The objective of this study was to evaluate the distribution of cotton (Gossypium hirsutum cv GSA 71) root system under four depths of trickle irrigation emitter. Cotton was grown on an Amarillo fine sandy loam (fine loamy, mixed thermic Aridic Paleustalf) at Lubbock, Texas. Trickle irrigation was applied at the surface (0), 0.15, 0.3, or 0.45 m depths in amounts calculated using daily pan evaporation, crop coefficient, and crop development stage. Effects of depth of trickle irrigation were evaluated by measuring the cotton root-length density in 0.15 m increments to the 0.90 m depth at distances of 0, 0.25, and 0.5 m perpendicular to the cotton rows. The root-length distributions were not significantly (0.05 %) different in the top 0.15 m when the surface irrigated treatment was compared with the 0.45 m irrigation depth treatment. Root-length densities for the 0.15 and 0.3 m treatments were not significantly different from one another nor were they different from treatments within the evaluated zones (0 to 0.3 m or 0.15 to 0.45 m). Root-length density within the 0.3 to 0.6 m zone, was not significantly different for the 0.45 m treatment compared with the surface irrigation treatment. Root-length density in the 0.6 to 0.9 m zone was not significantly different for any treatment. Emitter depth of trickle irrigation did not significantly influence the depth or distribution of cottonroot development.Knowledge of root-length distribution as a function of depth and lateral distance from the plant is essential to development of models used in irrigation management. However, little information is available on rooting patterns of cotton under trickle-irrigated conditions. Rooting distribution of field-grown cotton has been documented (Weaver 1926;Bruner 1932;Long 1959) and McMichael (1986) described the development of the cotton-root system. Klepper et al. (1973) discussed the interrelationship between soil-water status and cotton root-

In Sri Lanka, coconut (Cocos nucifera L.) has been classified into three varieties, typica, nana and aurantiaca based mainly on their stature and breeding behaviour. Typica and nana are tall and dwarf coconuts respectively while aurantiaca includes intermediate types. Different phenotypes within a variety have been classified as forms of coconut. Current Sri Lankan coconut classification includes 19 different forms within three varieties. There are four different forms of coconut within variety nana, as green, yellow, red and brown dwarfs based on the colour of the epicarp of the fruit. They all conform to the morphological features of variety nana except the yellow dwarf population which was observed to be a phenotypic mixture of different types of coconut. The current study was conducted to differentiate the phenotypes within the Sri Lankan Yellow Dwarf (SLYD) population and to classifY them based on existing criteria. A sample of 200 yellow dwarf palms were studied to determine their breeding behaviour, and the morphological characteristics related to stem, leaf, inflorescence and fruit morphology and yield. Based on quantitative and qualitative data generated, the pure Sri Lanka yellow dwarf coconut form could be distinguished and in addition a new coconut form which was named as Sri Lanka Yellow Semi Tall (SLYST) which was classified within the variety aurantiaca, was identified within the yellow dwarf population. There was a further group of coconuts which could not be placed within the existing varietal classification and they were hypothesized to be a population resulting from cross pollination between SLYD and SLYST.

The examination of ultra-thin sections by the scanning electron microscope was performed comparing spring wheat Laval-19 with several types of wheat of increasing hardness (Yorkstar, Concorde and Neepawa) using both whole kernels and the corresponding flours. The hardness properties of each type of wheat was determined using an Instron Tensile Testor. In hard wheats, we observed a tightly packed structure showing little or no air space in the endosperm. The adhesion between proteins and starch granules was strong enough to break the starch granule rather than to separate at the interface. The soft wheats appeared to have a much looser structure with many intergranular air spaces and no broken starch granules. The microscopy of the flours illustrated the integrity of the starch-protein matrix in the hard flours and a mixture of free starch granules, free protein, and small aggregates of protein and starch in soft flours. The Instron measurements were in agreement with the microscopic examination and both techniques indicated clearly the differences in hardness between soft and hard, leaving the spring wheat Laval-19 a type of wheat intermediary between soft and hard.