12. Glasshouse fumigation

Contents - Previous - Next

For many years, fumigants were used for the routine control of insect pests in glasshouses. Calcium cyanide dust, generating hydrogen cyanide (HCN) with the aid of the moisture of the air, and nicotine, volatilized as a gas by ignition or other means, were both commonly employed.

Recently, the aerosol method of dispersing insecticides and evaporation from steam pipes have found wide favour for glasshouse work, largely because considerable volumes of glasshouse space can be treated quickly and economically.

Although it was previously stated that the subject of aerosols lay outside the scope of this manual, it is impossible to discuss glasshouse treatments without taking into account the various ways in which the insecticides are dispersed. Therefore, a brief outline of the use of fogs, liquified gas aerosols, smokes and actual fumigants will be presented with the intention only of indicating some of the general methods employed to control insects and related pests on plants in glasshouses.


In computing the dosage of insecticide to be applied, the cubic capacity of the glasshouse must be known. In calculating this, any space occupied by plants, benches, soil and other material should be ignored.

If the house consists of one span (lean-to-type), or is divided at the pitch of the roof into two even spans, the area of the cross-section may be calculated as the average height multiplied by the width. Thus, capacity of house = average height x width x length.

When the roof is of unequal span, it may be considered as consisting of two lean-to sections and the volume is calculated as the sum of these two sections.


When aerosols , fogs, smokes, steam pipe fumigants and gases are applied, it is necessary to close all vents. Temperatures should be maintained above 20C but kept below 35C. For this and other reasons, application is usually made in the evening. It is not practicable to carry out treatments when the wind velocity is above 16 kilometers (10 miles) per hour because when the wind outside is too strong, there is uneven distribution of the insecticide in the glasshouse, resulting in underdosing in some places and overdosing in others to the point of plant injury.


The time of exposure for aerosols, fogs and smokes is usually two to three hours, after which all the vents are opened. By this time most of the material has dispersed or settled. However, when fumigation takes place with most of the available insecticides and acaricides (chemicals which kill mites), and with HCN generated from calcium cyanide, it is usual to leave the house under gas until early the following morning.


Liquified Gas Aerosols

In connexion with glasshouse work, the term "aerosol" is used to describe the form in which the insecticide is dispersed after it leaves a special cylinder (sometimes called bomb) in which it is dissolved in a volatile liquid held under pressure. On discharge into the open the liquid vaporizes and leaves the particles of insecticide suspended as a fine mist in the air. This form of application is distinguished from the generation of fogs and smokes.

A wide range of common insecticides and acaricides have been used for dispensing from small aerosol cylinders carried in the hand. These include such materials as lindane, malathion, dichlorvos (DDVP, Vapona) and DDT, used principally as insecticides.

Not all insecticides can be effectively dispensed from liquified gases, because there are sometimes technical difficulties in finding suitable solvents and in producing formulations that do not cause clogging of the nozzles at the time of discharge. As a general rule, the toxicants with higher vapour pressures are more effective; this may be partly due to A fumigation effect (see Introduction). At present, too little is known about the mode of action of aerosols on insects and mites.

Before discharging the aerosol, many growers place the cylinder or bomb in warm water at,but not above, 38C for half an hour or so. This increases the pressure in the container and thus gives better distribution when the aerosol is released.

Low cost. The aerosols are popular at present among growers because a glasshouse of 1 400 m (50 000 ft ) volume may be treated in about three minutes at a very low cost.

Pests controlled. Aerosols are used to control many of the pests that attack glasshouse crops. However, aerosols cannot penetrate soil and they are not effective against slugs and snails. Scale insects are controlled to some extent but, for satisfactory results, treatments have to be repeated at regular intervals.

Precautions. Many of the materials used in aerosols are extremely toxic to human beings, including the carrier gases such as methyl chloride. It is necessary, therefore, that persons applying aerosols should wear a respirator and protective gloves and clothing. The respirator should be of the industrial type described in Chapter 3 and the canister recommended is for "organic vapours, acid gases, smoke and dust."

The protective clothing consists of a work suit completely covering the arms and body and long rubber gloves. After application is completed the protective clothing and gloves are removed and all skin areas exposed to the aerosol are thoroughly washed with soap and water. If operations are extensive, the working shoes should also be removed.

Warning. On no account should smoking be allowed in the vicinity while aerosols are being discharged or at any time during treatment until such time as the glasshouse has been fully aerated. Aerosols may be sources of fire or explosion; more particularly, methyl chloride is a highly flammable gas.


Insecticidal and acaricidal smokes are particularly useful in smaller glasshouses. The toxicants are mixed with a combustible material in a special container, which allows ready ignition and discharge. These containers are usually tins or waxed containers and are popularly called "pressure fumigators" or "smokes". The range of materials available in this form is more restricted than in the aerosols. Azobenzene, nicotine, lindane, dichlorvos and others have been used.

Generally speaking, the smokes cause less plant injury than the aerosols and, therefore, the grower has to pay less attention to choice of material and to other technical considerations.

Warning. Smokes diffuse readily into offices and living quarters attached to glasshouses. Care must be taken either to prevent diffusion or to evacuate such spaces while treatment is in progress.


The production of "fogs" is, in effect, a modification of a spraying technique. By means of compressed air, spray concentrate solutions are forced through atomizers so that small droplets are formed. This method has been used for dispersing a number of insecticides and acaricides by means of portable fog-generating machines (Pritchard, 1949).

Gaseous Fumigation

True vapours or gases are still used in certain ways for glasshouse pest control.

Steam pipe fumigation. A method of some value is to paint a slurry of some toxicant on the steam pipes. The gas is usually volatilized during the night. The insecticide lindane has been used in this way and the vapours of naled (Dibrom) and dichlorvos have also been successfully used for the control of a large number of pests. In commercial application of both naled and dichlorvos, protective clothing and suitable respirators are required.

It is often necessary to repeat this type of treatment regularly once a week to obtain adequate control. (It should he mentioned that sulphur for the control of powdery mildew is often volatilized in this way or by using thermostatically controlled heating units.)

Calcium cyanide. HCN, generated from calcium cyanide, once very generally used, is still employed occasionally to solve certain problems or to serve as a variation of the aerosol technique. The form of calcium cyanide best suited for glasshouse work has the consistency of sea sand. Although moisture in the air is necessary for the generation of HCN from calcium cyanide, excessive moisture in the glasshouse cuts down the efficiency of the treatment by dissolving the gas; this may also lead to considerable plant injury. It is important that there should be no standing water on the walks where the dust is applied because a complex reaction takes place which reduces the amount of gaseous HCN given off. Therefore, it is the practice not to water the plants for several hours before fumigation.

Wearing a respirator with the special canister for HCN, the operator starts from the far end of the house and works toward the final exit door at the opposite end. He carefully pours a thin stream of cyanide from the tin container onto the concrete floor of the house taking care always to move away from the dust and never to cross back over a place already treated.

The dosage of calcium cyanide varies according to the pests to be controlled and the susceptibility of the varieties of plants present. Many plants will not tolerate more than 0.5 to 0.75 g/m (0.5 to 0.75 oz/ 1 000 ft). Hough and Mason (1951) have published full lists both of the glasshouse pests which can be controlled with certain dosages and also of the plants which will tolerate these treatments.

Calcium cyanide is particularly useful for a clean-up campaign in empty glasshouses. Under these conditions, dosages of 16 to 32 g/m (16 to 32 oz/l 000 ft) of the powder are employed for exposures of not less than 24 hours. The maximum dose is required to eradicate red spiders. As no plant injury is involved in this treatment, a considerable amount of sealing can be done.

Calcium cyanide, which reacts with moisture to give off HCN, must be handled with great care (See Chapter 6).

Methyl bromide Chamber

Methyl bromide is not usually a suitable fumigant for general release in a glasshouse, mainly because it diffuses so rapidly through small leakage holes that it is difficult to maintain the concentrations needed to kill the pests. However, for glasshouses that can be made sufficiently gas tight, it has been used effectively for disinfestation purposes. A chamber for atmospheric fumigation with methyl bromide is of considerable help for dealing with localized infestations on a few plants or for treating stock coming in and going out. Establishments that ship considerable amounts of material under quarantine often operate their own chambers under the supervision of quarantine officials.

A small chamber of a type described in Chapter 8 with an exhaust vent leading outdoors, can be installed in a well-ventilated part of the premises. Usually such a chamber need not exceed about 6 m (200 ft) in capacity.

A handbook giving the tolerances of many plants to methyl bromide has been produced by the U.S. Department of Agriculture (USDA, 1977).


The development of light weight plastic sheets has made it possible to cover sizable areas of land so that volatile chemicals can be contained for sufficient time to effect a treatment either of the soil or of vegetation above the ground. The use of this technique for the control of weeds and of soil-infesting insects and nematodes, is well known. This method has also been used successfully to deal with infestations above ground level.

In California, large beds of strawberry plants infested with cyclamen mites are covered with white polyethylene sheets (tarpaulins), normally 30 x 60 m (100 x 200 ft) in size (Allen, 1957). The tarpaulins are sealed down at the edges with earth or placed under a narrow moat of water. Methyl bromide from cylinders is discharged as "hot gas" under the sheets through plastic hose used for irrigation (soil soakers), so as to give an even distribution of fumigant over the area. Recirculating blowers are also used.

When transparent plastic sheeting is used in full daylight, there is considerable overheating of the plants. This is avoided for the most part by using opaque material. Nevertheless, temperatures under the sheets must be carefully observed by using thermocouples or thermistor thermometers. Exposure is for two to three hours, according to seasonal conditions, with the dosages varying from 24 to 48 g/m (1.5 to 3 lb/ 1 000 ft ) as the temperatures decrease under the sheets from 30 to 10C.

In summer, fumigation should be done between crops. Some differences in tolerance to the fumigation are shown by varieties of strawberries, and injury may often be avoided by treating the plants when they are dormant. In the summer, fumigation is best done early in the morning or during the evening. There is evidence that this treatment has a stimulating effect on plant growth. However, it is not recommended that fumigation be done solely for this effect when the mites are not present, because stimulation may sometimes result in small leaves, flowers and fruit (Allen, 1957).

Plastic-coated nylon fumigation sheets were used in a campaign to eradicate overwintering larvae of the oriental fruit moth in the okanagan Valley of British Columbia (Monro, 1958b, c). The insects were lodged on or in the soil and debris in an orchard into which had been dumped some infested waste from a canning factory.

Measures to control the European pine shoot moth by field fumigation with methyl bromide of ornamental pines in commercial nurseries and private properties are described and illustrated in detail by Carolin et al (1962), Klein and Thompson (1962) and Carolin and Coulter (1963).

13. Plant quarantine treatments

In plant quarantine work, the object of fumigation is to obtain complete mortality of 81 l stages of the pest against which the treatment is directed. For each prescribed treatment, experimental work has defined certain conditions required to bring about this degree of control. A minimum certification statement of conditions should always contain the dosage, exposure time, and temperature. More exact conditions may be added, such as the minimum concentration x time (c x t) product of fumigant required, or a load factor may be introduced to modify dosage according to the amount of commodity present. Sometimes other factors are stipulated, such as the maintenance of a given humidity in the system during treatment or the amount of vacuum, when applicable.

To be effective, the stipulated conditions must be attained in practice every time the treatment is made.

It may be helpful to review here the more important technical considerations involved in plant quarantine fumigation.

Volatilization of fumigant. A particular treatment calls for a certain dosage or concentration of fumigant for a certain length of time. To attain the desired result, the full concentration must be present from the beginning of the treatment. The beginning of the exposure period can be counted only from the time the fumigant is fully volatilized. Furthermore, if the fumigant evaporates slowly after it is introduced into the system, progressive sorption of the fumigant by the commodity may prevent the attainment of the concentration x time product needed to kill the pest organism in the time allotted.

This consideration is particularly important in treatments of perishable products, such as flowers, growing plants, fruit and vegetables, where the exposures are comparatively short periods of two to four hours. Even when volatile fumigants, such as methyl bromide, are used in treatments at temperatures below 20C, the fumigants evaporate slowly if they are not artificially heated before or after they enter the chamber. If at all possible, methyl bromide should be introduced as a "hot gas", in the ways already described. If that is not convenient, it should be discharged onto a heated evaporating pan inside the chamber. Ethylene dibromide may be evaporated rapidly in pans placed on hot plates or by means of efficient vaporizing nozzles.

Methyl bromide is recommended for use in certain treatments down to 4C. Such treatments are likely to fail if the fumigant is not fully volatilized when it is introduced.

Leakage. Leakage from the structure in which treatment is being done may render the treatment ineffective. It is necessary to undertake regular checks of the gas tightness of all chambers, railway cars, trucks or other spaces used for quarantine fumigations. Methods of testing structures for gas tightness are described in Chapters 8 and Il.

Chemical analysis. In practice, the attainment of effective concentration x time products is the only way of ensuring complete control. Therefore, during actual treatments of the quarantined commodities, determinations should be made at regular intervals of time of gas concentrations in different parts of the system to ensure that the necessary concentration x time products are attained.

Test insects. Usually, it is inconvenient to make chemical analyses during every treatment. These analyses may be supplemented by the regular use of test insects. The response of the test insects should be approximately the same as that of the most resistant stage of any pest likely to be encountered. If the reactions of the test insects are related to the results of periodic analyses of the fumigant concentrations, a constant check can be made on the effectiveness of the treatment.

Loading. In the fumigation of produce, such as fresh fruit and vegetables shipped in the normal carrier, the disposition and amount of the load will have to conform to commercial practices at reasonable costs of transportation. It is essential, however, for adequate space to be left at the top and bottom of the load to allow for proper circulation of the fumigant/air mixture. This may usually be ensured by the use of false floors and by the allowance of 30 or 60 cm (1 or 2 ft) of space between the top of the load and the ceiling. The regulations or administrative instructions of quarantine authorities often contain specific requirements for the amount and arrangement of the loads.

Circulation of the fumigant. In short exposures of two to four hours, proper distribution of the fumigant throughout the structure can be attained only by artificial circulation or by thorough premixing of the fumigant/air mixture. Proper conditions of uniform distribution by such means must always be provided in all plant quarantine treatments, including vacuum fumigation.

When tender plants, fruit and vegetables are being fumigated, excessive and continuous air movements may lead to plant injury. With such materials circulation should be gentle and the output of the fans should be adjusted accordingly. Such adjustment, however, must not be made at the expense of the efficiency of the treatment. Destruction of the quarantined pest is the primary and overriding consideration.

Humidity. Plant injury is minimized if high humidity is maintained during treatment without the deposition of free moisture. For practical purposes this means a relative humidity of 75 percent or slightly over. On the other hand, too much water on the fumigated material or condensed on the inside surfaces of the chamber will seriously reduce the efficiency of hydrogen cyanide (HCN) and ethylene dibromide (EDB), and may lead to considerable injury; with the less water-soluble methyl bromide, it may prevent the fumigant from reaching some of the pests.

Effective humidity control systems are available commercially. These employ spray nozzles through which a fine spray of water is driven by compressed air. Close humidity control is ensured by sensitive "humidistats."

Postfumigation treatment of plants. Injury to growing plants may be reduced by treatment appropriate to the fumigant used. Plants to be exposed to methyl bromide and ethylene dibromide should not be watered for 24 hours unless there is danger of wilting. On the other hand, growing plants fumigated with HCN should be gently but thoroughly washed with water to minimize subsequent injury.


For some plant quarantine treatments, fumigation is combined with cold storage to obtain greater control of insects. The San Jose scale, Quadraspidiotus perniciosus (Comstock), and larvae of the coaling moth, Laspreyresia pomonella (1.), are controlled more effectively on apples with methyl bromide treatment followed by storage at -0.5C than by the use of either of these treatments alone (Moffitt, 1971; Morgan et al 1974, 1975). Also a combination of cold storage for two to four days at 0 to 2C followed by fumigation with methyl bromide (c x t product 75 mg h/l and 10C) is effective in eliminating eggs of the cutworm Spodoptera littoralis (Boisd.) from chrysanthemum cuttings (Powell, 1979). Dosage schedules for fumigation plus refrigeration of fruits are given by USDA (1976).

The possibility of causing injury to fruit with methyl bromide or other fumigants should be taken into account whenever these combined treatments are planned. Benschoter (1979) showed that injury to 'Marsh' grapefruit from methyl bromide varies with early, mid and late season fruit and is greater when fumigation is followed by cold storage than when the fruit is held at ambient temperature. Fumigators are encouraged to make test treatments on small quantities of fruit before applying the combined treatments to commercial shipments.

14. Experimental fumigations

From time to time, users of fumigants may want to carry out a small scale "test" fumigation to determine the suitability of a treatment. For instance, when a commodity or pest organism, not normally fumigated, requires treatment, a number of questions will arise. The choice of fumigant, conditions of the treatment, including dosage, exposure time and temperature, possible effects on the commodity, possible residues, as well as effectiveness in controlling the pest organism, all need to be considered. Even in well-established fumigations, unexpected problems sometimes arise and tests are needed to determine the cause. Small scale tests are also used to check for resistance of insects to fumigants (see FAO, 1975 for methods for detection and measurement of resistance to fumigants).

In addition to conducting experimental treatments for testing procedures, it is good practice to use insects themselves for test purposes in assessing effectiveness of routine treatments. It is rewarding to spend time on the examination of fumigated insects because the observations and recorded results may also be useful for planning future treatments. The assessment of insect mortality requires great care and good judgement to make reliable decisions as to the effectiveness of a fumigation.

In this chapter, procedures for carrying out small-scale experiments are described. The descriptions refer mainly to the two fumigants methyl bromide and phosphine; however, somewhat similar procedures will apply for other fumigants. The practice of using insects for testing effectiveness of treatments is outlined along with methods of handling and rearing the insects. Also included in this chapter are examples of some of the problems encountered in fumigation treatments, along with discussion on ways to avoid these problems.

Experimental treatment

Small-scale tests are carried out in specially constructed small fumigation chambers or other small containers that will accommodate the test material and can be made gas tight. Metal containers such as the portable drum fumigator described in Chapter 8, glass desiccators or flasks, may be adapted for such tests. Other equipment and material will consist of apparatus for measuring and transferring small amounts of fumigant, a source of the appropriate fumigant and in some cases equipment and methods for analysing gas concentrations. In addition, test insects are often needed for determining effectiveness of the treatment.


The general requirements for the fumigation chamber are similar to those outlined for larger chambers in Chapter 8. Good sealing of the door or cover of the chamber is especially important; if a glass desiccator is used, silicone grease on the ground-glass surfaces between the cover and flange will provide a good seal. In addition, some means of introducing the gas through a gas tight seal is necessary. Silicone rubber stoppers or septa fitted in the wall of the chamber to allow injection by a gas syringe are satisfactory for this purpose. If ampoules of liquid methyl bromide are used, an ampoule breaker similar to that shown in Figure 27 is required.

A small fan, magnetic stirrer or simply a swinging "punkah" type fan should be provided for stirring the gas mixture in the chamber. If an electric fan is used, the motor should be of the spark-proof type. Fumigation chambers designed for phosphine should not have motors or other equipment containing copper.

Measurement of Volume

The volume of the chamber should be carefully determined so that dosage of fumigant can be calculated with reasonable accuracy. For rectangular or cylindrical chambers, measuring the dimensions may be adequate for determination of volume. For small chambers, such as desiccators or flasks, the volume is best determined by filling with water and measuring the volume of water in a suitable measuring container, e.g. a graduated cylinder.

Gas Syrinnes

For small chambers around 50 to 100 l in size, methyl bromide and phosphine are most easily applied as gases using gas tight syringes. The size of syringe required will be influenced by the volume of the fumigation chamber as well as the fumigant concerned. For example, a 500 ml syringe is likely to be suitable for most tests with methyl bromide using a chamber of 50 1 volume, while a 100 ml syringe would be adequate for the lower concentrations of phosphine that are needed.


Methyl bromide

Small quantities of this fumigant can be obtained from 454 or 681 9 (1 or 1.5 lb) cans using a can-tapping device specially designed for the purpose (Figure 44). This device consists of a piercing unit made from a 6 mm brass connector fitted with a piercing tube and neoprene gasket at one end and two silicone rubber septa (discs) held in place over the outer end of the connector by a screw cap with a 1 mm hole drilled in the centre. A contoured stainless steel plate for receiving the connector is held firmly to the can by two 90 mm stainless steel clamps.

To pierce the can, the connector is screwed into the threaded hole of the contoured plate until finger tight and then given a quick half turn with a wrench to make the puncture and simultaneously to provide a gas tight seal as the gasket is compressed against the can. Can tapping devices such as this are available from scientific suppliers or they can be made in a machine shop. It should be noted that the use of materials that do not react with methyl bromide, e.g. neoprene, silicone rubber and brass, is important; aluminium fittings should nut be used.

FIGURE 44. Schematic drawing showing cross-sectional view of the can tapping device.

Samples of the gas are taken by placing the can on its side with the piercing unit pointing upwards, inserting the needle of the syringe through the hole of the brass cap to pierce the septa and allowing the methyl bromide to expand into the syringe as shown in Figure 45. Special syringe-needles, with the hole in the side rather than at the tip, should be used to avoid the problem of blocking by particles of septa during the piercing operation. Best results are obtained when the screw cap is loosened slightly before piercing to allow some expansion of the septa. After the needle is withdrawn, the cap is again tightened to produce a tight seal and prevent escape of methyl bromide. The syringe may be flushed with a small amount of methyl bromide to remove air before the required sample is taken. If the fumigant is taken directly to the fumigation chamber, very little gas will diffuse from the syringe needle. However, shut-off valves for syringes are available or the needle tip may be inserted into a silicone septum to provide a temporary seal.

The fumigation chamber should be located nearby with appropriate ventilation to prevent exposure to the fumigant. Gas masks and suitable detection equipment should be available.


This gas is easily collected from commercial aluminium phosphide pellets immersed in a container of water, as described by Kashi and Bond (1975). A glass tube 27 mm in diameter drawn out to a narrow neck at one end is filled with water in a graduated cylinder and fitted (under water) with a silicone rubber septum, as shown in figure 46. A pellet is dropped into the cylinder of water and bubbles of phosphine are collected by displacement of the water in the glass tube.

The required quantity of phosphine is withdrawn through the septum with a gas syringe and transferred to the fumigation chamber.

NOTE. It is advisable to flush the syringe with nitrogen before filling with phosphine to avoid the flame that may occur if sufficient oxygen is present. When drawing phosphine into the syringe, the plunger should be pulled slowly to prevent sudden pressure changes.


The volume of undiluted methyl bromide or phosphine gas required to give a prescribed concentration in any small chamber can be easily calculated from the data given in Chapter 2, Table 2. for example, at 25C the weight of methyl bromide gas that can be contained in 1 m is 3 874.1 9, i.e. 3 874.1 mg in 1 lithe or 3.874 mg in 1 millilitre. To obtain the volume (no. of ml) of methyl bromide required to give 16 mg/l for a 10 1 fumigation chamber, divide the total weight of fumigant required (e.g. 160 mg) by the weight contained in 1 ml (e.g. 3.874 mg) to give 41.3 ml. Similar calculations are made for other temperatures and other fumigants.


After the test material is placed in the fumigation chamber the opening is sealed and the fumigant applied. If test insects are used, care should be taken in selecting representative samples for reliable results.

Experimental fumigations are usually carried out under uniform temperature and humidity conditions, usually 25C and 77 percent relative humidity, with exposure periods of five hours for methyl bromide and 20 hours for phosphine. However, treatments can be altered to simulate conditions present in commercial operations. At the end of the exposure period the chamber is opened to allow the fumigant to disperse in a fume hood or through properly constructed exhaust ducts.

In conducting experimental fumigations all of the precautions that are applicable to commercial treatments should be observed.


The type of information derived from a test fumigation will obviously depend on the objective of the treatment; however, much of the information can be obtained soon after the test is made. Effects on commodity, such as corrosion, colour, flavour or odour changes, may be evident immediately after the treatment. In the fumigation of living plants phytotoxic effects are usually evident within 24 hours. Other effects may be delayed depending on the type of commodity and type of treatment. Gas may continue to desorb from commodities for considerable periods after treatment, making establishment of residue levels difficult. One relatively well-known effect on wheat flour treated with methyl bromide only becomes apparent when a pungent odour is given off from hot bread made from the flour (see Chapter 6).

In assessing the results of test fumigations it may be useful to hold the test material for two to three days, so that due allowance can be made for unexpected effects.

Contents - Previous - Next