Study design/Experiment 1

Counts were taken within each of 15 fields of view on each slide, the number 15 being chosen in order to handle the counting within a reasonable period of time. A pollen grain was considered as germinated if the entire pollen tube produced was greater than the pollen grain diameter. Pollen grain germination was then scored by expressing the number of germinated grains as a percentage of all grains in the field of view.

Note that the d.f. for bud length and bud/bud length add up to 5, which is the number of buds minus 1.

Note that the sampling was done at two layers; a top level with six buds and eight sucrose concentrations to give the 48 (6x8) slides, and a lower level with 15 sets of pollen counts for each field of view within a slide. The experiment can thus be considered as a form of 'randomised block' with bud length as 'block' and with two buds per block, and with sucrose concentration as 'treatment'. Fifteen fields of view were made for each slide resulting in additional replication at the slide level. This results in the structure for the analysis of variance shown above.