This sea ice atlas follows Swithinbank's atlas (Reference Swithinbank1960) (reviewed in Journal of Glaciology, Vol. 4, No. 35, 1963, p. 643–44) in the series produced at the Scott Polar Research Institute, and the method it employs is essentially similar. There are 96 frequency charts, half of them covering the waters immediately adjacent to Graham Land, and half of them the South Atlantic and South Pacific waters to east and west of the peninsula. Each pair of charts covers a period of a quarter month, or roughly a week. The main feature of presentation is that isopleths indicating the limits of ice of particular concentrations at particular times of year are not used; that method is rightly judged to be misleading, both because of the large amount of interpolation required in the present state of knowledge, and because of the dangerous tendency, encouraged by the method, to think in terms of an “average” state of the ice. Instead, sector diagrams are used, which show the ice or open water actually observed in a particular area at a particular time of year, and reflect also the number of such observations. This last point is important, as it affects the reliability of any sector diagram. If it is required to know in which years these observations were made, the ice summaries, which occupy the remaining part of the atlas, supply this information. The user must then draw his own deductions. What the atlas seeks to provide, therefore, is the same as its predecessors of this type, namely the fullest and least distorted summary of all known observations of seasonal change in the state of the ice.

This information can be used for a large number of purposes, of which the most frequent are likely to be the planning of shipping or sledging routes, and the correlation of ice data with meteorological or oceanographical data in climatological or related studies. The method used has the further advantage that new observations may be incorporated relatively easily as they come to hand. Enough information on the compilation of the atlas is given in the one page of explanatory matter for the user himself to be able to recalculate the contents of a given sector diagram after adding, say, five years’ new observations. The whole atlas could, and it is to be hoped will, be brought up to date every decade or so.

The atlas is the first application of the sector diagram method to southern waters. The area selected is probably the only one for which there were enough data when the work was started in 1955, Graham Land waters being the most frequented part of the Southern Ocean. The sources consulted, listed chronologically on four pages at the end of the atlas, fall into three groups: ships’ logs, ice reports from shore stations, and (the smallest group) published accounts. Heap covers the period from 1898 to 1961, but probably three-quarters of his data relate to the last seventeen years of this. The volume of data will continue to increase, as activity increases and aircraft reports (hitherto a rarity here) start to displace the others, as they have in the Arctic. Before many years, it is possible that the greater part of sea ice observation will be by satellite, and the volume of data will further greatly increase. None of this, of course, lessens the value of the information about the past presented here—quite the contrary. For the rest of the Southern Ocean, we rely still on the U.S. Navy Hydrographic Office's atlas (1957) (reviewed in Journal of Glaciology, Vol. 3, No. 27, 1960, p. 659–60)—an excellent compilation, but with the disadvantages inherent in an isopleth method of presentation. As more observations are recorded, the sector diagram method could be extended to cover other parts of the Southern Ocean.

The reader will by now be aware that critical comment on Heap's method is not to be expected from this reviewer, who must admit to having played some part in its formulation, and who will therefore not go beyond saying that he can think of no better method, and is delighted to see how effectively it has been applied to this area. The detailed working out, which includes introduction of a number of improvements and necessary modifications, has been carried out with extraordinary application, accuracy and patience, and it may surprise some to know that it is entirely the work of one man, unaided by any team of assistants. The Admiralty Hydrographic Department have made an excellent job of the printing. Criticism of their contribution may be restricted to two small points: absence of any pagination can be irritating, and binding in limp covers makes a work of this bulk (certainly the weightiest work on sea ice) difficult to handle. The price, which is fixed by the Stationery Office rather than the Admiralty, is excessively high; not that the work is not worth it, but it may deter many, and not only individuals, from finding out how good the work is.