecticides which are less poisonous
to human beings than HCN and ethene oxide, for instance, and
yet possess good insecticidal activity.

What is the position of methallyl chloride in this respect ?
I carried out experiments with this product on white rats in
comparison with ethene oxide and carbon disulphide. The
animals used originated from two breeders. One batch was
from a carefully selected breed, while the other batch was derived
from a commercial source. No difference was noticed between
the experiments upon these two batches.

In the laboratory they were kept at a constant temperature
of 22° C. They were fed on 1/3 of wheat meal and % of whole-
milk powder. The cages were provided with running water.
The weight and the behaviour of the animals was checked
regularly. Abnormal specimens were removed.

For fumigating these animals I devised the following appa-
ratus. A is a glass aquarium tank with a capacity of about 24
litres, with a loose, three-ply wood cover. The bottom of this
cover is provided with soft rubber. It is clamped on to the
aquarium by means of 10 rubber bands. A glass, cylinder B is
fixed in the cover and is big enough to hold a rat of about 200
grams. This cylinder is closed at the top and bottom by means

of rubber stoppers. The stopper at the top has two holes to take
the thermometer t and the glass tube a. Underneath the aquarium
tank there are two bottles C and D with a capacity of about lo
litres each. The two botdes are connected by means of a glass
tube, while a glass tube ƒ runs from C to the bottom of the
tank. The bottle D is connected to the tube-pump E, the other
side of which is connected to the aquarium and the cylinder via
the wash bottle F.

At the beginning of the experiment C and D are evacuated.
Then the liquid to be evaporated is pipetted into the wash
botde, which is placed in a hot water bath. Air is admitted
through b, while c is closed. The evaporated liquid is carried
into the bottles C and D by the air, which process may be
checked by means of the manometer G. All the liquid should
be evaporated before the atmospheric pressure in the two
bottles has been restored. As soon as this the case b is closed
and c and d are opened, while a remains closed.

Now the pump is started. The gas circulates through A, C
and D until a homogeneous mixture is obtained. Then B is
opened at the top and a rat is placed inside the cylinder. It is
closed again and the bottom stopper is pushed out by means
of the thermometer. The rat then drops into the tank. The
cylinder is therefore used as a sluice. The animals remained
in the apparatus, which had a total capacity of 46 litres, for 25
minutes at the most. The rubber strips could be removed very
quickly, so that the animals could be easily taken out at the
end of the experiment. (Fig. 3, photograph 6).

In this apparatus rats were gassed with M, ethene oxide and
carbon disulphide. In all cases the dose was 100 g per cub. m,
the temperature was 20—21° C.

They ran around restlessly sniffing and blinking. It was
evident that they found it difficult to breathe. The movements

Fig- 3

became less and less decisive and after a few minutes they became
unconscious and were lying on their sides. Generally foam
appeared at the nose. After fumigation they came to within
a few minutes. Legs and ears were bright red, later they acquired
a bluish colour.

Dissection of the succumbed ones showed that the lungs
had been affected (filled with liquid, lung oedema).

The animals appear to remain normal. There is nothing
peculiar to be seen. Dissection showed that the lungs and
sometimes also the liver had been affected.

They run around, are restless and their movements are
spasmodic. In many cases they become very wild and after-
wards unconscious. After they have been taken from the tank
they sometimes die with a shock. In other cases they were
temporarily or permanently paralysed.

Immediately after fumigation it was found that their weight
had decreased a little. It is remarkable that in many cases a
marked increase of weight occurred a short time after fumigation,
which was especially the case with M.

The results of the fumigation are given in the table on page 90.
The sign denotes that the animals succumbed as a result of
the fumigation. Generally death occurred one or a few days
after fumigation.

From these results it is clear that there is a very great
difference between the effects of ethene oxide and M and that
the latter is only a litde more poisonous to higher animals than
is carbon disulphide.

With ethene oxide the animals can stand at the most a
fumigation period of 1I/2 minutes, while this period is 15
minutes in the case of M and 17^ minutes in the case of carbon
disulphide. In all cases the dose was 100 g per cub. m.

Furthermore the irritating odour of M and the irritating
effect upon the eyes are a clear warning that this gas is present.
It is out of the question that any one would by accident enter
a room which is being fumigated and remain there. Even
people who lack the sense of smell would be warned by the
irritation of the eyes. Even at very low concentrations its
presence may be easily detected. Ethene oxide does not possess
this property.

Yet it remains essential for persons who enter the rooms
which are being fumigated to wear a gas mask. When the liquid
is being atomized, this mask should also be provided with a
mist filter.

I. The approximated c x t value of M for a fumigation period
of I, 2, 3, or 4 hours is about 160, in any case lower than 200.

2.nbsp;The c.t figure cannot be regarded as a measure for the practic-
al value of a gaseous insecticide.

3.nbsp;In actual practice the gram-hour value of M lies between
500 and 1000, dependent on the conditions under which it
is applied. If losses are high it may also be higher than 1000.

4.nbsp;M is considerably less poisonous to higher animals than
ethene oxide and only a little more poisonous than carbon
disulphide.

It was shown that the activity of this product is very high.
Furthermore it has an irritating effect on the insects, so that
It is not necessary to add other irritants.

If the gas is properly applied the quality of most products is
not affected. A method was worked out for the objective
determination of the effect on the taste of the products. The
germinating power of seeds is not affected; bulbs, tubers, and
many plants are not damaged.

The concentration to be applied — not to be confused with
dose — lies far below the explosion limit.

It is only a little more poisonous to higher animals than
carbon disulphide and considerably less so than ethene oxide.
Several practice trials are described.

It was shown that the so-called c.t value determined in the
laboratory is of no value for comparing the effect of gaseous
insecticides.

A graph is drawn by which to find the dose required in practice
for different fumigation periods and under different conditions.

The effect is expressed in gram-hours, that is, the number of
grams required multiplied by the number of hours. This number
of hours is the period of time that passes between the atomizing
of the gas and the moment at which ventilation is started.

The possible cases have been divided into three categories
in which a small, an average or a large dose is required, both
with respect to the products and the spaces. Products which
absorb a large quantity of the gas require a high dose. A large
dose is also required in rooms containing a lot of absorbent
material, such as old wood, carpets, curtains, or if leakage occurs.

Large dose — almonds, nuts, beeswax, furs, carpets, furniture.
Probably also tobacco and cattle-cakes.

The smallest dose is therefore required for pulse in vacuum
installations, for instance; the largest dose for furs in old ware-
houses with a lot of mouldered wood.

Some products, such as pulse and cocoa beans can stand a
long fumigation period very well. Other products such as
grain should not be fumigated for more than 12 hours and if
possible even for a shorter period of time.

Generally the duration of the fumigation should be shorter
as the number of gram-hours required is higher. It is advisable,
however, to use a dose not higher than 80 g per cub. m. The
parts of the lines in the graph which correspond to a higher
dose are dotted.

Empty rooms and houses may be fumigated for a long time
if this is desired. Up to now the gas has never been found to
affect metals, furniture or colours. Care should be taken that
objects which are likely to be damaged do not come into direct
contact with the mist blown off from the nozzle.

The number of gram-hours required for the fumigation times chosen
here, are indicated for each of the three categories in the graph page 96.

The extent to which the space is filled may also influence the
dose required. In general it may be said that with products which
absorb only a small quantity of gas and therefore only need a
small dose, this dose should be even a little lower if the space
contains a larger quantity of the product. In the case of products
which require an average dose, this dose remains the same even
if the rooms contain a larger quantity of the product. For
products which already require a large dose, because they
absorb a large quantity of the gas, the dose should be increased
if the space contains a large quantity of these products.

After fumigation the rooms should be well ventilated for
from 6—24 hours, the period being dependent on the conditions.

All Open fires and lights should be extinguished, also pilot
flames of gas heaters.

During fumigation the temperature should be at least 15° C,
preferably a litde higher.

Any person entering the rooms which are being fumigated
must wear a gas mask provided with a carbon filter, e.g. Degea
Canister A. During the time the liquid is being atomized, the
mask should be provided with a mist filter.

If necessary the rooms should be sealed before fumigation
for which purpose strips of wood, putty, strips of paper, etc.
may be used.

Suppose pulse must be fumigated in a gassing chamber. A
small dose is required. The duration of the fumigation is fixed
at 24 hours. According to the graph the number of gram-hours is
5 60. The dose per cub. m should then be 5 60 : 24 or, roughly, 23 g.

If the same products are to be fumigated for 10 hours, the
number of gram-hours required is 3 70 and the dose is 3 7 g per
cub. m.

Suppose bags of wheat must be gassed in a ware-house
constructed of reinforced concrete which can be properly
sealed. This is an average case. The best fumigation time for
wheat is less than 12 hours. The fumigation period may be fixed
at 8 hours. The number of gram-hours required is 520 read
from the graph. The dose should be 520 : 8 or 65 g per cub. m.

I will describe here the method which I succesfvdly applied
for obtaining Calandra species. With slight alterations this
method may be applied for many other insects.

The material needed for a laboratory culture must be obtained
from storages, in the case of Calandra, for instance, from gra-
naries, if possible from more than one. If this is not done it may
occur that a selected strain is obtained.

The collected Calandra are transmitted on to wheat of good
quality and placed in an incubator at a constant temperature
between 20—28 C.

The cultures may not be overcrowded. Overcrowding is
indicated by migration of the insects out of the wheat. To detect
this the cultures must be observed while they are standing in
a quiet place, because a tendency to migrate also occurs if the
insects are disturbed.

One month after starting the culture the adults are screened
of and destroyed. About one or two weeks later their progeny
emerge from the wheat. A sufficient number of these are
collected and a fresh culture is started with them. The natural
rate of mortality of these adults is also determined. If this
exceeds a few percents in one week, other material must be
looked for. Otherwise, the insects are kept for one month again
in the wheat; at the end of this period they are screened off
and destroyed.

The descendants that hatch after some weeks are now screened
off daily. The age of these adults is thus known exactiy.

-1200
lt;400

900
800
yoo
6oo
500
400
500
2 00

-loo
0

It is advisable to keep the cultures in an incubator at
27—28° C, while the adults that have been screened off are
placed in an incubator at 20° C, which is the same temperature
as that at which the experiments are carried out.
The most suitable relative humidity is about 80 %.
As containers for the cultures I used aquarium tanks with
a piece of flannel and a glass pane. The adults that had been
screened ofl^ were put into jars as described on p. 34; a piece of
cupper gauze was soldered into the cover of the jar.

Besides upon Calandra species, the effect of M was tested upon
many other insects. A list of the insects which, as was ascertained
in our experiments, may be controlled with methallyl chloride,
is given below. The varieties marked with * were used in practice
trials. The common names of the insects were obtained from
a list published by the American Association of Economic
Entomologists (38). The names between brackets did not occur
on this list.

*nbsp;Dermestes lardarius L. Larder beetle.
Attagenus piceus Ol. Black carpet beetle.
Lasioderma serricorne F. Cigarette beetle.

*nbsp;Sitodrepa panicea L. (Drug store beetle.)
Ptinus tectus Boield. (Spider beetle.)
Tenebrio molitor L. Yellow meal worm.
Gnathocerus cornutus F. (Broad horned flour beetle.)

Araecerus fasciculatus De G. Coffeebean weevil.
Bruchus rufimanus Boh. Broadbean weevil.

The following products were successfully fumigated both
in the laboratory and in actual practice (* only in the laboratory).

Wheat, maize, rice, peas, cocoa beans, tobacco*, apricot-
stones, almonds, hazel-nuts, dried apricots, various kinds of
raisins and currants, sugar, corn-flour, tapioca*, vermicelli*,
grated cocoa-nut, corn-flakes, ginger, pressed copra-cakes*,
furs, beeswax*, carpets and furniture.

Flour became musty. There is no objection whatever to
fumigation of flour-mills, as long as the stock of flour itself
is not fumigated.

1.nbsp;Rank, J.: Die Cacao-Motte. Bull. Off. de VOffice Int. d. Fahricants
de Chocolat et de Cacao. 1933.

2.nbsp;Schulze, K.: Einfluss von Temperatur und Luftfeuchtigkeit auf die
Eiablage und Entwicklungsdauer des Kornkäfers. Mitt. Gesell.
Vorratsschutz, 1937, No. 2 amp; 3.

3.nbsp;Cotton, R. I. and Good, N. E.: Annoted list of the insects and
mites associated with stored grain and cereal products and of their
arthropods parasites and predators. U.S. Depart, of Agric. Mise
Bull. 258, July 1937.

4.nbsp;Zacher, F.: Die Vorrats-, Speicher- und Materialschädlinge und
ihre Bekämpfung. 1927.

5.nbsp;WEroNER, H.: Bestimmungstabellen der Vorratsschädlinge und des
Hausungeziefers Mitteleuropas. 1937.

7.nbsp;Mueller, Karl: Beiträge zur Kenntnis des Kornkäfers. Zeitschr.
f. angew. Ent. XIII, 1928.

8.nbsp;Teichmann und Andres: Calandra granaria L. und Calandra
oryzae L. als Getreideschädlinge. Zeitschr. ƒ. angew. Ent. VI, 1920.

9.nbsp;Zacher, F.: Der Kornkäfer und seine Bekämpfung. Mitt. Gesell.
Vorratsschutz. 1933, No. 9.

11.nbsp;U.S. Department or Agriculture: Fumigation against grain-
weevils with various volatile organic compounds. Depart. Bull
No. 1313, 1925.

13.nbsp;De Bussy, L. P.: Lasioderma in Deli en zijn bestrijding. Mede-
deelingen v. h. Deliproefstation. 1916/1917.

14.nbsp;Hinds, W. E.: Carbondisulphide as an insecticide. U.S. Depart.
Agric. Farmers Buil. 799, 1917.

15.nbsp;Scherpe, R.: Bekämpfung von Calandra granaria L. mittels
flüchtige Giftstoffe. Mitt. Gesell. Vorratsschutz. 1928, No. 2.

16.nbsp;Lentz, O. und Gassner, L.: Schädlingsbekämpfung mit hoch-
giftigen Stoffen. //^/Z/Blausäure 1934 (Iste und 2. Folge 1935/1936),
Heft II Aethylenoxyd 1934 (Iste Folge 1935).

17.nbsp;Wille, J.: Chlorpikrin in der Schädlingsbekämpfung, insbesondere
im Kampf gegen den Kornkäfer. Z«i^schr. f. angew. Ent. VII, 1921.

19.nbsp;Flury, F.: Über Phosphorwasserstoff. Anz. Schädlingskunde.
1937, No. 3.

20.nbsp;Kleine: Prüfung neuer chemischer Mittel zur Bekämpfung des
Kornkäfers. Zeitschr. angew. Eni. XIII, 1927.

21.nbsp;Cotton, R. T. and Roark, R. C.: Ethylene oxide as a fumigant.
Ind. and Eng. Chem. 1928, No. 8.

22.nbsp;Hazelhoff, E. H.: Regeling der ademhaling bij insecten en spinnen.
Diss. 1926.

23.nbsp;Mackie, D. B.: Methylbromide — its expectancy as a fumigant.
Fisk, F. W. and Shepard, H.: Laboratory studies of methylbromide
as an insectfumigant. Journ. Econ. Ent. 1938, No. 1.

25.nbsp;Briejer, C. j. Control of insects by methallylchloride. Nature,
June 1938, p. 1099.

26.nbsp;Briejer, C. J.: Neue Gase zur Bekämpfung von Vorratsschädlingen
und die Feststellung ihres praktischen Wertes. Paper read at the
Ith Int. Congress of Entomology. Berlin, 1938.

27.nbsp;Strand, A.: Measuring the toxicity of insect fumigants. Ind. and
Eng. Chem. {An. Ed.) 1930, No. 1.

29.nbsp;Shepard, H. H., Lindgren, D. L. and Thomas, E. L.: The relative
toxicity of insect fumigants, 1936.

31.nbsp;Kunike, G.: Beiträge zur Lebensv^^eise und Bekämpfung des Korn-
käfers. Calandra granaria L. Zeitschr. f. angew. Ent. XXIII, 1936.

32.nbsp;Page, A. B. P. and Lubatti, O. F.: Determination of Fumigants.
Journ. Soc. Chem. Ind. 56, 1937.

33.nbsp;Lepigre, a.: Contribution à l'étude de la désinsectisation des
grains par le mélange d'oxyde d'éthylène et d'acide carbonique.
Notes sur le bromure de méthyle. 1936.

34.nbsp;Peters, G.: Zur Thema „Neue Schädlingsbekämpfungsmittelquot;.
Chemiker Zeitung. 14. Jan. 1939.

35.nbsp;Peters, G.: The biological and chemical tests of efficiency of
gaseous insecticides. Paper read at the 1th Int. Congress of Entomology.
Berlin, 1938. Anz. f Schädlingskunde, 1938, Heft 10.

36.nbsp;Peters, G. und Ganter W.: Zur Frage der Abtötung des Korn-
käfers mit HCN. Z^itschr.f. angew. Ent. XXI, 1935.

Het experimenteel in het laboratorium bepaalde Habergetal
heeft voor de onderlinge vergelijking van gasinsecticiden
geen waarde.

2. Bij de bestrijding van insecten met chemische middelen
bestaat het gevaar, dat door overleven van de sterkste in-
dividuen een averechtsch effect wordt bereikt.

3. Chemische middelen kunnen bij de bestrijding van schade-
lijke insecten niet worden gemist.

4. Bij het onderzoek van den invloed van chemische middelen
op zaaizaden dient behalve de kiemkracht ook de verdere
ontwikkeling te worden nagegaan.

5. De meening van Flury, dat de oorlog met gassen „humanerquot;
is, dan de oorlog met explosieve stoffen, is onjuist.

6. Bij de bemesting van land- en tuinbouwgewassen wordt te
weinig aandacht besteed aan de mogelijkheid tot natuurlijke
ontwikkeling van koolzuur.

7. Kimstmatige koolzuurbemesting van gewassen op het vrije
veld is uitvoerbaar en heeft groote praktische waarde.

8. Er dienen proeftuinen aangelegd te worden, waarin de
genotypen van land- en tuinbouwgewassen, die bij de
selecde terzijde gesteld worden, bewaard blijven.

9. Het beïnvloeden van de kiemcellen van land- en tuinbouw-
gewassen door Röntgenbestraling e.d., met het doel kunst-
matig een groot aantal nieuwe varieteiten te doen ontstaan,
is economisch niet te verdedigen.

10. Het geelziek der hyacinthen kan bestreden worden door
verandering van cultuurmethode.

ii. De mededeeling van Kabos, dat de narcisvlieg Merodon
equestris Fabr. door een warmwaterbehandeling bij 43.5° C.
lang niet altijd gedood wordt, is onjuist.

Voor bestrijding van de narcisvlieg is het noodzakelijk, dat
verwilderde narcissen in bosschen, parken en tuinen, ge-
legen in de bloembollenstreek, vernietigd worden.

13- De fabricage van suiker uit beetwortelen kan vereenvoudigd
worden, door de uit de wortelen verkregen snijdsels te
drogen en op te slaan en de rest van de bewerking over het
geheele jaar te verdeden.

14. Het construeeren van curven uit de resultaten van biolo-
gische proefnemingen blijve beperkt tot het allernood-
zakelijkste.

15. Geregelde kunstmatige zuivering van de lucht in labora-
toria, fabrieken en werkplaatsen zal de arbeidsprestaties
belangrijk verhoogen en het aantal ziektedagen vermin-
deren; zij dient van overheidswege te worden voorge-
schreven.

16. Bij het onderwijs aan de Nederlandsche Universiteiten
wordt onvoldoende aandacht besteed aan de toegepaste
biologie.

Date	Adults screened off		Date	Adults screened off
Date	non-fumig.	fumigated	Date	non-fumig.	fumigated
July 15	81	4	July 27	7	3
„ 16	31	I	„ 28	7	I
,, 19	31	I	» 29	3	2
„ 20	13	0	» 30	3	3
» 21	13	5	Aug. 2	3	2
» 22	14	9	» 3	I	I
„ 23	8	5	» 4	2	I
„ 26	II	10	» 5	2	I
			» 6	0	0

Date	Adults screened off		Date		Adults screened off
Date	non-fum.	fumigated	Date		non-fum.	fumigated
July I	2	0	Aug. 4		50	0
» 2	0	0		5	49	0
» 5	I	0	99	6	49	0
» 6	3	0	»	9	100	0
» 7	8	0	9gt;	10	138	0
» 8	26	0	99	II	148	2
» 9	42	0	99	12	114	0
„ 12	98	0	99	13	95	0
» 13	202	0	99	16	143	0
» 14	139	0	99	17	178	0
» 15	141	I	99	18	146	0
„ 16	124	I	99	19	71	0
» 19	164	0	99	20	49	0
„ 20	149	0	99	23	38	0
jgt; 21	170	0	99	24	16	I
„ 22	quot;7	0	99	25	12	I
» 23	97	I	99	26	II	I
„ z6	143	2	99	27	6	2
» 27	148	0	99	30	4	5
„ 28	97	0	Sept. I		II	4
» 29	80	0	99	2	5	2
» 30	67	I	99	3	4	0
Aug. 2	94	0	99	6	0	0
» 3	ICQ	0

Date	Adults screened off serie 775 gr.		Adults screened off serie 900 gr.
Date
	non-fum.	fumigated	non fum.	fumigated
Aug. 4	112	0	198	0
» 5	67	0	99	0
» 6	75	0	132	0
» 9	179	0	190	0
» 10	159	0	199	0
» II	103	0	97	0
» 12	136	0	137	0
» 13	81	0	133	0
„ 16	130	0	213	0
» 17	143	0	189	0
„ 18	126	0	185	0
» 19	109	0	148	0
» 20	117	0	139	0
» 23	200	0	250	0
„ 24	99	0	229	I (C. gran.)
» 25	122	0	236	0
„ 26	126	0	242	0
» 27	141	0	182	0
» 30	238	0	437	0
Sept. I	194	I (C. gran.)	280	0
» 2	117	0	247	0
„ 3	83	0	149	0
„ 6	97	0	222	0
» 7	72	0	170	0
„ 8	91	0	124	0
.. 9	44	0	82	0
„ 10	34	0	60	0
13	52	0	125	0
» 14	27	0	41	0
» 15	17	0	27	0

Date	Adults screened ofif		Date	Adults screened off
Date	non-fum.	fumigated	Date	non-fum.	fumigated
July 21	13	0	Aug. 12	50	0
„	5	0	» 13	37	0
» 23	10	0	» 16	44	I
„ 26	13	0	» 17	22	0
» 27	26	0	„ 18	38	0
„ 28	22	0	» 19	27	0
„ 29	29	2	„ 20	23	0
» 30	22	I	» 23	54	0
Aug. 2	31	0	» 24	22	0
» 3	42	0	» 25	42	I
» 4	34	I	» 26	33	0
» 5	29	2	,, 27	24	0
„ 6	47	0	» 30	32	0
» 9	57	0	Sept. I	36	0
» 10	53	0	» 2	31	0
„ 11	40	0


	\. dosed here / \ - /
	ioo\/
1 1
// 1 1 / --- ' ^ --- 1 1	looV^
------ 1 / \ y / ✓	XlOO,^^
/ /■ / ^—		/
/ ^^^^ ^^ —--	10^	/

Counted		Dose in grams per cub.				m.		Temperature
Counted	6.Z5	9-37	12.50	15.60	18.75		21.87	Temperature
Dir.	___	0%	_	5 %		17%	73 %	ca. 19° C
24 h.	—	15	—	81		81	ICQ
I w.	—	69	—	100		100	100

Counted	Dose in grams per cub. m.						Temperature
Counted	6.25	9-37	12.50	15.60	18.75	21.87	Temperature
Dir.		0		_	18	—	ca. 18° C
24 h.	—	II	—	—	90	—
I w.	—	69	■—■	—	100	—-
Dir.	—.	.—.	2	26	—	—■	ca. 17° C
24 h.	—	—	10	55	■—'	.—
I w.	—	—■	55	88	—	—
24 h.	—■	6	II	77	82	96	ca. 17° C
1 w.	—	54	73	100	100	100
24 h.	■—■	12	19	73	89	100	ca. 19° C
I w.	—	37	59	96	100	100
24 h.	—.	4	6	81	76	98	ca. 17° C
I w.	—	60	65	98	100	100

Dir.	18	32	—	87	92	-	ca.	19° C
24 h.	68	64	—	100	IDG	-
I w.	95	98	—	IDC	100	-
Dir.	—	52	—	-	I GO	—	ca.	18° C
24 h.	—	100	—■	-	100	■--
I w.	—	100	—	-	I GO	-
Dir.	—	.—•	80	ICO	—	-	ca.	17° c
24 h.	—	—	100	100	—	-
I w.	—	—	ICO	100	-	-
24 h.	18	58	71	100	I GO	-	ca.	17° c
I w.	64	92	99	IOC	IGG	-
24 h.	30	84	94	100	100	-	ca.	19° C
I w.	52	98	98	100	IDG	-
24 h.	52	—■	92	98	92	-	ca.	17° c
I w.	91	—■	IDG	100	IGG	-

Counted		Dose in	grams per	cub. m.		Temperature
Counted	12.6	13.0	14.5	17.4	21.75	Temperature
Dir.	98 %	69%	80%	too %	100 %
24 hours	100	—	100	100	100
I week	100	ICQ	100	100	IOC

On the bottom of the box		Halfway up the box		Against the roof of the box
after 24 hours	I -week	after 24 hours	I week	after 24 hours	I week
14	84	12	98	16	98
20	88	16	94	28	90
9	90	26	100	28	94
36	98	18	98	24	100
10 (middle)	90	22 (middle)	97

36	100	36	96	36	94
34	94	38	94	44	100
28	98	35	91	28	98
30	86	38	93	26	97
40 (middle)	100	60 (middle)	98

Bag. No.	»/„ killed	Bag. No.	»/„ killed	Bag. No.	»/„ kiUed
I	91	6		II	96
2	84	7	84	12	96
3	88	8	80	13	84
4	72	9	80	14	96
J	84	10	20	15	93

Bag.		Top		Bag.	Middle			Bag.		Bottom
No.	percent killed			No.	percent killed			No.	percent killed
	dir.	72 h.	I W.		dir.	72 h.	I w.		dir.	72 h.	I w.
I	8	80	98	6	15	76	93	II	4	74	96
2	14	84	100	7	4	76	96	12	14	67	94
3	30	82	100	8	2	80	96	13	7	67	93
4	46	84	98	9	4	73	98	14	14	78	98
5		98	100	10	4	70	92	15	I	62	88

kg of wheat	M per kg in nil	M per jar in ml	Mon 24 hours	tality I week
2	0.025	0.050	34	73
2	0.025	0.050	22	73
2	0.030	0.060	60	98
2	0.030	0.060	42	92
2	0.040	0.080	72	100
2	0.040	0.080	86	100
2	0.050	O.IOO	94	100
2	0.075	0.150	100	100

kg of wheat	M per kg in ml	M per jar in ml	Mort 24 hours	ality I week
Z	O.IOO	0.200	100	100
1-75	0.025	0.044	16	65
1-75	0.030	0.053	36	70
1-75	0.030	0.053	49	92
1-75	0.040	0.070	96	100
1-75	0.040	0.070	73	100
1-75	0.050	0.088	92	100
1-75	0.075	0.131	100	100
1-75	O.IOO	0.175	100	100
1.50	0.030	0.045	40	83
1.50	0.030	0.045	51	100
1.50	0.040	0.060	94	100
1.50	0.040	0.060	96	100
1.50	0.050	0.075	94	100
1.50	0.075	O.I 13	100	100
1.50	O.IOO	0.150	100	100
1.25	0.050	0.063	69	90
1.25	0.050	0.063	100	100
1.25	0.075	0.094	96	100
1.25	O.IOO	0.125	100	100
1.00	0.050	0.050	62	94
1.00	0.050	0.050	60	100
1.00	0.060	0.060	100	100
1.00	0.075	0.075	98	100
1.00	0.075	0.075	100	100
1.00	O.IOO	O.IOO	100	100
0.75	0.040	0.030	25	40
0.75	0.050	0.038	30	49
0.75	0.050	0.038	27	71
0.75	0.050	0.038	43	80
0.75	0.060	0.045	36	82
0.75	0.075	0.056	98	100
0.75	O.IOO	0.075	100	100
0.50	0.050	0.025	24	52
0.50	0.075	0.038	63	94
O.JO	0.075	0.038	12	61
O.JO	0.085	0.043	62	97
O.JO	O.IOO	0.050	100	100


.-■fv. gt; • -
ic- -

Date	Blank	Areginal	M	Date		Blank	Areginal	M
Jan. 3	54	0	0	Jan.	24	63	0	0
» 4	40	0	0		25	67	I	0
» 5	15	0	0	gt;5	26	39	I	0
» 6	23	0	0	»	27	34	0	0
» 7 » 10	21 66	I 7	0 0	jj	28 31	69 64	5 2	0 0
» II	75	10	0	Febr. 2		99	7	0
» 12	39	9	0	gt;3	3	30	I	0
» 13	40	8	0	gt;5	4	22	0	0
» 14	35	2	0		7	36	0	0
» 17	70	6	0		8	9	0	0
» 18	36	3	0	3gt;	9	7	0	0
» 19	30	3	0	5gt;	10	8	2	0
» 20	12	0	0	JJ	II	7	I	0
» 21	14	4	0	5gt;	14	12	0	0

Fumigation time	Dose per cub. m in ml	24 hours	I week
3 hours......	..... 69.6	48	98
Duplicate.....		32	84
Triplicate.....		40	88
3 hours......	..... 69.6	46	ICQ
Duplicate.....		20	IOC
Triplicate.....		20	98

Fumigation time	Dose per cub. m in ml	24 hours	I week
		36	98
		46	100
Triplicate.......		52	98
6 hours........		20	98
Duplicate.......		32	96
Triplicate.......		20	98
Average 3 hours.....		34	95
Average 6 hours.....		34	98

ml M per kg	ml M per jar	Mort 24 hours	rality I week
0.3	0.6	100	100
0.3	0.6	ICO	100
0.3	0.6	100	ICG
0.2	0.4	80	ICO
0.2	0.4	94	100
0.2	0.4	90	100
0.15	0.3	33	92
0.15	0.3	32	95
O.IO	0.2	8	86
O.IO	0.2	4	88
Second series
0.3	0.6	100	100
0.3	0.6	100	100
0.25	0.5	87	100
0.25	0.5	98	100
0.20	0.4	82	100
0.20	0.4	79	100
0.15	0.3	43	98
0.15	0.3	65	100
O.IO	0.2	33	83
O.IO	0.2	13	48
0.05	O.I	4	19
bianco	bianco	0	12
bianco	bianco	8	II

kg of wheat per jar	M per kg in ml	M per jar in ml	Mort 24 hours	ality I week
2	0.16	0.32	86	ICO
1-75	0.17	0.30	96	100
1-75	0.17	0.30	90	ICO
1.50	0.18	0.27	93	100
1.50	0.18	0.27	80	IOC
1.25	0.20	0.25	90	100
1.25	0.20	0.25	72	100
1.00	0.24	0.24	86	100
1.00	0.24	0.24	76	ICO
0.75	0.28	0.21	54	IOC
0.75	0.28	0.21	50	98
0.50	0.36	0.18	27	97
0.50	0.36	0.18	37	99

Tin	Capacity in litres	Dose p. tin	Dose per cub. m in ml
I	10	0.50	50
2	5	0.25	50
3	3-3	0.17	51
4	2-5	0.13	52
5	2	O.IO	50
6	1-7	0.08	48
7	1.4	0.07	49

Capacity in Ktres	Dose	Pet cent killed in
Capacity in Ktres	Dose	24 h.	2 days	I w.
10	0.40	47	79	100
5	0.20	50	82	100
3-3	0.13	32	64	98
2.5	O.IO	36	80	97
2	0.08	40	78	100
1-7	0.07	42	80	98
1.4	0.06	18	58	96
This confirms that in the case

Capacity in litres	Dose	Per cent killed n
Capacity in litres	Dose	24 h.	2 days	I w.
10	0.40	32	78	100
5	0.20	37	87	100
5-3	0.13	20	78	ICQ
2-5	O.IO	26	80	96
2	0.08	40	84	98
1-7	0.07	29	73	97
1.4	0.06	32	89	ICQ
of wheat fumigatio			n the	filling

Capacity in litres	Dose per tin	Killed in
Capacity in litres	Dose per tin	24 hours	2 days	I week
5	0.50	78	100	100
3-3	0.33	18	78	100
2-5	0.25	20	56	98

Capacity tin	FiUing per cub. m in kg	Per cent killed
5	100	100
3-3	150	78
2-5	200	56

Capacity in litres	Fining per cub. m in kg	Dose per tin in ml	24 hours	2 days	I week
10	50	I	93	100	100
5	100	0.50	68	100	100
3-3	150	0.33	10	81	98
2-5	200	0.25	29	79	100
2	250	0.20	0	26	74
1-7	300	0.17	0	14	80
1.4	350	0.14	0	8	78
10	50	I.OO	98	100	100
5	100	0.50	68	100	100
3-3	150	0.33	26	86	100
2-5	200	0.25	30	78	100
2	250	0.20	4	48	92
1-7	300	0.17	4	20	84
1.4	350	0.14	0	16	74

Cap. tin in litres	Filling per cub. m in kg	Dose	24 hours	2 days
10	100	0.30	36	76
5	200	0.15	24	58
3-3	300	O.IO	22	66
2-5	400	0.08	48	82
2	500	0.06	46	84
1-7	600	0.05	44	84
1.4	700	0.04	68	94
10	100	0.30	6	42
5	ZOO	0.15	10	54
3-3	300	O.IO	24	66
2-5	400	0.08	34	78
2	500	0.06	46	86
1-7	600	0.05	32	82
1.4	700	0.04	58	94

Cap. tin in litres	Filling per cub. m. in kg	Dose	24 hours	2 days	I week
10	50	0.35	8	64	98
5	100	0.175	8	66	80
3-3	150	0.12	6	58	92
2-5	200	0.09	II	94	100
2	250	0.07	6	42	86
1-7	300	0.06	6	34	70
1.4	350	0.05	2	14	66
10	50	0.40	50	98	100
5	100	0.20	58	90	100
3-3	150	0.135	55	84	99
2-5	ZOO	O.IO	44	84	100
2	250	0.08	38	76	98
1-7	300	0.07	48	90	100
1.4	350	0.06	28	74	96

Time in hours	Dose in ml	Time in hours	Dose in ml
24	50	3	100
6	200	3	69
6	100
6	50
6	25

Number of seeds	Percentage Germinated
Number of seeds	Not fumigated	fumigated
100	97	90
ICO	75	72
ICO	91	93
100	25	28
100	75	74
100	79	82
100	96	97
100	86	87
100	95	88
100	35	32
100	65	65
100	96	92
50	96	96
100	92	91
50	92	96
50	90	92
50	86	92
50	100	100

3rd floor	I		2	sorting loft
2nd floor	5		4	3
I St floor	passage	room in which beans are crushed	landing	paper loft

Fumigation time	Dose per box in ml.	Per cent killed in
Fumigation time	Dose per box in ml.	24 hours	I week
3 hours	6	16	78
3 »	6.5	14	84
3 »	7.0	42	98
6 „	3-5	22	100
6 „	4.0	46	100
6 „	4-5	40	100
24 „	1-5	66	96
24 „	1.6	60	100
24 „	1-7	80	100
Blank I		0	0
„ n		4	4

71
Fumigation period	Dose per kg in ml	Dose per cub. m in ml	Dose per cub. m in gram	gram-hours
24	0.045	28.8	26.64	639
6	0.095	6i.o	56.43	338
3	0.160	103.0	95.28	286

Time	Kill in per cent												Average kill
2 hours 40 min.	94	98	82	94	90	96	98	94	94	90	96	90	. 94-7
2 „ 5° ..	98	98	88	96	92	100	98	98	96	96	100	96	96.3
3 .. — ..	96	100	94	96	92	94	100	98	94	98	98	98	96.5
3 .. 10 ..	100	100	100	100	88	94	100	100	96	98	100	100	98.0
3 „ 20 „	100	100	100	100	100	100	100	100	94	100	100	100	99-5

Time	KiU in per cent																				Aver- age
2 h. 40 min.	95	94	94	95	94	90	98	90	98	96	98	98	84	94	100	100	100	98	98	100	95-7
2 „ 50 „	98	100	94	96	96	100	94	92	100	98	100	100	92	100	100	100	96	98	100	100	97.7
3 .. — „	98	100	100	98	98	98	100	100	98	96	100	100	94	96	100	100	100	100	98	100	98.7
3 10 „	100	100	100	100	100	100	100	100	96	96	100	100	100	98	100	100	98	100	100	100	99.4
3 20	100	100	100	100	100	100	100	100	100	100	100	100	100	100	100	100	109	100	100	100	100

Time	Kill in per cent															Average
0 h. 50 min.	98	98	98	94	98	98	96	98	90	88	98	94	88	100	92	95.2
0 „ 55 ..	100	96	100	100	100	100	98	98	86	98	98 96	98	94	94	96	97.1
I „ — .gt;	100	100	98	100	100	ICO	100	100	96	100	98 96	98	98	92	100	98.5
I ,, 5 ..	98	100	100	100	100	100	100	100	96	98	98	98	96	94	98	98.4
I „ 10 quot;	100	100	100	100	100	100	100	100	96	100	100	100	100	100	100	99-7

Time								K	iU in	per cent						Average
3	hours	20	min.	92	96	88 96	92	94	96	98	98	100	94	98	98	95-3
3	„	40	JS	96	98	88 96	100	100	96	100	98	100	98	100	94	98.2
4		—	„	98	100	100	100	100	100	100	100	100	100	100	100	99.8
4		20	„	100	100	100	100	100	100	100	100	100	100	100	100	100
4	»	40	»	100	100	100	100	100	100	100	100	100	100	100	100	lOO

Fumigation period	Ethene oxide	Methallyl chloride	Catbon disulphide
I ^ minutes	13 —	_
	26 —2	—	___
3	14—26	—	—
15 minutes		51 —	6 —
17% »		16 — J	8 —
20		24 —15	17 —2
25		2—4	9—9