An analysis of the feeding behavior of three stages of Toxoptera citricida by DC electrical penetration graph waveforms

With the purpose of studying the feeding behavior of the brown citrus aphid pest, Toxoptera citricida (Kirkaldy) (Hemiptera: Aphididae), we compared stylet probing behaviors of third and fourth instars and adults on Citrus unshiu Marc (Rutaceae) seedlings using the electrical penetration graph (EPG) technique. EPG waveforms exhibited the full suite of stylet behaviors – stylet pathway, intracellular stylet puncture, phloem salivation (E1), sieve ingestion (E2), and xylem sap ingestion activities, plus the non‐penetration (Np) waveform. Before the phloem phase, the number of probes was significantly higher for third‐instar nymphs than for adults. Overall duration of Np events by adults was significantly lower than the duration of third and fourth instars. The number of short probes of the fourth instars was significantly higher than that of the adults. In the phloem phase, adults made more frequent and longer E1 events than the third and fourth instars. Third instars made more frequent but shorter E2 events, whereas adults made fewer but longer events. These results showed adults gained nutrients by increasing feeding time during phloem ingestion. Thus, the probability of phloem‐associated virus acquisition and transmission of T. citricida was higher in adults than in nymphs.     

Xylem ingestion by winged aphids

When aphids and their host plant are incorporated in a DC electrical circuit, phloem and xylem ingestion register as separate waveforms of the electrical penetration graph (EPG) signal. Aphids are primarily phloem feeders; xylem ingestion is seldom reported but can be induced experimentally by fasting the insects in desiccating conditions. In experiments with the black bean aphid, Aphis fabae Scop., young winged (alate) and unwinged (apterous) virginoparous adults were collected from their natal host plants (broad bean, Vicia faba L.) and allowed 3-h continuous EPG-recorded access to V. faba seedlings. Several aphids (47% of both morphs) showed ingestion from phloem sieve elements. Alate aphids also showed frequent xylem ingestion (60% of individuals), but no apterous aphids exhibited this activity. The EPG technique involves attachment of a fine gold wire electrode to each insect, a process that may affect normal behaviour at the plant surface. However, when the technique was modified to monitor the stylet activities of freely-settled aphids, high levels of xylem ingestion by alates were also recorded. The results suggest that the developmental physiology of winged aphids somehow predisposes them to xylem ingestion, possibly as a result of dehydration during the teneral period. Alate aphids may reduce their weight by fasting before take-off, giving aerodynamic benefits, but making rehydration, via xylem uptake, a priority following plant contact.     

Electrical penetration graphs from Cicadulina mbila on maize, the fine structure of its stylet pathways and consequences for virus transmission efficiency

Five distinct electrical penetration graph waveforms characterising the feeding behaviour of the leafhopper Cicadulina mbila Naudé (Homoptera: Cicadellidae) on maize (Zea mays L.) were obtained using a DC based system. The waveforms were distinguished by spectral features and by statistical analysis of their median voltages, durations and time to first waveform recording. By changing the polarity of the system voltage and the level of the input resistor it was shown that the waveforms are mainly determined by the electromotive force (emf) component. Based on the correlation between waveforms and the fine structure of the stylet pathways observed by transmission electron microscopy, insect's activities have been associated with five waveforms: stylet pathway formation (waveform 1), active ingestion (waveform 2), putative stylet work (waveform 3), salivation (waveform 4) and passive ingestion (waveform 5). Like waveform E1 and E2 of aphids, waveforms 4 and 5 of C. mbila correspond to feeding activities in sieve tubes. However, unlike aphids which probe briefly in non-vascular cells, waveform 2 corresponds to active ingestion in cells, where the cell content is partially ingested and hence the organelles' integrity severely affected. These observations suggest that this specific feeding feature, typical of leafhoppers, determines their ability to acquire geminivirus virions located in the plant cell nucleus.     

New structure in cell puncture activities by aphid stylets: a dual‐mode EPG study

Intracellular punctures by aphid stylets appear as potential drop (pd) waveforms in DC electrical penetration graph (EPG) recordings. We used a dual‐EPG device that recorded in one channel the ‘full EPG’ with R‐plus emf‐components (i.e., the usual DC EPG) and concurrently in a second channel the ‘R‐EPG’ with R‐components only. The circuit of the latter channel was an optimised amplitude modulation (AM) version derived from early (before 1990) AC systems. We also made some ‘emf‐EPG’ recordings using a separate high input resistance ‘emf‐amplifier’ sensitive to emf‐components only. The intracellular pd waveforms have previously been divided into three subphases, and we aimed to distinguish and separate these subphases more accurately by the dual‐EPG recordings than with the normal full EPG only. In this study, we temporarily distinguished five subphases (α–ε), but unequivocal distinction of only a few of these appeared possible, in spite of the information coming from the two signals. The lack of clearly separable features in R‐EPG signals often provided serious difficulties in pd recognition without the concurrent full EPG, but once located, only subphase II‐2 features were clear and supported the II‐2 data from the full EPG. Consequently, we could not distinguish subphases of complete pd waveforms better with additional R‐EPG information during cell punctures by Aphis gossypii Glover (Hemiptera: Aphididae). In Brevicoryne brassicae (L.) (Hemiptera: Aphididae), however, distinguishing II‐2 subphases in the full EPG was sometimes a problem. Our detailed dual‐EPG observations showed some waveform continuity from halfway into the II‐1 subphase (start of the newly recognised subphase β) until the end of the pd, with a strong but variable emf origin. This waveform tended to overrule other subphase waveforms in B. brassicae more than in A. gossypii and Myzus persicae (Sulzer) (Hemiptera: Aphididae). Subphase waveforms in full EPGs were especially difficult to recognise when pd periods had been interrupted in a virus inoculation experiment and additional R‐EPG information could then be useful. This inoculation experiment showed again that only the first subphase (II‐1) contributes to virus (Cucumber mosaic virus) inoculation by A. gossypii. In B. brassicae, the benefit of concurrent R‐EPG information in such virus experiments is presently under further investigation. Apart from this special application to virus experiments, we do not recommend the routine use of the dual‐EPG device. Furthermore, we do not advocate the distinction of more than the previously recognised three intracellular pd subphases as a feasible option in future studies. Analysis of EPGs with concurrent R‐EPGs requires substantially more analysis work without yielding consistently useful additional insights. This confirms earlier dual‐EPG results from thrips.     

Specificity of Agrobacterium-mediated delivery of maize streak virus DNA to members of the Gramineae

Parameters affecting the efficiency of agroinfection of maize streak virus (MSV) in maize have been determined. Monomeric units, cloned at a number of sites in the MSV genome were not infectious but multimeric units containing partial duplications were equally as infectious as complete tandem dimeric clones. Inoculation of tandem dimeric units conjugated into different strains of Agrobacterium showed that both A. tumefaciens and A. rhizogenes were able to transfer DNA to maize and this ability was Ti (or Ri) plasmid-specific. Nopaline strains of A. tumefaciens and both agropine and mannopine A. rhizogenes strains efficiently transferred MSV DNA to maize. A number of strains were capable of MSV DNA transfer to other members of the Gramineae, providing information which may be essential for Agrobacterium-mediated transformation of monocotyledonous plants.     

The influence of ant-attendance on aphid behaviour investigated with the electrical penetration graph technique

For the mutualistic interaction between the aphid Metopeurum fuscoviride Stroyan (Homoptera: Aphididae) and the ant Lasius niger L. (Hymenoptera: Formicidae) it has been shown that ant-tended aphids develop faster, reproduce at a higher rate, and live longer than aphids not tended by ants. We used electrical penetration graphs (EPG) to investigate if behavioural patterns differ between ant-tended and untended M. fuscoviride during 8 h experiments. Measurements were made on adult aphids from four different ant-tended colonies that continued to be tended by L. niger during the experiments, and from four different colonies where ant workers were excluded several days before the start of the experiment and that were also not tended by ants during the experiments. Ants readily tended wired aphids and ant tending did not interfere with the EPG measurements. There were no significant differences in the duration of sieve element penetration or in any other analysed feeding-related EPG parameters between ant-tended and untended individuals. However, the quality of the EPG recordings did not allow the distinction between the EPG-waveform E1 (salivation only) and E2 (salivation and ingestion). These results suggest that the changes in life-history traits of ant-tended aphids do not result from changes in time of sieve element penetration waveforms. Alternative mechanisms may involve an increase in the rate of sap uptake or a higher effectiveness in nutrient uptake in the presence of ants. Our study demonstrates that the EPG technique is a useful tool to investigate the feeding behaviour of aphids during interactions with ants.     

Leafhoppers on leaves: An analysis of feeding behavior using conditional probabilities

Feeding behavior of four deltocephaline leafhoppers,Graminella nigrifrons, G. oquaka, Amblysellus grex, andDalbulus maidis on maize and johnsongrass was analyzed using an electronic monitoring device. Five distinct waveform patterns were identified: secretion of sheath saliva (salivation), nonvascular probing, nonsieve element ingestion, x-waveform, and phloem ingestion. Waveforms were associated with feeding activities by correlation with light microscopic examination of salivary sheath termination points in leaf tissue and analysis of honeydew excreted by monitored leafhoppers. In previous studies x-waveforms have been reported to occur only when the stylets of homopterans are in contact with the phloem; the function of x-waveforms is poorly understood. There were no differences in time spent salivating or ingesting from nonsieve elements amongG. nigrifrons, G. oquaka andA. grex on either plant.D. maidis differed from other species in phloem probing and feeding behavior; only a small proportion produced x-waveforms, although those that did spent significantly more time in this behavior than other species. Also,D. maidis spent more time than other leafhoppers ingesting from tissues other than sieve elements. Kinetic diagrams of transition probabilities show that probing activities of all species were not random regarding the sequence of behaviors culminating in phloem ingestion. Thirty-five percent ofG. nigrifrons x-waveforms were followed by nonsieve element ingestion. This was consistent with observations showing that salivary sheaths of leafhoppers producing x-waveforms sometimes do not terminate in the phloem, but rather in nearby cells. Phloem ingestion was always preceeded by x-waveforms. The quantitative differences in probing behavior are discussed in relation to ability of these leafhoppers to transmit the phloem-associated maize chlorotic dwarf waikavirus.     

Aphids secrete watery saliva into plant tissues from the onset of stylet penetration

Aphid feeding requires the secretion of two types of saliva: gelling saliva (from the principal gland) that forms an intercellular sheath for the penetrating stylet, and watery saliva [from accessory salivary glands (ASGs)] that facilitates intracellular penetration and phloem feeding. Plant viruses can be used as salivary markers to investigate key steps in aphid feeding, and penetration can be monitored electrically using the electrical penetration graph (EPG) approach. We conducted a series of EPG‐controlled transmission experiments using Cucurbit aphid‐borne yellows virus [CABYV; Polerovirus spec. (Luteoviridae)], which is retained in the ASGs, as a marker for watery saliva secretions. The melon aphid, Aphis gossypii Glover (Hemiptera: Aphididae), was used as a vector and melon seedlings, Cucumis melo L. (Cucurbitaceae), as host plants. Viruliferous aphids were interrupted at various stages during stylet penetration, i.e., during intercellular penetration prior to intracellular puncture and following a potential drop within the first probe. Viruliferous aphids and leaf disc samples obtained from the stylet penetration site were used to detect CABYV by quantitative real‐time RT‐PCR. Approximately half of the inoculated leaf discs were found to be infected with CABYV after very brief (12.9 ± 1.9 s) intercellular stylet probes and before intracellular stylet puncture. The number of virus particles ejected during such probes was similar to the number ejected by aphids during longer probes including a single intracellular puncture. Our results therefore suggest that watery saliva is secreted by aphids from the onset of stylet penetration.     

Behavioral Comparisons of Ingestion and Excretion by Selected Species of Pentatomids: Evidence of Feeding on Different Food Sources Supports Pest Status

Laboratory studies were conducted on food ingestion and excretion by selected species of pentatomids on different food sources to support their pest status. We compared the frequency and time of feeding on vegetative (stem) and reproductive (seed) structures of soybean, Glycine max (L.) Merrill and of maize, Zea mays L. by Piezodorus guildinii (Westwood), Dichelops melacanthus (Dallas), and Edessa meditabunda (F.); in addition, the amount of excreta (feces) produced were compared for D. melacanthus feeding on seed of soybean and stem of maize seedling. The feeding behavior of E. meditabunda and P. guildinii on soybean, and of D. melacanthus on maize was recorded using the electropenetrography (EPG) technique. Excretion was estimated using water sensitive paper recording number and area of fecal drops. Results indicated that E. meditabunda on soybean stem repeated events of ingestion (both xylem and phloem sap) over four times per bug during the 8 h of recording for ca. 53 min per event. Dichelops melacanthus on maize seedling repeated each ingestion event over three times per bug for ca. 24 min per event. Piezodorus guildinii feeding on soybean stem repeated each ingestion 1.2 times per bug for ca. 40 min per event; on seed endosperm, it fed for a longer time, ca. 80 min per event, each event repeated only 0.5 times per bug. Number of excretory drops was higher (9.9 drops per bug) when D. melacanthus fed on maize seedling than on soybean seed (1.4 drops per bug). A larger amount of saliva/regurgitate liquid food was expelled when bugs fed on the former than on the later food.     

Electrical recording of plant penetration by western flower thrips

Plant penetration by western flower thrips (Frankliniella occidentalis (Pergande)) was analysed with the electrical penetration graph technique (EPG, DC-system). Thrips attached to a gold wire were included in an electrical circuit to record EPGs when penetrating the plant tissues with their stylets. Three basic EPG waveforms have been distinguished, correlated with stylet penetration into cells, salivation, and ingestion, respectively. The main difference with EPGs of Homoptera is the occurrence of continued separate penetrations that are not necessarily followed by ingestion. Insertion of the stylets causes strong voltage fluctuations in the EPG. We could confirm earlier evidence that penetration of cells and subsequent ingestion of (part of) the protoplast takes less than 20 seconds. Repeated short penetrations can be followed by a continuous feeding pattern during which the stylets are not withdrawn. The same sequence of waveforms is produced on other plant parts such as fruits or pollen grains. The specific waveforms are mainly caused by electromotive force (emf). The emf component was recorded with high resolution and the correlation of waveform details with activities of the cibarial muscle system is discussed.     

Host plant dispersion, leaf hopper movement and disease transmission

Abstract.

• 1 The plant-to-plant movement of the corn leafhopper, Dalbulus maidis Delong & Wolcott, and the spread of the leafhopper-borne maize rayado fino virus were investigated in four patterns of maize (Zea mays) dispersion.
• 2 D. maidis was less abundant and the spread of the virus was slower in dense stands of maize than in sparse stands.
• 3 When plant density was held constant, leafhoppers were more abundant in maize stands with relatively equidistant plant spacing (uniform dispersion) than in stands with densely-sown rows (linear dispersion) or double-sown hills (clumped dispersion), but there was no difference in virus incidence among these plant dispersion patterns.
• 4 Leafhoppers were less likely to move to adjacent plants in uniform plant dispersion patterns than in either linear or clumped dispersion patterns. This result may explain the lack of higher virus incidence in uniform stands, despite higher leafhopper abundance.
• 5 Leafhopper movement was consistent with a simple rule: the shorter the distance to the next adjacent plant, the more likely a leafhopper is to move between plants.
• 6 These results demonstrate that host plant dispersion can affect the abundance and behaviour of highly mobile herbivorous insects even when plant density is constant.

     

Effect of Manipulations on the Host Selection Behavior of Sitobion avenae (Homoptera: Aphididae)

The study of aphid host selection and feeding behavior is difficult because aphids have to penetrate the plant to reach their feeding site, phloem tissue. The activity of the stylets, salivation or food intake, can not be observed externally and requires an indirect visualization technique such as the Electric Penetration Graph (EPG). The plant selection behavior of Sitobion avenae on potato varied depending on whether an ethological or EPG method was used to study it. A similar variation did not occur with Myzus persicae or Rhopalosiphum padi. The application of water-based silver conductive paint onto the thorax, as normally used for EPG, or onto the abdomen of Sitobion avenae alates resulted in increased duration and frequency of probing compared to results from ethological observations. Our results indicated that EPG manipulations might have different effects on different species of aphids and that a comparison of EPG and ethological data is required to confirm that the EPG method does not bias aphid feeding behavior.     

Systemic antifeedant effects of azadirachtin on the peach-potato aphidMyzus persicae

The electrical penetration graph (EPG) method was used to analyse the feeding behaviour of apterous, adultMyzus persicae (Sulzer) (Homoptera: Aphididae) onNicotiana clevelandii (Gray) seedlings, treated systemically with azadirachtin. A preliminary experiment showed that the effects of tethering aphids for EPG recording were minimal. The percentage of the 9 h recording period devoted to non-penetration activities, and to stylet pathway patterns increased as the azadirachtin concentration in the root treatment increased. The number of probes initiated, and the numbers of sieve tube penetrations also increased with increased azadirachtin concentration. The mean time elapsing between the initiation of the first probe to reach a sieve element and contact with this tissue was not significantly altered by azadirachtin treatment. However, azadirachtin treatment significantly reduced the percentage of probes that reached sieve elements and increased non-penetration activity before and after the first perid of ingestion from the sieve elements. The percentage of the recording period spent in the EPG pattern associated with sieve tube penetration was significantly reduced by an azadirachtin concentration of 300 ppm, and the duration of each individual penetration was significantly reduced by an azadirachtin concentration of 100 ppm. When the total EPG was split into 3 h periods, significant interactions were seen between time period and azadirachtin concentration for the duration of non-penetration, pathway, and sieve tube penetration patterns.     

The role of sinigrin in host plant recognition by aphids during initial plant penetration

Plant penetration behaviour (probing) of the cabbage aphid, Brevicoryne brassicae, and the pea aphid, Acyrthosiphon pisum, was studied on excised leaves of broad beans, Vicia faba, kept in water or in a 1% aqueous solution of sinigrin. Using the DC EPG (Electrical Penetration Graph) technique it was shown that the cabbage aphid on sinigrin-untreated bean leaves showed numerous short probes into epidermis and mesophyll. None of these aphids showed either phloem salivation or ingestion waveforms on untreated leaves. In contrast, on sinigrin-treated bean leaves, 35% of the probing time was spent on phloem sap ingestion (E2) and almost all aphids reached phloem vessels and started feeding. The duration of phloem salivation before phloem ingestion and the mean duration of phloem ingestion periods were similar on a host and a sinigrin-treated non-host plant. However, the total probing time by B. brassicae was 10% longer, the total phloem sap ingestion time was twice as long, and the time to the first phloem phase within a probe was three times shorter on the host plant compared to sinigrin-treated broad beans. Acyrthosiphon pisum also responded to the addition of sinigrin to broad beans, but in this case sinigrin acted as a deterrent. On sinigrin-treated leaves, A. pisum terminated probes before ingestion from phloem vessels, and none of these aphids showed phloem salivation and ingestion on treated leaves. Glucosinolates were detected in the mesophyll cells of the brassicaceous plant, Sinapis alba. Based on this finding and in addition to the foregoing EPG analysis of aphid probing on these plants and broad beans, our hypothesis is that aphids may recognise their host plants as soon as they probe the mesophyll tissue and before they start ingestion from phloem vessels.     

Characterization of the feeding behavior of three Erythroneura species on grapevine by histological and DC‐electrical penetration graph techniques

Feeding behavior of three leafhopper species – Erythroneura vitis (Harris), Erythroneura ziczac (Walsh), and Erythroneura elegantula (Say) (Hemiptera: Cicadellidae) – reared on grapevine, Vitis vinifera L. cv. ‘Seyval blanc’ (Vitaceae), was investigated using histological techniques and DC‐electrical penetration graphs (DC‐EPG). Histological studies revealed that the Erythroneura species induced white stipples on the leaves and that these leafhoppers produced thin salivary sheaths in grapevine leaf tissues. The DC‐EPG system allowed the characterization of five waveforms associated with stylet penetration and feeding in leaf tissues. These waveforms were characteristic of feeding phases corresponding to epidermis penetration pathway, salivation, and ingestion. We calculated 28 parameters (e.g., number of probes, duration of phases, and time spent in the various tissues) to describe and compare the feeding behavior of the Erythroneura species. We conclude that the three Erythroneura species are mainly mesophyll feeders but may probably also feed in other tissues such as xylem.     

Aphid activities during sieve element punctures

Aphid salivation in sieve elements and phloem sap ingestion were linked to waveforms in the Electrical Penetration Graph (EPG). Non-viruliferousRhopalosiphum padi (L.) (Hemiptera, Aphididae) on barley yellow dwarf virus (BYDV) infected wheat could acquire the virus, which was used as an indication for phloem sap ingestion, whereas virus inoculation by viruliferous aphids on healthy plants was associated with salivation in sieve elements or other phloem cells. Probing was monitored and the waveforms recorded were related to ELISA results of test plants. The EPG patterns A, B, and C are indicative of the stylet pathway phase, whereas patterns E1 and E2 reflect the phloem (sieve element) phase with an unknown activity (E1) or with ingestion and concurrent salivation (E2). Aphids showing pathway and E1 rarely acquired virus, suggesting that little or no phloem sap ingestion can occur during these patterns, whereas those showing additionally pattern E2 did so substantially, indicating phloem sap ingestion. The main pattern related to virus inoculation was E1, although some aphids were able to inoculate plants during pathway. Pattern E1 clearly reflects the most important salivation into sieve elements. Pattern E2 had no clear contribution to virus inoculation, supporting the present hypothesis that during this pattern the saliva is mixed with the phloem sap in the single canal at the stylet tips and ingested immediately, without reaching the plant tissue. Sustained sap ingestion did not affect virus inoculation. So, BYDV inoculation mainly occurs during the first period of a sieve element puncture which is always formed by E1. Implications on persistent virus transmission are discussed.     

Irreversible thelytokous reproduction in Muscidifurax uniraptor

Three encyrtid parasitoids Apoanagyrus (Epidinocarsis) diversicornis, Aenasius vexans, and Acerophagus coccois (Hymenoptera: Encyrtidae) are used to control the cassava mealybug Phenacoccus herreni Cox & Williams (Sternorrhyncha: Pseudococcidae), an important pest of cassava in South America. The influence of parasitism on the feeding behaviour of mealybugs was studied by observing honeydew secretion and by the electrical penetration graph technique (EPG, DC-system). Honeydew secretions were observed after parasitism until mummy transformation. No strong EPG parameter differences were found between parasitised and control insects. All results indicated that parasitised mealybugs keep feeding on the phloem sap after parasitism until mummy transformation. The main influence of parasitism on EPG parameters is the appearance of a new pattern resembling the E2 pattern at the extracellular level and labelled H. This pattern was also produced with control insects located on an unfavourable feeding site and could be associated with a stress response. It might be related to the still unclear E(c) pattern of aphids. The relationship of H to stylet activities is discussed.     

Plant penetration activities by the flatid planthopper Metcalfa pruinosa (Hemiptera: Fulgoroidea): an electrical penetration graph‐histology analysis

The citrus flatid planthopper, Metcalfa pruinosa Say (Hemiptera: Flatidae), is a very polyphagous native insect in North America and currently a serious pest in Europe and South Korea. To understand the feeding behaviour of M. pruinosa, stylet penetration behaviour of M. pruinosa was investigated with an electrical penetration graph (EPG) system. This study reports seven EPG waveforms related to M. pruinosa feeding behaviour: np (no stylet penetration), Mp1 (initiation of stylet penetration), Mp2 (stylet movement and salivation), Mp4 (phloem feeding), Mp4‐H (honeydew excretion), Mp5 (xylem feeding) and Mp6 (unknown). To determine respective feeding behaviour related to the Mp4 and Mp5 waveforms, stylets were cut with a laser beam, and the location of the stylet tip within plant tissue was examined. We found plant sap was exuded from the severed stylets only when the Mp4 waveform was observed, suggesting phloem sap ingestion. The stylet tip was located in the xylem region, indicating xylem‐feeding activity, when the Mp5 waveform was observed. The analysis of 24 different EPG parameters suggests that M. pruinosa stylets reached the vascular bundle of a plant within ca. 5 min and spend ca. 70% of the time feeding on xylem and phloem feeding. This is the first study that reports seven distinctive EPG waveforms with respect to the feeding behaviour of M. pruinosa which could help determine host specificity and host plant susceptibility.     

Waveform characterization of the soybean stem feeder Edessa meditabunda: overcoming the challenge of wiring pentatomids for EPG

Stink bugs (Hemiptera: Heteroptera: Pentatomidae) are in general robust and restless insects, which makes them difficult to wire for electropenetrograph (EPG) studies. In addition, cuticular lipids may reduce wire effectiveness, and their removal could improve success of wiring. We compared wiring effectiveness for three species of stink bugs, differing in walking behaviour and degree of cuticular waxiness, that is, Piezodorus guildinii (Westwood), Nezara viridula (L.), and Loxa deducta (Walker). Results indicated that removal of cuticular lipids by mechanical abrasion (via sanding) greatly improved attachment success with gold wire. Our hypothesis that heavier and bigger bugs would lose the wire attachment more quickly than lighter and smaller bugs was not confirmed, regardless of the sanding. In contrast, our hypothesis that greater movement of a bug would cause the wire to break more often was supported by extensive testing. Behaviour appears to be more relevant for successful wiring than body weight. We used the sanding and wiring technique to characterize and correlate direct current EPG waveforms for the large and restless stem‐feeding stink bug Edessa meditabunda (Fabricius) on soybean plants. This marks the first published example of pentatomid EPG waveforms. Edessa meditabunda recordings on soybean stems generated eight types of waveforms in three phases and two families, named as follows: non‐probing = Np and Z; pathway phase = Em1; X wave phase = X; ingestion phase, family I = Em2 and Em3; ingestion phase, family N = Em4 and Em5. These eight were described based on their frequencies, relative amplitudes, and level voltages. Histological studies of stylets within salivary sheaths correlated the Em1, Em2, and Em3 waveforms with specific penetration sites. The waveform with the longest duration when feeding was Em2, representing xylem sap ingestion; in addition, waveform Em3 (always preceded by an X wave) was correlated with phloem sap ingestion.