Host plant adaptation in the glasshouse whitefly.

<strong>Introduction</strong>Although much research towards the development and understanding of plant resistance and biological control as pest control strategies has been done, few studies have concentrated upon the interaction between these two control methods where they are combined. However, ecological research has provided a firm framework in which biological control and plant resistance can be studied within a tritrophic context. Such a framework emphasises how the changes in the response of a pest to its host plant, may consequently affect the dynamics of the interaction between it and its natural enemy. Conversely, how the use of a natural enemy to maintain pest density to below a certain level, may affect the interaction between the pest and its host plant. Within this framework, the pest's potential to adapt to the host plant, needs to be considered. Where adaptation occurs, the resistance of that host plant to the pest diminishes and also the efficacy of the biological control agent (s) used to control that pest may be affected. These tritrophic systems, should be studied from the viewpoint of evolutionary ecology. This stresses the adaptive character of the interactions, more than a pure ecological approach, which until now has provided the basis for studying the population dynamics of pest control systems.In relation to this tritrophic approach to pest control, the system Encarsia formosa (parasitoid) - Trialeurodes vaporariorum (pest; common name, glasshouse whitefly) - host plant was taken as a model system. In this thesis, aspects of the bi-trophic interaction between whitefly and its host plants are described. This system is a subject of continuing studies at the Department of Entomology in Wageningen. Here, research has concentrated upon preference and performance aspects of whitefly- plant relationships (van Lenteren and Noldus, 1990), the parasitization behaviour and parasitizing efficiency of E. formosa (Noldus and van Lenteren, 1990), the improved resistance of host plants to whitefly (de Ponti et. al., 1990) and the reduction of incompatibility problems of host plant resistance with respect to the deployment of E. formosa (van Lenteren and de Ponti, 1991). These studies have provided a fundamental basis for the improvement of whitefly control in commercial glasshouses by E. formosa (van Lenteren and Woets, 1988). Earlier studies e.g. van Boxtel et. al., 1978, had indicated the potential of whitefly to adapt from one host plant to another, within its host range. However the rate at which adaptation would occur and whether or not genetically distinct races of the glasshouse whitefly exist in relation to this process was not clear from such studies. The ability of whitefly to adapt rapidly to new host plants, or the existence of distinct races of whitefly in relation to this process may have repercussions upon the efficacy of whitefly control by E. formosa .<strong>Research questions and experiments</strong>The studies described in this thesis, aimed to answer the following questions:1 At what rate is a whitefly likely to adapt, when transferred from one host plant to another (chapter 2) ?2 Do 'trade-offs~ occur during adaptation with respect to the host plant from which whitefly originated (chapter 2) ?3 Can the rearing of whitefly upon a certain host plant, 'pre-condition' them to adapt to another host plant (chapter 3) ?4 Does the host plant from which whitefly originates influence its preference for other host plants (chapter 4) ?5 Do genetically distinct races of whitefly exist (chapter 5) ? The whitefly's rate of host plant adaptation was assessed by measuring changes in life history parameters over consecutive whitefly generations. The relative contributions of genotypic and phenotypic variation to the adaptation process were not assessed. This was a deliberate decision, because a simple measurement of changes in ecological parameters over consecutive whitefly generations would give an overall indication of intra and intergenerational variability, that could subsequently be used to calculate the degree of replication required, to accurately ascertain each component of genetic variation. Egg laying, immature mortality and immature development time, were assessed. These are all indicative of population fitness with respect to the net reproductive rate of the population.The major objective was to develop a method for assessing the rate and extent of host plant adaptation by whitefly. This required the solving of many logistical problems related to the rearing of insect and plant material, so that the material used in all whitefly generations was comparable. All experiments were conducted in a climate cell under standardised conditions of light, temperature and relative humidity. Whiteflies were maintained as rearings with discrete generations. In each generation tested, whiteflies were removed from each respective rearing using an aspirator. They were lightly anaesthetized under carbon dioxide and a single male and female were placed in a leaf cage, that was clipped to the appropriate host plant. There were typically 20 to 40 such leaf cages per whitefly population per host plant tested. After 24 hours the whiteflies and leaf cages were removed and the number of eggs laid per female counted. The development of the eggs was then followed to assess immature mortality and development time. The method developed was used to assess the rate of adaptation of a whitefly population originating from gerbera, to four cucumber cultivars. 'Trade-offs' with respect to the original host plant gerbera were sought once adaptation to cucumber had been ascertained. The same methodology was used to assess whether or not whitefly could be induced to adapt to a Dutch cultivar of sweet pepper and whether the host plant from which whitefly originated influenced this process or not.The preference of whiteflies reared upon three different host plants (cucumber, gerbera and sweet pepper), was assessed in a multiple choice test, to determine if the host plant from which whitefly originates influences its preference for other host plants with respect to their suitability for whitefly.Finally it was attempted to develop an electrophoretic technique for the glasshouse whitefly using starch and polyacrylamide gel electrophoresis methods for assessing variation in isoenzyme allele frequencies. With this technique, it may be possible to conduct tests for the existence of genetically distinct races between whitefly populations, based upon the genetic distances between them.<strong>Results</strong>Whiteflies originating from gerbera and exposed to cucumber showed adaptation within one whitefly generation. Thereafter the whitefly performance fluctuated so well between cucumber cultivars as over successive whitefly generations. The degree of variation hereby was circa 50 % in the number of eggs laid per female per 24 hours, circa 10 % in the immature development time (in days) and circa 25% in the overall percentage immature mortality. No 'trade-offs' were found, but this investigation was quite cursory in nature and examined just one possibility i.e. that between gerbera and cucumber. 'Trade-offs' may well be detected for other whitefly populations.The work with whitefly upon Dutch sweet pepper indicated that where whiteflies were reared for a single generation upon Lycopersicon hirsutum cv. glabratum prior to being transferred to sweet pepper, their performance was much better than where they had been reared upon gerbera, cucumber or tomato. Although the plant upon which the whiteflies were reared prior to being transferred to sweet pepper influenced their subsequent performance upon this host plant, this phenomenon did not appear to be related to the quality of the host plants concerned with respect to whitefly performance. Rather the specific secondary plant substances which they contained seems to be of importance.The preference tests indicated that the influence of the host plant was such that whiteflies originating from a given host plant, exhibited a greater preference for that host plant relative to that of other whitefly populations. However independent of which host plant a whitefly population originated from, there was an invariate rank order of preference of cucumber>gerbera>sweet pepper. This rank preference, is the same as that of the suitability of these host plants with respect to whitefly performance.The attempts over several months, to develop a reliable electrophoresis technique for the glasshouse whitefly were not successful. However there is still much scope for further improvements. I did not develop this further because of time constraints.<strong>Conclusions</strong>The problems faced in developing a suitable method for assessing host plant adaptation emphasised the importance of consistently using highly standardised plant and insect material of high quality. The glasshouse whitefly appears to adapt readily to plants in its host range, within a few generations. Upon more marginal host plants such as for example sweet pepper, the rate of adaptation appears to be lower than that upon better quality host plants and is influenced by the host plant from which the whitefly originates. Independent from which host plant whitefly originates, that its rank order of preference, so well performance will be cucumber>gerbera>sweet pepper. As the quality of the host plant from which whitefly originates decrease, then the magnitude of the difference between host plants within this rank order also decrease. Life history parameters could clearly be used to indicate differences in whitefly performance upon host plants. The rate at which a population adapts to a host plant measured by these parameters, may provide a good indicator as to the durability of a host plant's inherent resistance to whitefly. However all such tests upon whitefly performance should be conducted over several consecutive whitefly generations so as to account for the between generation variation, even where whitefly is already adapted to the host plant in question. In 'no-choice' tests of whitefly performance, care should be taken to differentiate between laying and non-laying females which indicate respectively antibiosic and antixenosic aspects of host plant resistance. Finally research must still be done to establish whether or not genetically distinct races of the glasshouse whitefly exist and also the relative contributions of genetic and phenotypic variation to whitefly performance and adaptation.<strong>Practical implications</strong>Although the relationship between observations made in my studies and their applicability in the commercial glasshouse situation has yet to be determined, some speculations are now made. If whiteflies are transferred from one crop to another under commercial glasshouse conditions, e.g. through infected plant material or through whiteflies migrating into a glasshouse through air ventilators, then these whiteflies would adapt to the crop concerned within a few whitefly generations, unless the crop was marginal with respect to whitefly use e.g. sweet pepper. In such a case, the chances of whiteflies becoming adapted to the crop are slight. If however air ventilators are covered with insect proof gauze and hygiene standards with respect to the movement and disposal of plant material are strictly maintained, the likelihood of whitefly movement between glasshouses will be reduced. Where more than one cultivar of an ornamental crop is grown in the same glasshouse, the possibility arises that whitefly will adapt to one cultivar and in thus doing become pre-conditioned to adapt to other cultivars in the glasshouse, to which it might not otherwise have adapted