A field experiment was designed to investigate whether the type of cutting planted affects the subsequent growth of willows. Cuttings of five different types (A–E, factor Type) were to be planted, and growth parameters would be measured over the following seasons, including yield at the end of the first year. At planting time, it was realized that the cuttings to be planted varied greatly in size, and that this might also have an effect on subsequent growth. Two options were considered here. Cutting size could be confounded with blocks, so that each block contained cuttings of the one size only. Alternatively, cutting size could be investigated as an extraneous factor, in addition to type. The second option was taken, and cuttings were classified as small (S), medium (M) or large (L, factor Size). Not all of the type × size combinations were available, and the total number of plots was fixed at 25. The design was based on a five-block RCBD with respect to cutting type, and the different sizes were allocated in as balanced a way as possible across blocks (factor Block) and cutting types. The yield (variate Yield) with allocation of size and type combinations to the five blocks.