The effect of parasitoid host-size preference on host population growth rates: an example of Aphidius colemani and Aphis glycines

Abstract.  1. Direct observations of Aphidius colemani foraging for Aphis glycines demonstrated that A. colemani attacks large aphid size classes selectively, in contrast to other Aphidiinae (Hymenoptera: Braconidae) species that generally prefer small or medium-sized nymphs of different host species.
2. To determine the effect of this size preference on the potential ability of A. colemani to control A. glycines outbreaks, the stage-dependent survival and fecundity of A. glycines were measured to parameterise a stage-structured demographic model for the density-independent population growth rate of the aphid.
3. Compared with hypothetical parasitoids that show either no size preference or preference for medium-sized host nymphs, the preference of A. colemani for large hosts caused a greater reduction in the population growth rate of A. glycines . This occurs in the model because, by attacking reproductive adults, A. colemani kills those aphids that have the greatest immediate effect on the population growth rate.
4. The strong effect of size preference by A. colemani is diminished somewhat by the continued reproduction of A. glycines adults, which can reproduce for up to 3 days following parasitism. Nonetheless, reproduction by parasitised aphids is not enough to compensate for the stronger, negative effect of the preference of A. colemani for large aphids, which removes individuals with the greatest reproductive value from the A. glycines population.     

Biology and host selection behaviour of the aphid hyperparasitoid Alloxysta victrix in association with the primary parasitoid Aphidius colemani and the host aphid Myzus persicae

The biology of the aphid hyperparasitoid Alloxysta victrix was investigated with respect to its interaction with the primary parasitoid Aphidius colemani and the host aphid Myzus persicae. Laboratory investigations of host selection behaviour indicated that naive A. victrix females could rapidly distinguish between unparasitized aphids and those parasitized by A. colemani, but showed little tendency to discriminate between aphids parasitized by the primary parasitoid three, five, seven, or ten days earlier (even though the latter were already mummified). Host suitability studies, in contrast, indicated that aphids parasitized by A. colemani became progressively less suitable as the primary parasitoid developed. When aphids were exposed to A. victrix more than seven days after being parasitized by A. colemani, survivorship of both primary and secondary parasitoid species decreased, with a significantly higher proportion of mummies producing no viable adult parasitoid of either species. The developmental period (egg to adult eclosion) of surviving hyperparasitoids was also longer in older host stages, and females developed significantly more slowly than males (14.2 days versus 13.6 days, respectively). The sex ratio of hyperparasitoid broods emerging from all host stages was consistently male-biased (31–46% female), but adult longevity was higher for females than for males (37.1 days versus 23.6 days).     

Host-attacking behavior of a eulophid parasite,Kratochviliana sp., to the leaf mining host,Phytomyza ranunculi (Diptera: Agromyzidae) during its stay on the leaf

The present paper studies how the female parasite of Kratochviliana sp. visits and attacks its host larvae of Ranunculus leaf mining fly, P. ranunculi at a single leaf visit. The parasite visited its hosts at random on the leaf. The frequency of host visits was independent of the host density and the proportion of hosts survived from the parasite attack, in a leaf and its distribution was expressed as a single straight line. It almost always attacked living hosts at the first host visit after isolated from them for one day but with the rate of about 0.5 at the subsequent visits. In consequence, the relationships of the number of host attacks and killed hosts to the host density drew satulated curves in each. A model of host attack by this parasite at its single leaf visit was formulated by modifyingBakker et al.'s model (1972) basing upon these observations and the attack avoidance by the parasite to already attacked hosts previously reported.     

Impact of the entomopathogenic fungus Verticillium lecanii on development of an aphid parasitoid, Aphidius colemani

The parasitoid Aphidius colemani developed normally (approximately 90% adult emergence) when its cotton aphid (Aphis gossypii) host was treated with Verticillum lecanii conidia 5 or 7 days after parasitization. Fungus exposure 1 day before or up to 3 days after parasitization, however, reduced A. colemani emergence from 0 to 10%. Also, numbers of spores and mycelial fragments in aphid homogenates were much higher in aphids exposed to the fungus up to 3 days after parasitization than in aphids treated after 5 or 7 days. Our results suggest that the parasitoid and fungus may be used together for aphid biocontrol as long as fungus applications are timed to allow late-instar development of the parasitoid.     

Time allocation of the parasitoidAphidius colemani (Hymenoptera: Aphidiidae) foraging forAphis gossypii (Homoptera: Aphidae) on cucumber leaves

The time allocation of individualAphidius colemani female parasitoids foraging forAphis gossypii nymphs on cucumber leaves has been investigated. Apart from experiences on the current leaf (such as density of hosts on the current leaf, density of hosts on a neighboring leaf and encounters with hosts on the current leaf), the effect of previous leaf visits on the time allocation was studied. Behavioral records were analyzed by means of the proportional hazards model, to determine the tendency of leaving the current leaf. The leaving tendency decreased only on leaves with a high host density (100 aphids), thus increasing the giving up time since the latest encounter. Rejection of aphids had no influence on the leaving tendency. To assess the effect of the number of hosts encountered on the leaving tendency, we considered three classes: 0–30 encounters, 31–100 encounters, and 100 or more encounters with hosts. The effect of the number of hosts encountered differed at different aphid densities. When fewer than 10 aphids were present the leaving tendency was much greater after 30 encounters than beforehand. At a density of 100 aphids the leaving tendency was lower than at the other aphid densities and increased only after 100 encounters. The density of hosts on a neighboring leaf, ranging from 0 to 100 hosts, had a negligible effect on the leaving tendency. Repeated visits to leaves with 10 unparasitized aphids resulted in an increase in the leaving tendency after 10 visits. It is argued that the parasitoids have some innate expectancy of host availability and that they concentrate on high-density patches.     

Direct and Indirect Impacts of Infestation of Tomato Plant by Myzus persicae (Hemiptera: Aphididae) on Bemisia tabaci (Hemiptera: Aleyrodidae)

The impacts of infestation by the green peach aphid (Myzus persicae) on sweetpotato whitefly (Bemisia tabaci) settling on tomato were determined in seven separate experiments with whole plants and with detached leaves through manipulation of four factors: durations of aphid infestation, density of aphids, intervals between aphid removal after different durations of infestation and the time of whitefly release, and leaf positions on the plants. The results demonstrated that B. tabaci preferred to settle on the plant leaves that had not been infested by aphids when they had a choice. The plant leaves on which aphids were still present (direct effect) had fewer whiteflies than those previously infested by aphids (indirect effect). The whiteflies were able to settle on the plant which aphids had previously infested, and also could settle on leaves with aphids if no uninfested plants were available. Tests of direct factors revealed that duration of aphid infestation had a stronger effect on whitefly landing preference than aphid density; whitefly preference was the least when 20 aphids fed on the leaves for 72 h. Tests of indirect effects revealed that the major factor that affected whitefly preference for a host plant was the interval between the time of aphid removal after infestation and the time of whitefly release. The importance of the four factors that affected the induced plant defense against whiteflies can be arranged in the following order: time intervals between aphid removal and whitefly release > durations of aphid infestation > density of aphids > leaf positions on the plants. In conclusion, the density of aphid infestation and time for which they were feeding influenced the production of induced compounds by tomatoes, the whitefly responses to the plants, and reduced interspecific competition.     

Initial floral visitor identity and foraging time strongly influence blueberry reproductive success

Priority effects occur when the order of species arrival affects subsequent ecological processes. The order that pollinator species visit flowers may affect pollination through a priority effect, whereby the first visitor reduces or modifies the contribution of subsequent visits. We observed floral visitation to blueberry flowers from honeybees, stingless bees or a mixture of both species and investigated how (i) initial visits differed in duration to later visits; and (ii) how visit sequences from different pollinator taxa influenced fruit weight. Stingless bees visited blueberry flowers for significantly longer than honeybees and maintained their floral visit duration, irrespective of the number of preceding visits. In contrast, honeybee visit duration declined significantly with an increasing number of preceding visits. Fruit weight was positively associated with longer floral visit duration by honeybees but not from stingless bee or mixed species visitation. Fruit from mixed species visits were heavier overall than single species visits, because of a strong priority effect. An initial visit by a stingless bee fully pollinated the flower, limiting the pollination contribution of future visitors. However, after an initial honeybee visit, flowers were not fully pollinated and additional visitation had an additive effect upon fruit weight. Blueberries from flowers visited first by stingless bees were 60% heavier than those visited first by honeybees when total floral visitation was short (∼1 min). However, when total visitation time was long (∼ 8 min), blueberry fruit were 24% heavier when initial visits were from honeybees. Our findings highlight that the initial floral visit can have a disproportionate effect on pollination outcomes. Considering priority effects alongside traditional measures of pollinator effectiveness will provide a greater mechanistic understanding of how pollinator communities influence plant reproductive success.     

Temperature influences the handling efficiency of an aphid parasitoid through body size-mediated effects

It is well known that increasing the ambient temperature increases the metabolic rate and consequently, the foraging rate of most insects. However, temperature experienced during the immature stages of insects affects their adult size (an inverse relationship). Because body size is generally correlated to foraging success, we hypothesized that temperature indirectly influences the foraging efficiency of adult insects through developmental effects. We first investigated the role of parasitoid: host body size ratio on the handling time of Aphidius colemani (Viereck) (Hymenoptera: Braconidae), then tested the prediction that increasing temperature during immature development increases the handling time of adults. As expected, parasitoids took longer to handle large aphids than small aphids. However, large parasitoids did not have shorter handling times than small parasitoids except when attacking large (adult) aphids. Developmental temperature had the predicted effect on parasitoids: Individuals reared at 25°C were smaller than those insects reared at 15°C. Parasitoids reared at 15°C had similar short handling times for both first instar and adult aphids, whereas parasitoids reared at 25°C took longer to handle adult aphids than first instar aphids. The size-mediated effect of temperature through development on parasitoid efficiency was opposite to the more familiar direct effect of temperature through metabolic rate. We conclude that the net effect of temperature on foraging insects will depend on its relative influence on immature and adult stages.     

Consumption of snowdrop lectin (<Emphasis Type="Italic">Galanthus nivalis </Emphasis>agglutinin) causes direct effects on adult parasitic wasps

Honeydew is a common sugar-rich excretion of aphids and other phloem-feeding insects and represents the primary sugar in many agricultural systems. When honeydew-producing insects feed on genetically modified plants, the honeydew can contain amounts of the transgene product. Here we address whether this route of exposure poses a risk for non-target insects. Three species of parasitic wasps were selected: i.e. Aphidius colemani, Trichogramma brassicae and Cotesia glomerata, all of which are known to use honeydew as a carbohydrate source in the field. Wasps were fed sucrose solutions with varying concentrations of snowdrop lectin ( Galanthus nivalisagglutinin, GNA), a protein that has been engineered into crops to confer resistance against homopteran pests and that has been detected in honeydew. Parameters evaluated included gustatory response, longevity, fecundity, progeny emergence and sex ratio. While A. colemani and T. brassicae, but not C. glomerata, were able to detect GNA, this gustatory recognition had no effect on the acceptance of a GNA-sucrose solution. In all three species, GNA ingestion reduced parasitoid survival significantly. However, in respect to fecundity, negative effects were observed for T. brassicae but not for A. colemani. The results suggest that the effect of GNA consumption may depend on the specifics of a parasitoid's biology, especially its longevity and its mode of egg maturation.     

Host instar preference and functional response of Aphidius transcaspicus, a parasitoid of mealy aphids (Hyalopterus species)

Aphidius transcaspicus Telenga (Hymenoptera: Braconidae: Aphidiinae) has been selected for use as a biological control agent against Hyalopterus pruni Geoffroy (Hemiptera: Aphididae) in prune orchards in California. In this study we focus on the host instar preference and functional response of A. transcaspicus. Host instar preference was determined using laboratory arenas by allowing female parasitoids to attack aphids on field-collected leaves of Prunus domestica L. supporting a natural range of H. pruni instars. A relative preference index indicated that A. transcaspicus was 1.4 times more likely to attack intermediate instars (2–4) than 1st or adult instars. The functional response of A. transcaspicus to H. pruni density was determined on enclosed P. domestica shoots in a prune orchard. The functional response to host density was linear (Type I) at the lowest aphid densities, and the asymptotic number of aphids attacked at higher aphid densities was 64.3?±?6.7 aphids day^?1.     

Foraging behavior of the coccinellid Nephus includens (Coleoptera: Coccinellidae) in response to Aphis gossypii (Hemiptera: Aphididae) with particular emphasis on larval parasitism

This study assessed the effect of parasitism of Nephus includens (Col.: Coccinellidae) larvae by Homalotylus flaminius (Hym.: Encyrtidae) on the predation rates of the predator on the cotton aphid, Aphis gossypii Glover (Hem.: Aphididae) by deriving functional responses for second- and fourth-instar predators at prey densities ranging from 10 to 80 aphids per arena. The relationship between the functional and numerical responses of adult females of N. includens also was determined for prey densities ranging from 10 to 140 aphids per arena. Predation rates of unparasitized and parasitized second-instar N. includens were both fit by a type II functional response model with parameters as follows: unparasitized (a = 0.0768 hours(-1) and T(h) = 0.975 h) and parasitized (a = 0.0787 hours(-1) and T(h) = 0.8823 hours). Predation rates of unparasitized and parasitized fourth-instar N. includens were fit by type III and II models, respectively, with the following parameters: unparasitized (b = 0.1702 hours(-1) and T(h) = 0.2369 hours) and parasitized (a = 0.038 h(-1) and T(h) = 0.539 h). The unparasitized fourth-instar was the most voracious stage, having the highest attack rate and lowest handling time. Considering these attributes, it would seem to be the most effective stage of this predator against A. gossypii. Adult female lady beetles (N. includens) showed a type III functional response and their numerical response increase to prey density was curvilinearly related to prey density, with the highest number of eggs being produced at highest prey densities. The maximum saturation level for both predation and egg production for adult females occurred at a prey density of 120 aphids. Thus, a ratio 1:120 (predator:prey) should be used when releasing this species for augmentative biological control. Release of either fourth-instar or adult stage N. includens should be minimized the potentially negative effect of parasitism by H. flaminius on early developmental stages, and hence increase its efficiency in biocontrol programs.     

Dissecting components of population-level variation in seed production and the evolution of masting behavior

Theory predicts that at higher trophic levels species are especially vulnerable to habitat fragmentation due to small population size and dependence on particular prey species. Using aphids as model organism, we tested the hypothesis that herbivore abundance increases in fragmented habitats as a result of reduced predator and parasitoid densities. In a 3 year-study, we examined the population dynamics of aphids with respect to host plant abundance and ant nest density in experimentally fragmented calcareous grasslands at two sites in the northern Jura mountains. Fragments of different size (area: 20.25 m², 2.25 m² and 0.25 m²) were isolated by a 5 m wide strip of frequently mown vegetation and corresponding control plots were situated in the adjacent undisturbed grassland. Aphid density was higher in fragments than in control plots. This was a combined result of a higher frequency of aphid-infested plants and larger aphid colonies in fragments than in control plots. Furthermore, a larger proportion of aphid colonies was ant-attended in fragments than in control plots. Aphid colonies were also more frequently visited by ants in fragments than in control plots in one of the 3 years. Parasitoid pressure on aphids was not influenced by the experimental fragmentation. Neither were aphid species richness and diversity affected by the fragmentation. Our study shows that even small-scale habitat fragmentation can have profound effects on the abundance of herbivorous insects. The effect on aphid density was consistent over 3 years and two sites with slightly different aphid communities.     

Strategies for the Control of Aphis gossypii Glover (Hom.: Aphididae) with Aphidius colemani Viereck (Hym.: Braconidae) in Protected Cucumbers

Aphis gossypii Glover (Hom.: Aphididae) is a damaging pest of protected cucumbers in the UK, and control measures are required which are compatible with other components in the overall cucumber integrated pest management programme. Two methods of establishing the parasitoids, Aphidius colemani Viereck (Hym.: Braconidae), in cucumber crops prior to invasion of the aphids were developed. The first involved weekly releases of small numbers of parasitoids beginning before A. gossypii became established on the plants. The second method used open-rearing units based on maize, wheat and rygrass plants infested with Rhopalosiphum padi L (Hom.: Aphididae), which is a common host to many natural enemies of A. gossypii but not a threat to the cucumber crop. Both methods were found to be more efficient in summer than in late spring. Parasitoid release rates were established for the two control methods at both of these times of the year. Like all biological control measures, these methods will require careful management in practice and some fine-tuning to suit individual crop production systems.     

Behavioural evidence for detection of the repellent polygodial by aphid antennal tip sensilla

Abstract. Close-up video was used to record responses of the aphid Myzus persicae (Sulzer) (Homoptera: Aphididae) to the plant-derived repellent polygodial. When 0.1% (±)-polygodial was painted onto one half of a Chinese cabbage leaf disc, and the other half treated with the ethanol solvent, apterous adult aphids avoided walking onto the polygodial-treated area during 15 min access periods. Similarly, when the rim of a small plastic Petri dish was treated with polygodial, aphids took significantly longer to walk off the base of the dish and onto the rim, compared with insects in solvent-treated control dishes. Analysis of video-recordings indicated that aphids were repelled following contact of antennal tips with the polygodial-treated areas of leaf or plastic. In contrast, aphids with both antennal tips surgically removed showed no apparent response to polygodial treatment of leaf or plastic surfaces. Aphids were placed between polygodial- and solvent-treated angled glass cover slips to enable unilateral presentation of polygodial on either the left or right side of the insects. Most aphids with antennae intact moved away from the polygodial-treated side, but removal of both antennal tips again negated the response. When the tip was removed from only one antenna, and the other remained intact, unilateral presentation caused a repellent response only if polygodial occurred on the same side as the intact antenna. Presentation of polygodial on both sides of aphids (bilaterally) caused no response if both antennal tips were intact or removed, but aphids with only one antennal tip removed were repelled from the side where the antenna remained intact. The results indicate that aphids detect polygodial upon contact with sensilla located on the antennal tips, and that the repellent response occurs as a negative chemotropotaxis following unilateral stimulation of these sensilla. Contact between the antennal tips and glass substrates was demonstrated using a fluorescent pigment.     

Effects of the antifeedant polygodial on plant penetration by aphids, assessed by video and electrical recording

Adult apterousMyzus persicae (Sulz.) discriminated within 2 min between mature Chinese cabbage (Brassica pekinensis L.) leaf halves treated with (±)-polygodial solution and solvent alone, and walked off leaf areas treated with polygodial faster than off the solvent-treated areas. However, when aphids were attached to a fine gold wire and stylet penetration of cabbage leaves was recorded electrically, polygodial treatment did not affect the number or duration of electrically-recorded penetrations, time taken to initiate a first penetration, or total penetration time. As the tethering of aphids causes behavioural restrictions which may negate the response to polygodial, indications of stylet penetration by freely-moving insects were sought. Simultaneous electrical recording and video monitoring showed that stylet penetration duration could be accurately inferred from antennal and body movements, enabling assessment of penetration times without attachment to the wire tether. When freely-movingM. persicae were video recorded during 15 min access to cabbage seedlings, polygodial treatment again had no apparent effect on stylet penetration. However, when aphids were presented with a choice of polygodial- and solvent-treated sides of floating mature cabbage leaf discs, video recordings revealed that the insects spent less time and made fewer penetrations on the polygodial-treated side. In addition to this rapid repellent effect, prolonged exposure to polygodial also produced behavioural changes. After being held for 24 h on polygodial-treated leaves or green paper prior to behavioural examination, aphids penetrated seedlings fewer times but for longer periods. The relevance of the results to virus transmission studies is discussed.     

Role of xylem consumption on osmoregulation in Macrosiphum euphorbiae (Thomas)

Aphids are phloem feeders that occasionally ingest xylem sap. The duration of xylem consumption by Macrosiphum euphorbiae (Hemiptera: Aphididae) was positively correlated with the level of dehydration of alate aphids of different ages after a period of starvation, supporting the hypothesis that aphids ingest xylem sap to replenish their water balance. However, the duration of xylem sap ingestion but not phloem sap consumption varied in unstarved alate adults of different ages. Furthermore, both alate and apterous aphids ingested xylem sap at the end of their life, when aphids were not dehydrated but when fecundity started to decrease. Fecundity was negatively correlated with the proportion of time spent ingesting xylem sap, and that over the entire reproductive life of alate and apterous aphids. The lower proportion of xylem ingested by apterous than by alate aphids during the first few days of adult life may be related to a higher symbiont density in apterous morphs. As previous studies have demonstrated a relationship between sucrose assimilation, which is directly influenced by fecundity and symbiont density, and osmoregulation, we suggest that xylem consumption may play a role in the osmoregulation of haemolymph of aphids.     

Development and functional response of Coelophora inaequalis (Coleoptera: Coccinellidae) feeding on brown citrus aphid, Toxoptera citricida (Homoptera: Aphididae)

1 Development and survivorship of Coelophora inaequalis (F.) were evaluated in the laboratory on the brown citrus aphid, Toxoptera citricida (Kirkaldy) at 20 and 25 °C. Coelophora inaequalis could complete its life cycle feeding on T. citricida at these two temperatures. The developmental period from egg to adult was significantly longer at 20 °C (24.7 d) than at 25 °C (15.9 d). The survivorship of combined immature stages were higher at 25 °C (39.2%) than at 20 °C (34.3%). 2 The effects of prey densities (4–64 aphids per leaf disk) on the functional responses of C. inaequalis fourth‐instar larvae and female adults were investigated using grapefruit leaf disks over a 24‐h period at 25 °C and a photoperiod of LD 14 : 10 h. Logistic regression analyses indicated that fourth‐instar larvae and adults exhibited a type II functional response to aphid density. Fourth‐instar larvae had a higher search rate (0.1305) than C. inaequalis female adults (0.0989). The handling time of fourth‐instar larvae (30.4 min) was significantly longer than that of female adults (5.8 min). Collectively, C. inaequalis appears to be a promising biological control agent of T. citricida in the citrus groves.     

Host instar and host plant effects on Aphidius colemani

Abstract:  The effect of Myzus persicae age (I–IV instar nymphs) on parasitism by Aphidus colemani was investigated in choice experiments with either aubergine ( Solanum melongena L.) or pepper ( Capsicum annuum L.) as the aphid host plant species. Aphidius colemani could develop successfully to the mummy stage in all ages of aphid tested, on both pepper and aubergine plants. The greatest proportion of parasitoid mummies was found with II and III aphid instars on pepper and on I, II and III aphid instars on aubergine. On aubergine, I instar nymphs were also parasitized to an appreciable degree. Successful parasitism of IV instar nymphs was consistently low on both crops. The potential implications of host plant effects on the effectiveness of A. colemani as a biological control agent are discussed.     

A functional response evaluation of pre-infestation with <Emphasis Type="Italic">Bemisia tabaci</Emphasis> cryptic species MEAM1 on predation by <Emphasis Type="Italic">Propylea japonica</Emphasis> of <Emphasis Type="Italic">Myzus persicae</Emphasis> on host plant tomatoes

Herbivore feeding on host plants may induce defense responses of the plant which influence other herbivores and interacting species in the vicinity, such as natural enemies. The present work evaluated the impact of pre-infestation with the tobacco whitefly Bemisia tabaci cryptic species MEAM 1, on the predation ability of the ladybird Propylea japonica, to the green peach aphid Myzus persicae, on tomato plants. The results show that B. tabaci pre-infestation density, duration, and leaf position, can impact prey consumed by P. japonica under various aphid densities. The aphids consumed by P. japonica in each treatment were fit using the Holling type II functional response equation. The predatory efficiency (a/T _h) of P. japonica was the highest in the treatment with 60 aphids and 48-h infestation directly on damaged leaves. The predatory efficiencies of P. japonica decreased with a reduction of pre-infestation density and duration. We also observed that pre-infestation on young and undamaged leaves increased predation by P. japonica.     

Variability and Parasitoid Foraging Efficiency: A Case Study of Pea Aphids and Aphidius ervi

When a parasitoid is searching for hosts, not all hosts are equally likely to be attacked. This variability in attack probability may affect the parasitoid functional response. Using a collection of experiments, we quantified the functional response of Aphidius ervi (Hymenoptera: Braconidae), an insect parasitoid of the pea aphid Acyrthosiphon pisum (Homoptera: Aphididae). We measured variability in the number of hosts attacked by a foraging parasitoid both among plants and within plants. At the first scale, A. ervi, searching among plants containing different numbers of aphids, showed both aphid-density-dependent and aphid-density-independent variability in the number of aphids attacked per plant. Within plants, A. ervi selectively attacked second and third instar aphids relative to other instars. Furthermore, there was variability in the susceptibility of attack among aphids independent of instar. Variability in attack rates among aphids both among and within plants decreased parasitoid foraging efficiency, with the greatest decrease caused by among-plant variability. Furthermore, the decrease in foraging efficiency was greatest when the average number of aphids per plant was low, thereby transforming a strong Type II functional response into one approaching Type I.