This note is a critique of a source of errors in the work of Lloyd and Haerendel (1973). The computational region in which they place their cloud is small, and the artifically close boundaries introduce spurious electric fields. Instability growth rate are artifically reduced, and at late times, spurious maximums and minimums appear on the potential contour diagrams. The density contours may exhibit artifical spreading or compression. In an analytic study of a one-dimensional slab model with artifically close boundaries we have reproduced some of the spurious properties.