Abstract
The crystal structure of fresh samples of dry growth accreted ice grown on cylinders rotating in an icing tunnel has been determined by two methods: 1) by photographing thin sections between crossed polaroids and 2) by allowing microtomed surfaces to etch and then photographing the etched surface through a metallurgical microscope. The latter technique is capable of far higher resolution than the former. The existence of substructure in dry growth samples, previously reported by Knight and others and Ashworth and others has been confirmed. Under certain conditions, this substructure plays an important role in influencing the observed crystal structures and measured crystal sizes. The crystal sizes (mean lengths, widths and areas) are dependent on both the ambient and deposit temperatures, but are smaller than those reported previously. This is due to differences in the analytical techniques used. The possible values of mean crystal lengths and widths occurring in fresh accreted ice samples are shown to be limited to a relatively narrow set of mean length and width combinations.
