Journal of
Entomology and Nematology

  • Abbreviation: J. Entomol. Nematol.
  • Language: English
  • ISSN: 2006-9855
  • DOI: 10.5897/JEN
  • Start Year: 2009
  • Published Articles: 135

Full Length Research Paper

Geomorphologic and allometric variations in the populations of adult cocoa moth, Ephestia cautella (Lepidoptera: Pyralidae) from Southwestern Nigeria

Oyedokun A. V.
  • Oyedokun A. V.
  • Entomology Section, Cocoa Research Institute of Nigeria, Km 14, Ibadan- Ijebu Ode Road P.M.B. 5244 Ibadan, Oyo State, Nigeria.
  • Google Scholar
Omoloye A. A.
  • Omoloye A. A.
  • Department of Crop Protection and Environmental Biology, University of Ibadan, Ibadan, Oyo State, Nigeria.
  • Google Scholar

  •  Received: 14 November 2018
  •  Accepted: 26 February 2019
  •  Published: 31 May 2019


Damage by the tropical warehouse moth, Ephestia cautella (Walker) to stored plant products especially dried cocoa beans is enormous. This is expressed in holing with frass and webbing of the beans leading to nutritional loss; turning the valuable commodity into powder and reducing the value in the export-import market. Effective management of the pest is contingent upon correct identification and bioecological expressions in relation with the host(s) and other environmental influences. Therefore, the morphological traits (phenotypic) variability between sexes and populations of E. cautella across different locations in Southwestern Nigeria were investigated with a view to identifying the population structure of E. cautella in the region. Thirty-two ecotypes of E. cautella (16 male; 16 female) populations (n = 10/ecotype) were evaluated for eight morphometric traits including forewings (length and width), hind-wings (length and width), body length, antenna length and abdominal dimension (length and width). Significant variations (P ≤ 0.05) occurred among the 32 E. cautella ecotypes for the eight measured traits with respect to sexes, locations and the interactions of the two sources of variation (sex by location). By magnitude, the females had significantly (P ≤ 0.05) higher mean for the fore and hind wings (length and width), body length and abdominal dimension; however, the antenna length was longer in males than in females’ samples. The overall mean Gower genetic distance for the 32 E. cautella ecotypes was 0.656, with the range of 0.273 to 0.968. At 0.1 level of similarity index, four clusters (I, II, III, IV) emerged; with the membership of 7, 8, 12, and 5 respectively. This study showed that four ecotypes of E. cautella exist in south-western Nigeria with sex by location assessment and this can be useful in control programmes of the pest.
Key words: morphological traits, population structure, variability, ecotypes, similarity index.


The tropical warehouse moth, Ephestia cautella (Walker), is a notorious pest of dried cocoa beans in storage where it causes great economic losses. The veracity in feeding habit of an insect pest  may  be  well  correlated  with  the innate damage potentials and ecotype’s body size variation of the pest. Insect growth in body size is influenced by the larval environment during developmental stages in terms of nutrient available to the developing larvae (Igushi 1998; Chapman et al., 2013). Knowledge of insect morphology is required in insect ontology and taxonomy to classify insects’ species (Zhou et al., 2006) as well as an impetus in control programmes.
Variations observed in the morphological traits of the sexes of the same insect species may equally create groupings with respect to locations which may give rise to ecotypes, geo-types, morphotypes, biotypes or evolving of new species entirely. An organism’s shape is defined by sizes of its body parts (traits) in relation to the size of the whole (body size) with scaling pattern relating trait size to body size and this is known as trait allometry (Bonduriansky and Day, 2003). Measurements of morphometric traits are collected from individual insect of the same species with similar age, different sexes and/or varied geo-ecological origin to characterize the precise relationship between the dimensions of each trait and individual variations in overall body size  (Cheverud, 1982; Klingenberg, 1996).
The varied morphological traits could serve as a tool for assessing ecological variations in insects of the same species which could lead to evolutionary emergence of new species, ecotypes or biotypes. Although size is regarded as a quantitative phenotypic trait; its variations which; oftentimes result from some complex interactions between genetic and environmental factors (Pavkovic-Lucic and Kekic 2013) could provide a good lead on the impacts of locations on an organism. Allometric variations within species could be determined to access intra species sexual isolation or interactions in insects, which can be categorized as static allometry (Gould, 1966; Gayon, 2000) particularly when they are from different geographical locations which could prove useful when incorporated into non-chemical control of such insect.
Knowledge of the morphology of E. cautella populations, the extant variability between its sexes across different ecologies and/or locations is important for a scientific assessment of the potential risks of its introduction into new areas via cocoa trade activities. The information on E. cautella allometry may form the basis of understanding the likelihood of actual damage through its feeding activities that could be caused by this pest in stored cocoa bean; which might be directly correlated with the body size of the insect. Therefore, in this study, the morphotypic traits of E. cautella species across the studied locations were investigated via intra-specific allometry approach. This is with a view to evaluating the phenotypic variations as well as the plasticity in E. cautella species in the Southwestern Nigeria (where about 80% of Nigerian cocoa is being produced) as a tool for assessing its generational success and fitness of the pest.



Insect culture
Samples of adult E. cautella (128 pairs) were collected from a survey of cocoa warehouses and stores in four Southwestern States of Nigeria. In each state, four locations were randomly selected based on available production data of high, medium and marginal production (Table 1) and were reared on cocoa bean samples from each location in the laboratory at ambient tropical conditions to obtain a stock culture per location. To obtain homogenous culture the method described by Oyedokun and Omoloye (2015) was adapted, in which adults (male and female) from each location were paired singly in mating and oviposition cages (18.5 cm top x 13.5 cm base x 19.5 cm height) diameter lined with black filter paper prior to introducing the adults to provide the contrast needed to collect the milky-white laid eggs. The eggs laid after 24 h were collected and placed in 90 mm diameter Petri dishes lined with black filter paper for eclosion. The lid of the Petri dishes was perforated to allow for aeration and was placed on a table with its legs dipped in water polluted with fresh engine oil so as to prevent ants from predating on the eggs. Ten day-old, freshly hatched first instar larvae of E. cautella were carefully introduced into each rearing cage (12.5 cm top x 11.5 cm base x 6.5 cm height) containing 300 g of dried cocoa beans fermented at 6 days in ten replicates per location. Prior to introduction of the day-old larvae onto the beans, the beans were sterilised in a Gallenkamp Oven at 40°C for 4 h to kill any insect eggs that might have been laid on the beans during outdoor drying of the cocoa beans. After cooling, the larvae were introduced into the cages containing sterilized cocoa beans using a soft, pointing-tip camel hair brush dipped in sterile water and adequately drained. The rearing cages (cut at the sides and covered with muslin cloth) containing the dried cocoa beans and day-old larvae were kept undisturbed and observed daily in the laboratory at the ambient tropical temperature (28 ± 2°C) and relative humidity (70 ± 5%) till adults started to emerge. 
Assessment of geomorphic variations in E. cautella population of from southwest Nigeria
Twenty pairs of E. cautella emergents taken from each culture were sexed and kept temporarily in separate sterile sample bottles containing 70% ethanol solution before being dissected in distilled water in the laboratory. To assess the morphometics vis-à-vis ecotypes, body parts were dissected and separated. The variations in general body morphology of E. cautella were assessed following the modified method described by Bernitez et al. (2011).
To measure the total body length (BL) for male and female per location,  ten  individuals  were   selected   per   sex,   mounted   on calibrated stage graticle (0.1 mm x 100) and were viewed and measured using a Celestron USB Microscope x 200 magnification. The body length was measured from head tip (frons) to abdomen end (cercus). Also, to measure the antennae length (from suture to the tip of flagellum), ten antennae comprising five from right and five from the left side of male and female E. cautella from each location were carefully dissected out and measured in the laboratory using calibrated stage graticle (0.1 mm x 100) and viewed with a Celestron USB Microscope x 200 magnification. The length and width were measured in five replicates and compared for variations with and between the sexes and locations following the modified method described by Bernitez et al. (2011). To assess the abdominal dimension (length-from the end of tergite to the tip of Cercus- and width- from tergum end to sternum base) per sex and location, 10 individuals were selected per sex and/or location, mounted on calibrated stage graticule (0.1 mm x 100) and were magnified using a Celestron USB Microscope x 200 magnification.
To assess variations in the wing dimension; forewings and hind-wings were spread out from the axillary end of the wings to the tip of the radius part for measurement. Four morphological components - forewing length (FWL), forewing width (FWW), hindwing length (HWL) and hindwing width (HWW) were measured on a stage graticule (0.1 mm x 100 calibration) using Celestron USB microscope x 200. Thereafter each measured part was mounted and examined for qualitative assessment of characters such as the colours, size and shape of wing scales on both wings.
Data analysis
Data collected were analyzed using Analysis of Variance (ANOVA) and significant means were separated by Least Significant Difference (LSD) at (P ≤ 0.001; P ≤ 0.01 and P ≤ 0.05) levels of significance and also allometric trait parameters were further analysed using Principal Component Analysis (PCA), WARD Dendogram/Inter-cluster variability of Statistical Analysis System, SAS-V9.2 (SAS Institute Inc., 2007).


The geographical locations of the surveyed Southwestern Nigeria states are shown on Plate 1. The eco-geographical positions of the surveyed areas ranged between Latitude 3.533°N for Eruwa, Oyo State to 5.416°N for Ise Ekiti, Ekiti State and Longitude 6.733°E for Ore, Ondo State to 7.883 for Iragbiji, Osun State (Table 1). All the surveyed locations were within derived savannah and derived forest agro-ecological zones of Nigeria. There were significant (P≤0.05) variations in BL and AL among the 32 E. cautella samples assessed for morphometric traits with respect to sexes, location (that is where they were obtained) and the interaction of the two sources of variation (Sex by Location) (Table 2). While the sexes and the location of E. cautella differed for all the eight traits, there were significant (P≤0.05) interactions between sex and location with respect to BL and AL; and (P≤0.001) for HWL ABDL and ABDW. Also, the coefficient of  variation  (3.51  to  11.89%)  for  all  the traits measured was low (Table 2), implying very low dispersion with each treatment assessed.
Significant (P<0.05) differences occurred with respect to the eight morphometric traits assessed in both male and female E. cautella (Table 3). By magnitude, the fore wing and the hind wing length and width, body length and abdominal length and width were significantly higher in the female ecotypes than in the male ecotypes. However, the antenna length was longer in male ecotypes than the female ecotypes of E. cautella (Table 3).
The mean values of the eight measured morphometric traits were compared across the 16 locations and means were separated using the Least Significant difference (LSD at P = 0.05) (Table 4). The Iragbiji ecotypes had the longest forewings while The Ile-Oluji ecotype had the highest morphometric value for the remaining seven measured traits. The extremely least morphometric and/or allometric features of E. cautella ecotypes was found in Iwo and Ido (FWL and HWL), CRIN and Ido (FWW), Adejare (HWW), Ise-Ekiti (BL), Iwo (AL), Aramoko-Ekiti (ABDL) and CRIN (ABDW) ecotypes.
In Table 5, the Eigen value and the variance proportion of the five principal components decreased from PC1 to PC5. While the five PC-axes explained 99% of the total variation within the thirty-two E. cautella ecotypes, the first three explained an approximate variance proportion of 90%. Across the five PC axes, the eigenvector of the abdominal length contributed significantly (≥0.20) and positively to the total variation. Positive and significant (≥0.20) contribution of the eigenvector of the abdominal width was in the first three PC axes. The eigenvector of the eight morphometric traits was positive and significant (≥0.20) in PC1. Within the axis, HWL and ABDL had the highest and the least (Table 5).
At the 0.1 similarity index (Figure 1), four clusters (I, II, III and IV) emerged; with the membership of 7, 8, 12 and 5 respectively. The ecotype membership in cluster I included six male and one female (from Aramoko Ekiti). Cluster II and III had mixtures of the two sexes of E. cautella from various locations. The five ecotypes in the fourth cluster were all females from Ondo, Ado-Ekiti, Ore, Idanre and Ile-Oluji. At the similarity index point of 0.2 to 0.37, two clusters emerged. From this point range, clusters I, II and III formed a single cluster while cluster IV remained unique. A single unit (cluster) representing the organism emerge at the similarity index point beyond 0.37 (Figure 1).
The discriminatory role of each of the eight morphometric traits in distinguishing among the cluster was presented in Table 6. The fore wing width and the antenna length significantly (P<0.001) differentiated the four clusters. Cluster IV significantly (P<0.001) differed from clusters I, II and III. None of the cluster is a duplicate of  the  other  one  with  respect  to  fore  wing  width  and antenna length. Fore and hind wing length and body length could not distinguish between Clusters I and III. The morphological features of fore and hind wing length and antenna length were similar for clusters I and III. Clusters II and III were significantly similar with respect to hind wing width and the length of the abdomen (Table 6). Each of the four clusters significantly held E. cautella ecotypes with specific distinguishing morphometric features.
Table 7 presents the intra-cluster variation. The longest antenna   was   among  the  seven  ecotypes  in  cluster I. However, the least abdominal length and width occurred in the cluster (Table 7). None of the eight morphological features differentiated among the seven ecotypes in clusters. Seven out of the eight features distinguished the 12 ecotypes in cluster III. Ecotypes in cluster IV had the highest mean for all the features except antenna length.  Cluster I was the most homogenous while cluster III was the most heterogeneous cluster in the E. cautella samples. The eight morphological features of the organism could not distinguish the seven E. cautella ecotype in cluster I.  Different  levels  of variability existed in clusters II, III and IV.


In   this   study,   variations   occurred   in  the  E. cautella species from across locations (Latitude 7°10' N to 8°00'N and Longitude 4°00'E to 5°25'E) in Southwestern Nigeria (Ondo, Osun, Ekiti and Oyo States) with respect to the measured traits. Significant variations in ecotypes and sexes of the test insect sample as revealed in this study corroborates  earlier studies  (Pavkovic-Lucic   and Kekic 2013) that environmental factors and diets contribute to the variations in insect morphological traits and these factors in turn determine the intraspecies morphological and/or allometric variations in insects. Expression of varied morphological traits in insects is influenced by a number of factors as evident in this study, ranging from genetic to temperature, humidity, altitude, larval density, quality and quantity of food available to the developmental stages of such insect and macroecological patterns (David et al., 1983; Hoffmann et al., 2005; Takahashi et al., 2011; Sanzana et al., 2013). Earlier report (Hoffmann et al., 2005) also supports this assertion especially in relation to wing morphology that environmental factors may alter the genes that should activate wing development in Drosophila melanogaster
Generally, from this study, female E. cautella samples were larger in size than the male species from all surveyed locations which may be due to inherent generational fitness of the sexes. This of course can also be influenced by the food source(s) on which they develop but not actually altering the gene coding for sex and morphological expressions in terms of size. Other factors like developmental temperature (though not evaluated in this study) might have been responsible for size variations among E. cautella ecotypes as obtained in the results of this  study  and  this  can  be  buttressed  by earlier work by Josh 2004 on Drosophila sp. that developmental temperature causes size variations that can be correlated with male and female fitness components. Although, little is known about genes responsible for differences in body size of insects (De Jong and Bochdanovits 2003), geographical distribution of insects with respect to climatic variation also influence body size (Gibert et al. 2004). This trend was evident in E. cautella samples collected and/or emerged from Ondo State (within Latitude 7°10'N; Longitude 5°15') with bigger allometric traits that favourably outcompete samples from Oyo State (Latitude 8°00'N; Longitude 4°00') that are relatively smaller in size. Similarly, E. cautella samples from locations like Iragbiji, Idanre and Ile-Oluji with mountainous geo-positioning had relatively higher allometric traits and this supports earlier work (Trotta et al., 2010) that larger individuals of insects (Drosophila sp.) occur at higher altitudes and latitudes. 
Expression of larger body in form of exaggerated and non-exaggerated traits by insects gives such insects advantages that correlate with major fitness components, greater flight ability, dispersal ability and higher mating success (Heed and Manger, 1986; Joshi, 2004; Frazier et al., 2008; Vishalakshi and Singh, 2008). In the same vein, this study showed that size is a quantitative phenotypic trait and  its  effects  on  the  generational  success  of  E.
cautella ecotypes in Southwestern may depend largely on bio-ecological factors prevailing in each location surveyed. Generally, female species are with larger sizes and this indicates likelihood of higher succession of the species in the surveyed locations because of the potential ability to lay higher number of fertilized eggs to continue the generational success of the insect pest.
This study showed variations in allometric values of the body parts of E. cautella assessed which culminated in the ecotypes identification in Southwestern Nigeria; indicating   that    slight    changes   in   the   microclimatic components of locations could have significant effects on the biotic components of the ecology especially in insects.


There are four ecotypes of E. cautella in Southwestern Nigeria (where about 80% of Nigerian cocoa is produced) as identified in the studied states with regards to distribution of the insect pest and cocoa beans production output of the locations. For instance, Ile-Oluji, Ondo, Ore and Idanre axes that are endemic for high cocoa production in Southwestern Nigeria had the highest morphometric values and this trend goes along the production potentials of the locations in the cocoa agroecology in Southwestern Nigeria. The geomorphic and allometric variations in this insect species as assessed in this study will serve as vital information in applying control option in the insect’s management before, during or after the export of cocoa beans from Nigeria.


The authors wish to acknowledge the assistance of Mr. Kehinde Oyeledun and Mrs. Elizabeth Onifade of the Entomology Section, Cocoa Research Institute of Nigeria during the laboratory culture of homogenous population of E. cautella. Also, Dr. (Mrs.) Feyisara Okelana- the immediate past Acting Executive Director of Cocoa Research Institute of Nigeria- is appreciated for her moral support and advice in the course of this study.


The authors have not declared any conflict of interests.



Benítez HA, Briones R, Jerez V (2011). Intra and inter-population morphological variation of shape and size of the Chilean magnificent


beetle, Ceroglossus chilensis in the Baker River Basin, Chilean Patagonia. Journal of Insect Science 11(1):1-9.


Chapman RF, Simpson SJ, Douglas AE (2013). The Insects: Structure and Function, 5th edition. Cambridge University Press, Cambridge, U.K. Available at: 



Cheverud JM (1982). Relationships among ontogenetic, static, and evolutionary allometry. American Journal of Physical Anthropology 59(2):139-149.


David JR, Allemand R, Van Herewege J, Cohet Y (1983). Ecophysiology: abiotic factors. In: Ashburner M, Carson HL, Thompson JN (eds.), The Genetics and Biology of Drosophila, Vol. 3d. Academic Press, London, UK pp. 105-170.


De Jong G, Bochdanovits Z (2003). Latitudinal clines in Drosophila melanogaster: Body size, allozyme frequencies, inversion frequencies, and the insulin-signalling pathway. Journal of Genetics 82(3):207-223.


Frazier MR, Harrison JF, Kirkton SD, Roberts SP (2008). Cold rearing improves cold-flight performance in Drosophila via changes in wing morphology. Journal of Experimental Biology 211(13):2116-2122.


Gayon J (2000). History of the concept of allometry. American Zoology 40:748-758.


Gibert P, Capy P, Imasheva A, Moreteau B, Morin JP, Pétavy G, David JR (2004). Comparative Analysis of morphological traits among Drosophila melanogaster and Drosophila simulans: Genetic variability, clines and phenotypic plasticity. Genetica 120:165-179.


Gould SJ (1966). Allometry and size in ontogeny and phylogeny. Biological Review 41:587-640.


Heed WB, Mangan RL (1986). Community ecology in Sonoran Desert Drosophila. In Ashburner M, Carson HL, Thompson JN (eds.), The Genetics and Biology of Drosophila. Academic Press, New York. pp. 311-345.


Hoffmann AA, Hallas R, Anderson A, Telonis-Scott M (2005). Evidence for robust sex-specific trade-off between cold-resistant and starvation resistance in Drosophila melanogaster. Journal of Experimental Biology 18:804-810.


Igushi Y (1998). Horn Dimorphism of Allomyrina dicotoma-septentrionalis (Coleoptera: Scarabaeidae) as affected by larval nutrition. Annals of Entomological Society of America 91:847-92.


Joshi A (2004): Are bigger flies always better: the role of genes and environment. Journal of Genetics 83:13-15.


Klingenberg CP (1996). Multivariate Allometry. In: Advances in Morphometrics Marcus LF (ed.) New York. pp. 23-49.


Oyedokun AV, Omoloye AA (2015). Developmental response of tropi cal warehouse moth, Ephestia cautella (Walker)(Lepidoptera: Pyralidae)'s larvae to stored cocoa beans fermented at varied degrees. African Journal of Biotechnology 14(16):1364-1372.


Pavkovic-Lucic S, Kekic V (2013). Developmental temperature, body size and male mating success in fruit flies Drosophila melanogaster (Diptera: Drosophilidae). European Journal of Entomology 110(1):31.


SAS Institute Inc., (2007). SAS Integration Technologies Client 9.2. Available at: 



Sanzana MJ, Parra LE, Sepulveda E, Benitez HA (2013). Latitudinal gradient effect on the wing geometry of Auca coctei (Guérin)(Lepidoptera, Nymphalidae). Revista Brasileira de Entomologia 57(4):411-416.


Trotta V, Pertoldi C, Rudoy A, Manenti T, Cavicchi S, Guerra D (2010). Thermal plasticity of wing size and shape in Drosophila melanogaster, Drosophila simulans and their hybrids. Journal of Climate Research 43(1-2):71-79.


Vishalakshi C, Singh BN (2008). Mating Success is not correlated with fluctuating asymmetry in Drosophila ananassae. Current Science 94:375-381.


Zhou M, Geng G, Huang S (2006). Ontology development for insect morphology and taxonomy system. Proceedings of the 2006 IEEE/WIC/ACM International Conference, Hong Kong. Available at: