Abstract

D i c k , K . M . 1987. Pes t m a n a g e m e n t i n s t o r e d

g r o u n d n u t s . I n f o r m a t i o n Bu l le t in no . 2 2 . Patan-

cheru , A . P . 502 3 2 4 , I n d i a : I n t e r n a t i o n a l Crops

Research Ins t i tu te for the Semi -Ar id Tropics.

W h e n groundnuts are stored, either as pods or kernels,

they are susceptible to insect attack. The publ icat ion

describes the main pest species and br ief ly out l ines the i r

ecology and behavior. The use of traps to detect pest

populat ions in stores is discussed and methods of esti-

mat ing quant i tat ive losses are described. Recommended

pest management techniques are out l ined w i t h empha-

sis on prevent ion of infestat ion by improv ing storage

practices. The potent ia l inputs to an integrated contro l

program are examined. Screening for host-plant resis-

tance to storage pests is discussed and general guidelines

for these experiments given.

The International Crops Research Institute for the Semi-Arid Tropics is a nonprofit scientific educational institute receiving support from donorsthrough the Consultative Group on International Agricultural Research. Donors to ICRISAT include governments and agencies of Australia,Belgium, Canada, Federal Republic of Germany, Finland, France, India, Italy, Japan, Netherlands, Norway, People's Republic of China, Sweden,Switzerland, United Kingdom, United States of America, and the following international and private organizations: Arab Bank for EconomicDevelopment in Africa, Asian Development Bank, Deutsche Gesellschaft fur Technische Zusammenarbeit (GTZ), Global 2000 Inc., InternationalDevelopment Research Centre, International Fertilizer Development Center, International Fund for Agricultural Development, The EuropeanEconomic Community, The Ford Foundation, The Leverhulme Trust, The Opec Fund for International Development, The Population Council,The World Bank, United Nations Development Programme, and the University of Hohenheim. Information and conclusions in this publication donot necessarily reflect the position of the aforementioned governments, agencies, and international and private organizations.

Correct citation: Dick, K.M. 1987. Pest management in stored groundnuts. Information Bulletin no. 22. Patancheru, A.P. 502 324, India:International Crops Research Institute for the Semi-Arid Tropics.

Cover: Groundnut pods and kernels with characteristic damage caused by Caryedon serratus larvae.Note also eggs attached to pods and kernels, and distinctive pupal cocoons.

Pest Management

in Stored Groundnuts

K. M. Dick

ICRISAT

Information Bulletin no. 22

International Crops Research Institute for the Semi-Arid Tropics

Patancheru, Andhra Pradesh 502 324, India

1987

Contents

I n t r o d u c t i o n

M a j o r Pests o f S t o r e d G r o u n d n u t s

C a r y e d o n se r ra tus

T r i b o l i u m c a s t a n e u m

O r y z a e p h i l u s m e r c a t o r

T r o g o d e r m a g r a n a r i u m

E l a s m o l o m u s s o r d i d u s

C o r c y r a c e p h a l o n i c a

E p h e s t i a cau te l l a

P l o d i a i n t e r p u n c t e l l a

O t h e r species

D e t e c t i o n o f Insect Pest P o p u l a t i o n s

S t i c k y t r a p s

R e f u g e t r a p s

L i g h t t r a p s

P i t f a l l t r a p s

L o s s Assessment

S a m p l i n g

Q u a n t i t a t i v e loss d e t e r m i n a t i o n

Insec t Pest M a n a g e m e n t

P r e v e n t i o n o f i n f e s t a t i o n

C o n t r o l o f es tab l i shed i n f e s t a t i o n s

I n t e g r a t e d pest m a n a g e m e n t

B r e e d i n g f o r res is tance t o s to rage pests

Suggested f u r t h e r s tud ies

3

3

3

4

5

5

6

6

7

7

8

8

8

9

10

10

11

11

18

21

21

23

24

24

25

B i b l i o g r a p h y 26

Introduction Major Pests of Stored Groundnuts

G r o u n d n u t , A r a c h i s h y p o g a e a L . , i s an

i m p o r t a n t cash a n d f o o d c r o p i n m a n y

par ts o f the t r o p i c s , p a r t i c u l a r l y i n s e m i -

a r i d areas. W h e n s to red be fo re use the

g r o u n d n u t c r o p i s suscept ib le t o a t t a c k

by insect pests. T h e degree o f suscep t i -

b i l i t y depends l a rge l y o n w h e t h e r o r n o t

the g r o u n d n u t s are shel led a n d the

ex ten t t o w h i c h pods o r ke rne ls are d a m -

aged be fo re be ing p laced in s tore . Insect

i n f e s t a t i o n causes loss in d r y mass o f the

kerne ls , increased levels o f f ree f a t t y

ac ids i n the o i l ( t h e r e b y l o w e r i n g the

q u a l i t y ) a n d , i f t he seeds are heav i l y

d a m a g e d , r e d u c t i o n i n g e r m i n a t i o n

p o t e n t i a l . T h e heat a n d m o i s t u r e gener -

a ted by la rge insect p o p u l a t i o n s w i t h i n

heaps o r s tacks o f g r o u n d n u t s m a y a lso

increase the r i sk o f m o l d g r o w t h .

I n Wes t A f r i c a the ex ten t o f p o s t h a r -

vest losses has p r o m p t e d several s tud ies

o f insect p o p u l a t i o n d e v e l o p m e n t o n

g r o u n d n u t s s to red as pods a n d kerne ls .

H o w e v e r , f ew a t t e m p t s have been m a d e

to measure the ex ten t o f losses caused by

insects e i the r i n f a r m e r s ' s tores o r i n

large-scale c o m m e r c i a l s to rage . C o n s e -

q u e n t l y , there is a lack o f i n f o r m a t i o n on

a p p r o p r i a t e m e t h o d s f o r assessing pos t -

harves t losses.

T h i s b u l l e t i n p rov i des i n f o r m a t i o n o n

the m a j o r insect pests o f s to red g r o u n d -

nu ts i n the sem i -a r i d t r o p i c s ( S A T ) ,

m e t h o d s o f assessing the losses caused

by these insects, a n d pest m a n a g e m e n t

prac t ices t h a t can he lp to reduce losses.

T h e s to rage o f g r o u n d n u t p r o d u c t s such

as cake , mea l , o r o i l i s n o t d iscussed.

Records list over 100 insect species that infeststored groundnuts (Redlinger and Davis, 1982).Only those species considered to be major cos-mopolitan pests are described here.

Caryedon serratus (Olivier)Coleoptera:Bruchidae

Common names: groundnut borer,groundnut weevil

Synonyms: Bruchus serratus OlivierBruchus gonagra FabriciusCaryedon fuscus (Goeze)Caryedon gonagra (Fabricius)

Figure 1. Groundnut borer (Caryedonserratus) adult ( x10) and larva ( x10).

3

This species is found in many parts of tropicalAsia and Africa, breeding on common treelegumes such as Tamarindus indica L. (tama-rind) as well as on harvested groundnuts. It isgenerally regarded as the only species that canpenetrate intact pods to infest the kernels. Theadult is 4-7 mm long and reddish-brown, withdark irregular markings on the elytra (Fig. 1). Ithas large, prominent eyes and can be easily dis-tinguished from other storage pests by its broadhind femur, which bears a conspicuous comb ofspines.

Infestation of harvested groundnuts can occurwhile the crop is drying in the field or when it isstored near infested stocks or crop residues.Adult females attach their eggs to the outside ofpods or kernels. When the first instar larvahatches it burrows directly through the pod wallto reach the kernel. Each larva feeds solelywithin a single kernel (Plate a). When mature,larvae may partially or completely emerge fromthe pod, leaving a characteristic round holeapproximately 3 mm in diameter. Larvae oftenmigrate to the bottom of a stack or heap beforepupating in distinctive ovoid cocoons (seecover). Damage caused by subsequent genera-tions is, therefore, often heaviest in this part ofthe stock.

Under optimal conditions, 30-33°C and 70-90% relative humidity (RH), the mean develop-ment period is approximately 40 d, but there islikely to be wide variation round the meanwithin each individual population.

Tribolium castaneum (Herbst)Coleoptera:Tenebrionidae

Common name: rust-red flour beetle

This species is found throughout the tropics andis regarded as a major pest of shelled ground-nuts. The adults are 3-4 mm long, chestnut-brown in color (Fig. 2), and have a lifespan ofseveral months. The beetles are strong fliers andare often the first species to colonize storedgroundnuts. The females lay their eggs in cracksin the testa or in holes in the kernels created bythe adult while feeding (Plate b). Thus the firstinstar larvae, which cannot penetrate intact tes-tae, are able to feed directly on the cotyledons.

The larvae are cylindrical in shape with twoprominent 'horns' on the last abdominal seg-ment. They pupate inside damaged kernels with-out forming a cocoon. The adults and larvae arefacultative predators of eggs and pupae of otherstorage pests and are also cannabalistic. Themean development period from egg to adult isapproximately 32 d at 30°C and 90% RH but istwice as long at 70% R H .

4

Figure 2. Rust-red flour beetle (Triboliumcastaneum) adult ( x15) and larva ( x10).

Figure 3. Merchant grain beetle (Oryzaephilusmercator) adult (x20) and larva (x15).

Oryzaephilus mercator (Fauvel)Coleoptera:Silvanidae

Common name: merchant grain beetle

This cosmopolitan species has similar ecologyand behavior to Tribolium castaneum. Theadults are 2.5-3.5 mm long with a distinctiveridged prothorax bearing six large teeth oneither side (Fig. 3). The larvae are cream-coloredand possess more prominent thoracic legs thanthose of T. castaneum. The first instar larvaecannot penetrate intact testae and must feed onkernels damaged by the adults or by other stor-age pests. Under optimal conditions, 30-33°Cand 70% RH, the life cycle is completed in 28-35d.

O. mercator is difficult to distinguish morpho-logically f rom its sibling species O. surinamensis,the saw-toothed grain beetle, but the latter ismore usually associated with cereal products.

Trogoderma granarium EvertsColeoptera:Dermestidae

Common name: khapra beetle

This species is tolerant of hotter, drier conditionsthan many other storage pests and is commonlyfound in semi-arid areas of Africa, West Asiaand northern India. It has not been recordedfrom Southeast Asia, South America, or Austra-lasia. The adults are oval in shape, 2-3 mm long,and dark brown with black mottl ing (Fig. 4).Their dorsal surface is covered with fine hairs.Adults live for about 14 d and do not feed or fly.The larvae are straw-colored and from the 4thinstar onwards have dense tufts of hair on eachabdominal and thoracic tergite.

The bionomics of this species on groundnuthave not been rigorously examined in the labo-ratory, and the optimal conditions for its devel-opment are still a matter for debate. Its morecommon occurence in hot, dry areas is generallyattributed to its inability to compete with faster-breeding species in more humid environments.

Figure 4. Khapra beetle (Trogodermagranarium) adult (x20) and larva (*10).

5

Some larvae of certain strains can enter diapausein response to adverse conditions, e.g., extremesof temperature or overcrowding. When almostmature, the prediapause larvae often leave thestored commodity to enter crevices in the storagestructure where they can remain without feedingfor many months. In this state, metabolic activ-ity is low and the larvae are extremely resistantto insecticides. Complete disinfestation is di f f i -cult and for this reason many countries rejectconsignments of a commodity found to beinfested by this species.

Elasmolomus sordidus (Fabricius)Hemiptera:Lygaeidae

Synonym: Aphanus sordidus (Fabricius)

This bug is widespread in tropical Africa andIndia, occurring on pods left drying in the fieldand in store. The adult is dark brown, approx-imately 10-mm long, and 2-mm wide (Fig. 5). Inthe field females lay their eggs in the soil or ongroundnut haulms but, in store, eggs are laidloosely among the groundnuts or in sacking. Thefirst instar nymphs have a bright red abdomen;later instars become progressively darker. A l lstages feed on kernels, perforating the pod withtheir rostrum. This causes the kernels to shrivel(Plate c) and increases the free fatty acid contentof the oi l , producing a rancid flavor.

Figure 5. Elasmolomus sordidus adult (x5).

Figure 6. Rice moth (Corcyra cephalonica)adult (x5) and larva (x5).

Corcyra cephalonica (Stainton)Lepidoptera:Pyralidae

Common name: rice moth

This species has the ability to develop at lowhumidities (< 20% R H). This may account for itsprevalence in the SAT over other stored productlepidopteran pests. The adult is brown and 12-15mm long with its wings folded (Fig. 6). The labialpalps point directly forward and are long andpointed in the female, but short and inconspicu-ous in the male. The adults do not feed. Femaleslive for 1-2 weeks and scatter their eggs amongthe produce. The larvae are capable of damagingintact kernels and feed both on the surface andwithin seeds. They spin a tough silken fibre,

6

webbing together kernels, frass,and cast larvalskins (Plate d). This type of contamination, eas-ily distinguished from the fine dust that resultsfrom beetle infestation, may be of greater eco-nomic importance than the weight loss caused bylarval feeding.

Pupation takes place either within the foodsource, in sacking, or in crevices in storage struc-tures. At 28° C and 70% RH the life cycle fromegg to adult takes 40-50 d. Male moths emerge,on average, 1-2 d before females.

Ephestia cautella (Walker)Lepidoptera:Pyralidae

Common names: tropical warehouse moth,almond moth

Synonym: Cadra cautella (Walker)

This is a pest of many stored commoditiesthroughout the tropics but is less common inarid areas. The adults are greyish-brown with an

Figure 7. Tropical warehouse moth (Ephestiacautella) adult (x8) and larva (x6).

indistinct pattern on the forewings (Fig. 7). Thismoth is smaller than C. cephalonica, being 6-9mm long at rest. The labial palps in both themale and female point upwards. Its life cycle issimilar to that of other moths infesting storedproducts. The larvae are extremely mobile andmove freely through the produce, contaminatingit with fine silk webbing. At 28°C and 70% RHthe life cycle takes 40-50 d.

Plodia interpunctella (Hubner)Lepidoptera:Pyralidae

Common name: Indian meal moth

This species appears to be more prevalent incooler areas of the tropics, e.g., highlandregions. Its habits and life history are similar tothose of the moth species already described. Theadult larvae are easily recognized by the mark-ings on their forewings: the basal third is cream-colored while the rest is reddish-brown. Adultsare 8-12 mm long at rest, with labial palps thatpoint directly forwards (Fig. 8). Under adverseconditions, e.g., extreme temperatures or over-crowding, the life cycle may be prolonged by a larval diapause. During diapause, normal appli-cations of insecticides, especially fumigants, mayprove ineffective.

Figure 8. Indian meal moth (Plodiainterpunctella) adult (x6) and larva (x5).

7

Other species Sticky traps

Tenebroides mauritanicus (L.)Coleoptera:Trogossitidae

Lasioderma serricorne (F.)Coleoptera:Anobiidae

Latheticus oryzae WaterhouseColeoptera:Tenebrionidae

Cryptolestes sppColeoptera.Cucujidae

Alphitobius sppColeoptera:Tenebrionidae

Carpophilus sppColeoptera: Nitidulidae

These species occasionally infest stored ground-nuts but are only considered to be of minorimportance. They usually occur in associationwith one or more of the major pests describedabove. Illustrated keys for their identificationcan be found in T D R I (1984) and Freeman(1980).

Any surface coated with a sticky substance (suchas petroleum jelly, or the polybutene gel usuallysold as bird repellant) that prevents an insectfrom leaving after landing on it can be used as a sticky trap. Several trap designs include the useof a pheromone source. Although these mayvary in shape, the basic design (similar to thatshown in Figure 9) has a pheromone sourcesuspended near a surface covered in a stickysubstance. A 'fly paper' baited with the sexpheromone component common to Ephestiaspp. and P. interpunctella, is available. (MothIndicator® from Detia, P.O. Box 10, 6941 Lau-denbach/Bergstrasse, Federal Republic ofGermany.)

Sticky traps should be suspended from thestore roof, to hang above or between stacks orheaps of groundnuts. One problem with the useof these traps in stores is the short effective life ofthe sticky surface which soon becomes coveredin dust. This wil l be a particular problem if a trapis placed near a groundnut decorticating unit.

Suspensory cord-

Detection of Insect Pest Populations

It is very important to detect low-level infesta-tions of storage pests if control measures are tobe implemented in sufficient time to preventlosses. The use of traps for this purpose oftenrequires less time and effort than more conven-tional methods of inspection, such as 'spear'sampling. Traps cause less damage to the com-modity, and wil l often provide the first evidenceof an infestation that has developed betweenstore inspections. Although insect traps areeffective in detecting infestations, it is difficult toestimate insect population density from trapcatches. A variety of traps are available differingin cost, sophistication, and in the range of insectsattracted to them.

Card folded to form'tent' with open ends

Pheromone source

Removable card coated on upper surfacewith sticky polybutene gel

Figure 9. Sticky trap for flying insects.Adapted from Figure 11, Anon. 1983.Courtesy of T D R I .

8

Refuge traps

These traps can be made from waste materialsuch as cardboard packaging. They provide a refuge for insects such as moth larvae whichhabitually leave the food source to pupate insacking or in crevices in the storage structure.Refuge traps should be placed around the baseof sack stacks and between surface sacks. Insectswill be prevented from leaving the trap if the

interior is sprayed with insecticide. These trapscan also be made more attractive to a givenspecies by baiting them with the appropriateaggregation pheromone. A trap similar to thatshown in Figure 10, baited with the aggregationpheromone of T. granarium, is commerciallyavailable. (Trogotrap® from Degesch AG, 32-40Weissmullerstrasse, D-6000 Frankfurt, FederalRepublic of Germany.)

Folded trap,corrugations inside,held together with rubber band

Figure 10. Refuge trap made from corrugated cardboard tha t can be baited with a pheromone.Adapted from Figure 2, Anon. 1983. Courtesy of T D R I .

9

Light traps

These traps are most efficient at detecting mothinfestations since the adults are attracted to thelight when they leave the produce in order to flyand mate. This is not the case with stored productbeetles, however, which often remain deepwithin the commodity for generations and leaveonly when conditions become unsuitable, e.g., asa result of high population density or hightemperature.

Light sources of high intensity but low watt-age are most suitable for attracting insects insidea food store. Ultraviolet (300-400 nm) and greenlight (500-550 nm) appear to be the wavelengthsmost attractive to storage pests. Ultraviolet (uv)tubes of suitable size and wattage (6, 15, or 32 W)are commercially available. Tubes or bulbs that

6W, 21 cmuv fluorescent tube

Metal bracketfor wall mounting

Store wall

Flex to power supplyPlywood baffles paintedwhite and covered withpolythene sleeves, coatedwith sticky polybutene gel

Figure 11. Simple, wall-mounted, unidirec-tional light trap developed by T D R I for usein stores. Adapted from Figure 7, Anon. 1983.Courtesy of T D R I .

emit green light are less easy to obtain but can bemade by covering white light sources with a heat-resistant green filter. These light sources can beoperated at 110/240 V AC (mains electricity)using a choke, or 12 V DC (car battery). To savepower they need only be operated when insectswithin the store are most likely to be flying, i.e.,for 2-3 h around dusk.

In stores, unidirectional traps are most conven-ient since these can be wall- or ceiling-mounted,to avoid interference in stock movements. A sim-ple trap of this type, developed at the TropicalDevelopment and Research Institute (TDRI ,Storage Department, London Road, Slough,UK) consists of two baffles of wood or metal,painted white and positioned on either side of a straight 6 W uv light tube (Fig. 11). Polythenesleeves coated on one side with sticky polybutenegel are placed over the baffles. Insects areattracted to the light, collide with the baffles, andare caught on the glue. The plastic sleeves can beeasily removed to examine the catch.

Lightproofing the store will prevent largenumbers of other species being attracted to thetrap from outside. The efficiency of a light trapcan also be improved by mounting a fan with a collecting funnel behind the light source. How-ever, such a trap may prove impractical in manystorage situations in the tropics as it requires a mains electricity supply and regular main-tenance.

Pitfall traps

Some pests of stored products, for example,Tribolium spp, cannot climb smooth surfacessuch as glass. Therefore, a small glass test tube orvial buried up to its lip in a bulk of groundnutswill trap adult Tribolium which fall into it whilewalking across the surface of a heap of pods orkernels. A pitfall trap has also been developedfor insertion into bulked commodities (Fig. 12).This consists of a metal tube 20-cm long, the tophalf of which is perforated with holes that allowinsects, but not the commodity, to fall into thetube. Insects passing through the holes fall into a glass tube in the lower half of the trap, which

10

Figure 12. Pitfall trap suitable for use in bulkgroundnuts. Adapted from Loschiavo andAtkinson 1973.

could contain a small pheromone source. Thetrap is inserted into the grain with a pushrod andretrieved using a cord attached to the top of thetrap.

Loss AssessmentIf any of the pest species described above arefound, either in traps or in stored groundnuts, itis likely that some damage wil l have been causedto the kernels. Methods of estimating the extentof storage losses are outlined here, with particu-lar emphasis on standard techniques that requirea minimum of equipment. Only quantitative lossis dealt with because the techniques presentlyavailable for assessment of qualitative loss areless well defined. Where the methods describedare applicable to a number of different crops, theterm 'grain' is used synonymously with ground-nut 'pod' or 'kernel'.

Sampling

In estimating the losses caused by insect pests toa consignment of any stored commodity, it is notpractical to examine every grain. Measurementof the loss to the commodity as a whole has to bebased on the loss recorded from a number ofsamples. These samples cannot simply beremoved from the commodity where it is mostaccessible, e.g., from the top of a sack, or fromthe surface of a large heap. Insect infestations areseldom uniformly or even randomly distributedwithin a stock of stored grain. Thus, to obtainsamples that give a true indication of the loss,methods must be used which ensure that everygrain has an equal chance of selection.

The theoretical principles of representativesampling should be applied to all types ofgroundnut storage regardless of the scale of thestorage operation. However, the practical prob-lems involved wil l differ with the storagesituation.

Sack storage. Sampling from grain stored insacks usually involves numbering all the sacks ina stack or warehouse, and using randomnumbers to decide which of the sacks are to besampled on any one occasion. With large con-signments, the conditions of storage may varymarkedly between the sacks, e.g., the tempera-ture at the center of a stack may be different fromthat at the surface. These differences should betaken into account by using a 'stratified' sam-pling procedure. At its simplest, this involves thedivision of a single stack into a number of layers(or strata), each containing the same number ofsacks. A given number of sacks in each layer isthen chosen at random for sampling.

As a practical guide, the optimum number ofsacks to be sampled from consignments of differ-ing size is as follows:• 10 sacks or less, sample each sack;• between 11 and 100 sacks, sample 10 sacks;• more than 100 sacks, the number to be

sampled equals the square root of the total.

11

Spring-loaded capgiving access tocollection tube

Glass collection tube

Perforated brass cylinder,holes allow insects to enter

Socket for push rod

Nylon cord fortrap retrieval

Figure 13. T D R I Produce Flow Sampler used for sampling bagg ed groundnuts.Adapted from Ashman 1973. Courtesy of T D R I .

Sample collector Sack

Collectingfunnel

Supportingframe

• Cone

HopperBung

Operator

Ideally, the sacks wil l be numbered and thefirst samples removed, when the consignment isbeing placed in storage. This provides a baselinemeasure against which the losses recorded insubsequent samples can be compared. Once thesacks have been stacked many of them are in-accessible. To obtain representative samples, thestack must be dismantled and this wi l l inevitablyinvolve some expenditure on labor and disrup-tion of normal stock movements within thestore. When stacks are broken down for sam-pling, the sacks should be replaced in their origi-nal positions so that the distribution of insectswithin the stacks is affected as little as possible. Itis stressed that if samples are taken only from themost accessible sacks then the loss measure-ments obtained only represent that part of thetotal bulk from which they were collected. Sim-ilarly, if stocks are removed during the surveythen loss levels in subsequent samples must beapplied only to that part of the original materialstill in store.

Just as each sack to be sampled must beselected without bias, every grain within a sackmust have an equal chance of inclusion in thefinal sample. Specialized equipment is availablefor removing representative samples from sacksof groundnuts, e.g., the T D R I Produce FlowSampler (Fig. 13), or for reducing the size oflarge samples without bias in the laboratory,e.g., Boerner or Garnet® dividers (Fig. 14a and14b). However, some sampling devices in wide-spread use, such as sack 'spears' or probes, donot provide representative samples and can givemisleading results (Golob, 1976).

If no suitable equipment is available thenrepresentative samples can be obtained by 'con-ing and quartering'. This involves emptying a sack onto a smooth surface, thoroughly mixingthe pods or kernels by hand and forming a cone-shaped heap which is then divided into quartersusing a flat board. Two opposite quarters arereturned to the sack. The remaining two are re-combined and the process is repeated unti l a sample of the appropriate size is obtained (Fig. 15).

Figure 14a. Boerner divider, approximately1 m high, used in laboratory sampling.

Figure 14b. Garnet® divider, approximatelylm high, used in laboratory sampling.

13

Plate a. Groundnut kernels damaged by Caryedonserratus. Note larval feeding site inside cotyledons,pupal cocoons formed largely inside kernels, and cocoon bet ween kernel and pod wall.

Plate b. Groundnut kernels damaged by Tribolium castaneu m. Note holes in testae and fine dust-likefrass typical of beetle infestations.

Plate c. Groundnut kernels (left) shrivelled as a result of feeding by Elasmolomus sordidus. Fungalinvasion is common where the bug has penetrated the testa.

Plate d. Groundnut kernels damaged by Corcyra cephalon ica. Exposed cotyledons, coarse frassandmasses of webbing are typical of moth infestations.

Figure 15. Coning and quartering techniquefor representative sampling without equipment;a. forming cone; b. mixing pods; c. halving; d.quartering; e. returning opposite quarters tosack; f. remaining quarters remixed (left), andapproximate size of final sample (right).

Bulk storage. Representative samples can beaccurately taken from groundnuts in sacks,because the consignment to be examined can bedivided into easily identifiable units (the sacks)which can then be numbered and sampled atrandom. To some extent, this can also be donewith bulk storage, when the store is being filled.The containers used to transport the crop to thestore can be designated as the sampling units,these can be numbered and then selected at ran-dom for sampling. Thereafter, this division ofthe bulk into readily defined units is lost and it isnot practical to use the same methodology asthat recommended for sack stores.

Where the dimensions of the bulk store areknown, it may be possible to use a systematicsampling procedure similar to that recom-mended for the inspection of railway wagons(Fig. 16). Such a sampling procedure is of littlevalue, however, unless samples can be obtained

Container < 15 tonnes capacity : five sampling points (middle,andapproximately 50 cm from sides).

Container 15 to 30 tonnes capacityeight sampling points.

Container 30 to 50 tonnes capacity : eleven sampling points.

Figure 16. Recommended sampling positionsin bulk grain containers. Based on Anon. 1983.Courtesy of T D R I .

Figure 17. Double-tube sampling speardesigned for use with small-grainedcommodities.

from below the surface of the bulk. This can bedone using a variety of devices. The simplest ofthese is the double-tube sampling spear (Fig. 17).These can be up to 3.7-m long, and as the collect-ing tube is divided into a number of compart-ments it is possible to record the depth fromwhich each part of the sample was removed.

17

Handle

Inner tube

- Outer tube

Opening

Opening inouter tube

Opening in inner tube

Partition

Rotation of innertube through 180°

Openings in outerand inner tube now aligned

Sampled produce

However, even with this equipment it may not bepossible to reach the bottom of a large heap orcontainer, where some insect species tend toaccumulate, and where damage can be mostsevere. Thus, the samples obtained with thespear may not be representative and data socollected can only be applied to that part of theproduce from which samples were removed.

Most of the devices available for samplingfrom deep bulk stores were designed for use withsmall-grained commodities such as wheat andrice. It is doubtful whether they would workeffectively for in-shell groundnuts withoutadapting the standard design.

Quantitative loss determination

Experience has shown that sampling at monthlyintervals is generally sufficient. When a delaybetween sample collection and analysis isexpected the samples should be placed in plasticbags with a drop of liquid fumigant such ascarbon tetrachloride in order to prevent furtherdevelopment of the insect population in thesample.

There are several acceptable methods for esti-mating the mass (weight) loss to stored cerealsand pulses caused by insect infestation. There is,however, little experience in using any of thesemethods with groundnuts. The procedureselected depends on factors such as the availabil-ity of equipment and the estimated number ofsamples to be collected. Whichever method ischosen, groundnut samples must be shelledbefore mass loss can be assessed. As the wet massof groundnuts wil l change with the ambient RHit is usually necessary to perform the calculationsusing the dry mass of each sample. This can beobtained either by placing a subsample of thegroundnut kernels in a suitably calibrated mois-ture meter, or by drying a number of sub-samples (5 x 10 g) in an oven at 103 ±2° C for 16 h.

Standard volume/mass method. The accuracyof this method depends on obtaining an exactstandard volume of grain using a simple appara-tus called a chondrometer (Fig. 18), that dropsan amount of grain from a fixed height into a

container of precise volume. The technique relieson the assumption that the volume occupied bythe same number of damaged or nondamagedgrains wil l be identical, but the mass of thisstandard volume wil l decrease as the level ofdamage increases.

The relationship between the dry mass and themoisture content of the standard volume of non-damaged grain at the time of storing is plotted ona graph. The dry mass of standard volumes ofgrain from later samples can then be comparedto that of the initial sample, and the percentagemass loss calculated as follows:

where:Dl = dry mass of the standard volume at the

beginning of the experiment (read fromthe graph using the same moisture con-tent as that obtained for DX) , and

DX = dry mass on occasion X.

In large-scale surveys, which may includenumerous sampling sites and different crop va-rieties, grain size often varies markedly betweensamples and a single volume/mass relationshipcannot be applied to all the samples.

The standard volume/mass method can beadapted to allow for situations where baselinesamples could not be collected. Each sample canbe divided into damaged and apparently non-damaged portions. The mass loss is the differ-ence in dry mass between the nondamaged anddamaged portions. With this 'adapted' methodrelatively large samples (> 1 kg) may be requiredin order to obtain enough damaged or nondam-aged grain to measure the mass of the standardvolume.

Use of a chondrometer wil l only provide accu-rate results if damage does not affect the packingof grains in the measuring container. If packingis affected, the mass of the standard volume mayactually increase as the damage levels increase.

When the 'adapted' method is used, 'hidden'infestation, e.g., by C. serratus, can result in anunderestimation of losses because grains withsuch an infestation are included in the nondam-

18

Filling tubeof fixed height

Dividing plate

Containerof precise volume

Figure 18. Chondrometer for use in standardvolume/mass method of quantitative lossassessment, approximately 30 cm high.

aged portion of the sample. Use of this methodintroduces another source of error because itassumes that insects select grains at randomeither for feeding or oviposition or both; anassumption known to be invalid for certainspecies.

Thousand-grain mass (TGM) method. In thismethod, a sample taken when the commoditywas placed in store is weighed, the number ofgrains counted and their moisture content deter-mined. The dry mass of 1000 grains is thenobtained using the formula:

where:m = wet mass of the working sample;H = percentage moisture content of the grain;

andN = number of grains in the working sample.

The mass loss to subsequent samples as a result of infestation is calculated by using theformula:

where:Ml = T G M of the grain at the beginning of the

study; andMX = T G M of the grain on occasion X.

This method resembles, in part, other proce-dures that relate numbers to mass of grains butdoes not involve separation of 'damaged' and'nondamaged' grains or adjustment of the work-ing sample to an exact mass or volume. It thusavoids several possible sources of error or bias.

Count-and-weigh method. This method in-volves weighing and counting only, and gener-ally requires smaller samples than the standardvolume/mass method. Each sample is dividedinto damaged and nondamaged portions. Thegrains in each portion are then counted andweighed, and the moisture content of the sampledetermined. Mass loss is calculated as:

where:Nu = number of nondamaged grains;Nd - number of damaged grains;

U = dry mass of nondamaged grains; andD = dry mass of damaged grains.

As this method involves the separation ofdamaged and nondamaged grains, the samesources of error are present as in the shortenedstandard volume/mass method. However, itdoes have an advantage over the other methodsdescribed here in that it allows damage by differ-ent species to be separately assessed. The infor-mation thus obtained could be of importance indetermining the most effective control measures.

19

The count-and-weigh method was used toassess losses to in-shell groundnuts stored insacks at an oil mi l l warehouse near Kurnool inAndhra Pradesh, India. The full procedure

employed in the laboratory analysis is given inFigure 19. After 5 months in store there was a 20% loss in dry mass of kernels. This was almostentirely due to damage caused by C. serratus.

Figure 19. Flow chart of laboratory analysis procedure used to assess losses to in-shell groundnuts bycount-and-weigh method.

20

Collected sample

Damagedpods

Insectdamage

Prestoragedamage

Shell

Insectscollectedfrom pods

Damagedseeds

Nondamagedseeds

Identifycausal agent

Corcyra TriboliumOryzaephilusTenebroides

Caryedon

Clean, Count, Weigh

Insectscollectedfrom seeds

Drysubsamples

Moisturecontent

of samples

Estimate'hidden'

infestation

Secondestimate ofmass loss

Firstestimate ofmass loss

Sieve, Count, Weigh

Damagedseeds

Nondamagedseeds

Shell

Nondamagedpods

The mass loss caused by early instar larvae (hid-den infestation) was estimated by dissecting sub-samples of kernels initially classed as non-damaged and recording the numbers containinglarvae. The mean mass of kernels containinghidden infestation was obtained by taking anaverage of the masses of 'damaged' and 'non-damaged' kernels. These values were used toestimate the total number and mass of kernelswith hidden infestation and the mass loss causedwas calculated using the formula given above.The results showed that if hidden infestation wasignored, losses were underestimated by 1-2% ofthe init ial dry sample mass.

Insect Pest Management

In the SAT the high cost of insecticides, thefrequent nonavailability of suitable formula-tions and packaging, and the increasing inci-dence of insecticide resistance necessitates anapproach to postharvest pest management basedon good storage practice. When determining theneed for insecticide application, the economi-cally acceptable level of insect infestation mustbe considered. This wi l l depend on whether thegroundnuts are destined for oil production, localconsumption, resale as seed, or export.

Prevention of infestation

Good storage management and hygiene are ofgreat importance in preventing insect infestationof stored produce. Groundnuts must be driedproperly after harvest, to reduce the moisturecontent of the kernels to below 7%, the upperlimit for safe storage. At high moisture levels,insect populations develop more rapidly andthere is an increased risk of invasion by toxigenicfungi, with a consequent danger of aflatoxincontamination.

Before stores are refilled they should be thor-oughly cleaned and the residues from the pre-vious crop removed and preferably burned. Ifthe store or container is known to have heldinfested stocks then it is advisable to apply an

insecticide to the interior surfaces of the storeafter it has been cleaned (Table 1). Alternatively,empty stores that can be made relatively gas-tight can be disinfested using dichlorvos as a space treatment. Aerosol formulations of thisinsecticide are available but these require the useof an aerosol generator. Strips of PVC impreg-nated with dichlorvos are cheaper and more con-venient for use in small stores. Placed in a storeat a rate of I strip 30 n r 3 these strips shouldcontrol stored-product moths for 3 to 4 months.

Used gunny sacks should be checked for thepresence of insects before they are filled again. Ifnecessary, sacks can be disinfested by rollingthem up together and placing them in a sealed oildrum with a single phosphine tablet.

In sack stores, stacks should be constructed onwooden pallets to reduce the possibility ofground water seeping into the bottom sacks.This would cause the groundnuts in these sacksto become moldy. To facilitate fumigation, thestacks should be of a uniform size, and a space ofat least 1 m should be left between stacks. In oi lmills or transit warehouses, consignments of dif-ferent ages should be stacked separately andremoved in sequence for processing or shipment,even if this involves extra expenditure on laborcosts of stock movements.

Pod storage. As most postharvest groundnutpests are unable to penetrate intact pods, leavingthe crop in the shell for as long as possible duringstorage is an effective method of l imit ing dam-age. C. serratus wi l l , however, attack ground-nuts stored as pods. Since infestation of cleanstocks wil l usually begin in the surface layers of a stack or bulk of groundnuts, the application ofan insecticide spray or dust wil l provide somemeasure of protection against this pest. In sackstores, the sacks on the surface of each stack canbe sprayed with any of the insecticides recom-mended for residual application on store wallsetc , although at a lower rate of application(Table 1). Spraying stacks layer by layer is likelyto be more effective than a single surface spraybut involves greater expenditure on insecticideand labor, and causes considerable interferencewith routine sack-stacking and stock movement.

21

The same insecticides can be applied to thesurface layer of pods in bulk stores. Alter-natively, dust formulations of insecticides suchas malathion, fenitrothion + carbaryl, or bromo-phos can be applied to the surface (Table 1).Bromophos dust (2% a.i.) applied at a rate of 200

g m-2 to the surface and base of large heaps ofpods in the open air, effectively controlled C. ser-ratus in Senegal (Pointel, Deuse and Hernandez,1979). If necessary, small quantities of pods canbe protected by the admixture of insecticidaldust. This should be done by emptying the

22

Table 1. Summary of chemical control measures for protecti on of stored groundnuts.

Control operations

Prevention of infestationApplication of insecticidalspray to interior surfacesof infested stores beforerefilling

Space treatment of emptystores before refilling

Direct application ofspray to pods or to sackscontaining pods or kernels

Surface application to,or admixture ofinsecticidal dust withpods

Control of established infestationFumigation of bagged or bulkstocks in gas-tight stores,or under gas-tight sheeting

Fumigation of small quantitiesof pods or kernels inpolythene-lined sacks orcontainers such as oil drums

1. a.i. = active ingredient; e.c. = emu

Insecticide common nameand formulation1

Malathion (w.p.)Fenitrothion (e.c.)Chlorpyriphos-methyl (e.c.)Pirimiphos-mcthyl (e.c.)lodofenphos (s.c.)Deltamethrin (w.p.)

Dichlorvos (resin strips)

Malathion (w.p.)Fenitrothion (e.c.)Chlorpyriphos-methyl (e.c.)Pirimiphos-methyl (e.c.)lodofenphos (s.c.)Deltamethrin + piperonylbutoxide (e.c.)

Malathion (dust) Deltamethrin (dust)Fenitrothion + carbaryl (dust)Bromophos (dust)

Methyl bromide + chloropicrin (gas)

Phosphine (solid aluminiumphosphide)

Phosphine (solid)Ethylene dibromide (liquid)Ethylene dibromide + carbontetrachloride, 1:8 mixture (liquid)

Application rate of whole productwith specified a.i. concentration2

250 g 25% a.i. in 5L water 100 m-2

200 ml 50% a.i. in 51. water 100 nr2

200 ml 50% a.i. in 5L water 100 nr2

200 ml 50% a.i. in 5L water 100 m-2

200 g 50% a.i. in 5L water 100 m-2

50 g 2.5% a.i. in 3.3L water 100 nr2

1 strip 30 m-3

Apply at half the raterecommended above

10 ml in 990 ml water t-1

500 g 2% a.i. t-1

500 g 0.2% a.i. t-1

500 g 1.2% + 0.8% a.i. t-1

200 g 2% a.i. m-2 of surface area

10-15 g m-3 for 24 h (dosage increasedfor control of T. granarium)

3-5 tablets or 15-25 pellets 57% a.i. t-1

for 7-10 d.

2-3 pellets 100 k g ' for 7-10 d.3 ml 100 kg-1

12 ml 100 k g '

lsifiable concentrate; s.c. = suspension concentrate; w.p. = wettable powder.2. If other formulations are used with a.i. concentrations differing from those

must be changed accordingly.given above, the amount of whole product used

groundnuts onto a smooth surface, then addingthe insecticide and thoroughly mixing the twowith a shovel or similar implement.

Kernel storage. The decision on when to shellgroundnut stocks is often based on factors otherthan good storage practice, e.g., the economicsof transporting a crop destined for export.Groundnuts destined for confectionary use orfor seed are often shelled soon after harvest sothat imperfect or damaged kernels can be dis-carded. This increases their susceptibility toattack by a number of insect pests. The directapplication of insecticides to shelled groundnutsis not recommended as this can result in thepresence of unacceptably high levels of toxicresidues. However, kernels in sacks can be pro-tected in the same way as nonshelled stocks byapplying one of the recommended insecticides asa spray or dust, to the outside surface of thesacks.

Control of established infestations

If stocks are already infested when placed instorage then insects may be present in the centerof a heap or stack before they are detected bytrapping or inspection. When this occurs, theonly effective method of disinfestation is by fum-igation. This involves the introduction of aninsecticidal gas into a gas-tight space around theinfested produce. The chemicals used in theseoperations are volatile and highly toxic to man.It is essential that they are used correctly andthat those performing the fumigation are prop-erly trained, and aware of the hazards involved.

Large-scale storage (>1t) . Some storagestructures can be effectively sealed to preventleakage of gas during fumigation. The entirecontents of the store can therefore be fumigatedat one time. In the majority of cases, however,the infested material must be covered with a gas-proof sheet (PVC, at least 0.25-mm thick; orpolythene, at least 0.13-mm thick). Where morethan one sheet is needed to cover a single stack,the sheets should be joined by tightly roll ing

together at least 0.5 m of overlap. The edges ofthe sheeting round the base should be weighteddown with sand bags to prevent gas escaping.

The most suitable fumigants for the treatmentof large, bagged or bulk consignments aremethyl bromide and phosphine (Table 1).Methyl bromide is usually applied from largecylinders under high pressure, and thereforerequires specialized equipment. The recom-mended dosage is 10-15 g m~3 for 24 h. In orderto effectively control T. granarium, this doseshould be increased to 15-25 gm - 3 .

There are drawbacks in using methyl bromideto protect stored groundnuts. Inorganic bro-mide is retained in the oil of groundnut kernels,so treated stocks must be promptly ventilated oncompletion of the operation, in order to mini-mize residues. Repeated application of methylbromide can reduce the germination potential ofgroundnuts kept for seed, particularly if theirmoisture content is high.

Phosphine is usually formulated as tablets orpellets of aluminium phosphide that releasephosphine gas when in contact with moisture inthe air. Ideally, to achieve good control in bulkstores, phosphine should be added to thegroundnuts as the store is filled at a rate of 3-5tablets or 25 pellets t-1. However, this is safe onlyif the operation is completed within 1-2 h ofexposure of the first phosphine tablets. If this isnot possible, tablets can be scattered over thesurface or introduced into the bulk produceusing a special applicator. Similarly, with sackstorage, tablets can be added to each layer ofsacks as they are being stacked, provided theoperation is completed within 1-2 h. Alter-natively, the tablets can be placed around thebase of the stack, or squeezed between sacks atthe sides and top of the stack.

After a fumigation period of 5-10 d the prod-uce should be aired thoroughly and if possiblethe spent fumigant should be removed. This ismore easily done if the tablets are enclosed inpaper envelopes when they are placed inposition.

Small-scale storage (< l t ) . Phosphine canalso be used to treat small quantities of infested

23

groundnuts, if these can be placed in gas-tightcontainers. One pellet (0.6 g) placed on top ofgroundnuts in polythene-lined sacks gave satis-factory control of insects infesting confectionarygrade kernels awaiting export (Proctor and Ash-man, 1972). Used oil drums, sealed with alumi-nium tape or strips of polyurethane foam, couldalso be used for this purpose.

One drawback to using phosphine for small-scale operations is that once the sealed containerin which tablets or pellets are supplied is opened,all the tablets have to be used immediately. Ifnot, the remaining tablets may start to decom-pose, giving off phosphine gas.

Liquid fumigants such as ethylene dibromide(EDB), carbon disulphide, and carbon tetra-chloride (CTC) can be used instead of phosphineto fumigate small quantities of produce. In WestAfrica, CTC has been recommended for single-sack fumigations and is marketed for that pur-pose in 10 ml plastic ampules. However, CTC ismore commonly used in admixture with otherfumigants to increase their penetration of thecommodity. In India, a 1:8 mixture of EDB:CTC (12 mL 100 kg-1) has been recommendedfor fumigating a wide range of commodities(Webley and Harris, 1977).

Owing to sorption of chemicals onto the com-modity, fumigant dispersal can be a problem.Better results wil l be achieved if the l iquid isapplied at several points while the container isbeing f i l led. Great care should be exercisedwhen dispensing these chemicals, as they arebelieved to be carcinogenic. Their use has beenrestricted in certain western countries resultingin a reluctance to recommend their use on storedcommodities in the tropics.

If properly carried out, fumigation shouldgive complete control of those insects exposed tothe gas. However, it offers no residual protectionand reinfestation can rapidly occur.

Integrated pest management

The need for alternatives to chemical controlmethods for the protection of stored productshas been indicated in the past but is now seen as

increasingly urgent. In particular, the discoveryof insect resistance to methyl bromide and phos-phine (the two most common fumigants) hasintensified the pressure to minimize use of con-ventional insecticides against postharvest pests.This in turn has encouraged research into alter-native control techniques. These include:• control of temperature, humidity, and atmos-

pheric gases in storage facilities to create con-ditions unsuitable for insect development;

• admixture of abrasive materials such as finesand, kaolin, or wood-ash to protect grain infarmer-level storage;

• use of certain plant materials such as crushedneem seed, neem leaves, or neem oil whichhave an antifcedant or repellant effect on stor-age pests. Vegetable oils have also been usedto protect stored pulses against attack by bru-chid beetles;

• dissemination of insect pathogens of storedproduct moths, e.g., the bacterium Bacillusthuringiensis Berliner or nuclear polyhedrosisand granulosis viruses, either by direct appli-cation onto the stored commodity or byattracting insects to traps containing a sourceof the disease;

• control of pests by natural enemies. This mayrequire modifications both to conventionalcontrol procedures and to some aspects of thestorage environment, to favor indigenous pre-dators and parasites; and

• use of genotypes resistant to attack by themain postharvest pests.

The integration of biological, physical, andchemical control measures is still in its infancyand the relative cost of integrated pest-management programs cannot easily be pre-dicted at this stage of their development. There isan urgent need to establish the potential andpracticality of this approach in order to reducethe problems created by reliance on syntheticinsecticides.

Breeding for resistance to storage pests

Breeding groundnut cultivars resistant to post-harvest insect infestation is one possible way of

24

reducing losses during storage. It is a strategyparticularly well-suited to improving storage inthe SAT where the financial resources and tech-nical expertise required for the effective use ofchemical control methods are often lacking.Unfortunately, breeding for resistance to storagepests is often antagonistic to selection forincreased yield. New high-yielding varieties havefrequently proved to be more susceptible toinsect attack during storage than the indigenousgenotypes grown by the small-scale farmer. Withthe large number of new varieties now beingdeveloped it is important that there is some col-laboration between breeders and storage ento-mologists to ensure that their release wil l notexacerbate storage problems.

Basic procedures for resistance screening. Theassessment of the inherent susceptibility of cropsto attack by storage insects usually involves themeasurement of parameters such as length ofdevelopment period, mortality of juvenilestages, amount of food consumed, and oviposi-t ion rate. To ensure that any differences in thevalues of these parameters obtained in a screen-ing trial reflect genuine differences betweengenotypes, both the grain and the insects used inthe experiment should be as uniform as ispractical.

Before insects are placed on samples of thegenotypes to be examined, the grain should bepreconditioned to the experimental temperatureand humidity levels for a period of at least 2 weeks to ensure that the moisture contents of thedifferent genotypes are as similar as possible.The moisture content of a subsample of eachgenotype should be determined (as describedabove) at the start of the experiment. If mass lossis being assessed, then the moisture contentshould also be determined at the end of theexperiment so that loss can be expressed as drymass of grain.

Ideally the experiment should be performed ina room or chamber with controlled temperatureand humidity. If this is not possible then therange of temperature and humidity should berecorded. This is essential when reporting thelength of the insects' development period.

Cultures maintained to supply insects forscreening should be of relatively constant den-sity so that the size of adults, and therefore theirlikely egg-laying capacity is comparable. If thescreening technique involves placing adults onthe experimental material then they should be ofidentical age and reproductive status (mated orvirgin). Whatever stage of the life cycle is used(adult, larva, or egg) a range of insect populationdensities per unit of grain mass should be exam-ined in a preliminary experiment to determinethe density at which differences between geno-types are most obvious and to check for anyinteraction between density and genotype.

Suggested further studies

In a study of the development of pest popula-tions on stored groundnuts in a warehouse inAndhra Pradesh, India, ICRISAT entomolo-gists recorded serious losses by the groundnutbruchid, C. serratus. Reports of damage togroundnuts by this species had previously beenconfined to West Africa. The findings of theICRISAT study highlight the need for moreinformation on postharvest pests of groundnuts,particularly from groundnut-growing areasother than West Africa. These data wil l be ofincreased value if they include a measure of themass loss caused by insect pests obtained usingone of the accepted methods outlined in thisbulletin.

There is also a need for more accurate infor-mation on the effect of insect infestation on thequality of groundnut oil. This might eventuallyallow the determination of economic thresholdsfor the main pests attacking stored groundnutsin the SAT.

25

Bibliography

Publications listed here include those referencesnot given in full in the text, and papers selectedto provide further reading, and more in-depthtreatment of specific topics.

Adams, J.M. 1976. A guide to the objective andreliable estimation of food Josses in small scalefarmer storage. Tropical Stored Products Infor-mation 32:5-12.

Anon. 1970. Methyl bromide dosage schedules.Tropical Stored Products Information 20:27-30.

Anon . 1983. Traps for use in detecting insectinfestation in stored products. Storage TrainingNotes. Slough, UK: Tropical Development andResearch Institute. (Limited distribution.)

Ashman, F. 1973. Methods and techniques ofassessing quality in stored products. TropicalStored Products Information 25:33-35.

Banks, H.J. 1977. Distribution and establish-ment of Trogoderma granarium Everts (Coleop-tera:Dermestidae): climatic and other in f lu-ences. Journal of Stored Products Research13:183-202.

Bell, C.H. 1975. Effects of temperature andhumidity on development of four pyralid mothpests of stored products. Journal of StoredProducts Research 11:167-175.

Blatchford, S.M., and Hall, D.W. 1963.Methods of drying groundnuts. 1. Naturalmethods (literature survey). Tropical Science5(1): 6-33.

Surges, H.D. 1962. Studies on the dermestidbeetle Trogoderma granarium Everts. V. Reac-tions of diapause larvae to temperature. Bulletinof Entomological Research 53(1): 193-213.

Burges, H.D. 1963. Studies on the dermestidbeetle Trogoderma granarium Everts. V I . Fac-tors inducing diapause. Bulletin of Entomologi-cal Research 54(3): 571-587.

Conway, J.A. 1976. The significance of Elasmo-lomus sordidus (F.) (Hemiptera: Lygaeidae)attacking harvested groundnuts in The Gambia.Tropical Science 18(3): 187-190.

Conway, J.A. 1983. Notes on the biology andecology of the groundnut seed beetle (Caryedonserratus) (01.) under field conditions in Sene-gambia. Tropical Stored Products Information45:11-13.

Davey, P.M. 1958. The groundnut bruchid,Caryedon gonagra(¥.). Bulletin of Entomologi-cal Research 49(2): 385-404.

Davey, P.M., Hall, D.W., Coveney P.L., andRaymond, W.D. 1959. The effect of insect infes-tation on the quality of decorticated groundnutswith special reference to storage at low and highhumidities. Tropical Science 1:296-307.

Delbosc, G. 1966. Les parasites des stocks d'ara-chide. Oleagineux 21:293-296.

Dobie, P. 1984. Biological methods for inte-grated control of insects and mites in tropicalstored products. I. The use of resistant varieties.Tropical Stored Products Information 48:4-8.

Dobie, P. 1984. Biological methods for inte-grated control of insects and mites in tropicalstored products. V I . Integrated control: the roleof biological methods. Tropical Stored ProductsInformation 48:37-45.

Feakin, S.D.(ed) 1973. Pest control in ground-nuts. PANS Manual No. 2. London, UK: Centrefor Overseas Pest Research. 197 pp.

Freeman, P. 1980. Common insect pests ofstored products. British Museum (Natural His-tory) Economic Series No. 15. London, UK:British Museum (Natural History).

Gillier, P., and Bockelee-Morvan, A. 1979. Laprotection des stocks d'arachide contre lesinsectes. Oleagineux 34(3): 131-137.

Golob, P. 1976. Techniques for samplingbagged produce. Tropical Stored ProductsInformation 31:37-48.

26

Colon, P. 1981. A practical appraisal of on-farm storage losses and loss assessment methodsin Malawi. 2. The Lilongwe land developmentprogramme area. Tropical Stored ProductsInformation 41:5-12.

Green, A.A. 1959. The control of insects infest-ing groundnut after harvest in The Gambia. I. A study of the groundnut borer Caryedon gonagra(F.) under field conditions. Tropical Science1(3): 200-205.

Green, A.A. 1960. The control of insects infest-ing groundnuts after harvest in The Gambia. I I .Field trials on the control of the groundnutborer, Caryedon gonagra (F.). Tropical Science2(1): 44-54.

Haines, C.P. 1984. Biological methods for inte-grated control of insects and mites in tropicalstored products. I I I . The use of predators andparasites. Tropical Stored Products Informa-tion 48:17-25.

Harris, K.L., and Lindblad, C.J. (eds.) 1978.Postharvest grain loss assessment methods. St.Paul, Minnesota, USA: American Associationof Cereal Chemists. 193 pp.

Heseltine, H.K. 1973. A guide to fumigationwith phosphine in the tropics. Tropical StoredProducts Information 24:25-36.

Hodges, R.J. 1979. A review of the biology andcontrol of the rice moth Corcyra cephakmicaStainton (Lepidoptera: Galleriinae). TropicalProducts Institute G. Report 125. London, UK:Tropical Development and Research Institute.20 pp.

Hodges, R.J. 1984. Biological methods for inte-grated control of insects and mites in tropicalstored products. I I . The use of pheromones.Tropical Stored Products Information 48:9-17.

Hodges, R.J. 1984. Biological methods for inte-grated control of insects and mites in tropicalstored products. IV. The use of insect diseases.Tropical Stored Products Information 48:27-32.

Howe, R.W. 1956. The biology of the two com-mon storage species of Oryzaephilus (Coleo-ptera, Cucujidae). Annals of Applied Biology44(2): 341-355

Howe, R.W. 1956. The effect of temperatureand humidity on the rate of development andmortality of Tribolium castaneum (Herbst)(Coleoptera, Tenebrionidae). Annals of AppliedBiology 44(2): 356-368.

Howe, R.W. 1965. Losses caused by insects andmites in stored foods and feeding staffs. Nutr i -tion Abstracts and Reviews 35:285-293.

Loschiavo, S.R., and Atkinson, J.M. 1973. Animproved trap to detect beetles (Coleoptera) instored grain. Canadian Entomology 105:437-440.

Narain, P., and Khosla, R.K. 1984. Estimationof post-harvest foodgrain losses. Journal of theIndian Society of Agr icul tural Statistics36(1):127-142.

Pointel, J.G., Deuse, J.P.L. , and Hernandez, S.1979. Evaluation et eVolution de l'infestation destocks experimentaux d'arachides en coque auSe'negal par Caryedon gonagra F. (Coleoptera:Bruchidae). Agronomie Tropicale 34(2): 196-207.

Prevert, P.F. 1964. The distribution of insects instacks of bagged groundnuts in northern Nige-ria. Bulletin of Entomological Research 54:689-713.

Proctor, D.L. , and Ashman, F . 1972. The con-trol of insects in exported Zambian groundnutsusing phosphine and polyethylene lined sacks.Journal of Stored Products Research 8:127-137.

Proctor, D.L., and Rowley, J.Q. 1983. Thethousand grain mass (TGM) method: a basis forbetter assessment of weight losses in storedgrain. Tropical Stored Products Information45:19-23.

Redlinger, L.M. , and Davis, R . 1982. Insectcontrol in postharvest peanuts. Pages 520-571 inPeanut science and technology (Pattee, H.E.,

27

and Young, C.T., eds.). Yoakum, Texas, USA:American Peanut Research and EducationSociety, Inc..

Rees, D.P. 1985. Review of the response ofstored product insects to light of various wave-lengths, with particular reference to the designand use of light traps for population monitoring.Tropical Science 25(3): 197-215.

Smith, K.G. 1963. The study of an insect popu-lation living on bagged groundnuts stored insouthern Nigeria with particular reference to thebehaviour of Trogoderma granarium Everts(Col., Dermestidae). Journal of the West Af r i -can Science Association 8(1): 44-51.

Srivastava, A.S. 1970. Important insect pests ofstored oilseeds in India. International Pest Con-trol 12(3): 18-20.

Tropica] Development and Research Institute.1984. Insects and arachnids of tropical storedproducts, their biology and identification (atraining manual). Slough, U K . T D R I .

Tyler, P.S., and Boxall, R.A. 1984. Post-harvest loss reduction programmes: a decade ofactivities; what consequences? Tropical StoredProducts Information 50:4-13.

U.S. National Academy of Sciences. 1978. Post-harvest food losses in developing countries.Washington D.C.,USA: National Academy ofSciences. 207 pp.

Webley D.J., and Harris, A.H. 1977. A compar-ison of fumigants for in-bag fumigation. Tropi -cal Stored Products Information 33:9-18.

28

RA-00102

ICRISAT

International Crops Research Institute for the Semi-A rid Tropics

Patancheru, Andhra Pradesh 502 324, India

ISBN 92-9066-124-0 Printed at ICRISAT Center ICR 87-0006