Constraints on nutritional compensation in acridids
Some aspects of the ability of locusts and grasshoppers (Acrididae) to compensate for nutritional shortfalls were studied, with a special emphasis on the factors which constrain this ability. Chapter 1 investigates the effects over the short-term (12 h) of the plant-produced allelochemical tannic acid on the ability of Locusta migratoria (L.) and Schistocerca gregaria (Forskal) to compensate for dilution of dietary proteins and carbohydrates by increasing consumption. Tannic acid had no effect on compensatory feeding by L. migratoria, and stimulated feeding by S. gregaria. Chapter 2 extends this study over the longer-term (fifth instar) for L. migratoria. Over this period, tannic acid restricted intake and reduced growth of those insects fed lowprotein diets, indicating an inhibitory effect on compensatory feeding for protein. In addition, the levels of dietary proteins influenced regulation for carbohydrate intake and, to a lesser extent, vice-versa. A detailed discussion is presented of the ways that some dietary components can influence the intake of others, and how failure to take this into account can lead to poor experimental design and interpretation. Chapter 3 investigates some mechanisms involved in dietary selection by the grasshopper Schistocerca americana (Drury). It was found that S. americana conditioned on distinctly flavoured protein-inadequate diets then tested on nutritionally similar diets with the familiar or a novel flavour, tend to eat more of the novel-flavoured diets. This suggests that conditioned neophilia, possibly in conjunction with aversion learning, may be a factor facilitating dietary selection in acridids. Chapter 4 investigates the patterns of feeding and dietary selection behaviour of the polyphagous grasshopper Taeniopoda eques (Burmeister) in its natural desert habitat. Despite the overwhelming thermoregulatory requirements and unpredictable variability inherent in ecological complexity, these insects nonetheless maintained a pattern of feeding comparable to that observed under controlled laboratory conditions. The patterns of dietary selection behaviour were concordant with some of the mechanisms observed to operate in the laboratory. Chapter 5 addresses an important inadequacy in the methodology currently used to investigate some aspects of nutritional compensation. A computer-generated data set is used to illustrate how the analysis of the currently popular ratio-based nutritional indices may be flawed, and how this may be overcome using as an alternative the analysis of covariance.