288 DISCOVERY REPORTS 



fleet. So far as the whaling grounds are concerned it is presumably based on abundant 

 material, but elsewhere, such as in the Pacific sector, the mean positions of the ice-edge 

 are no doubt to be taken as very tentative. 



CONSTRUCTION OF THE CHARTS 



PLOTTED OBSERVATIONS OF ICE 



The charts accompanying this report are primarily intended to show the seasonal 

 variations in the position of the ice-edge, and they include records from every month of the 

 year. In Plates LXXI to XCV each record is separately plotted, and the great majority 

 are derived from the Discovery Committee's ships, the ' Discovery', ' Discovery II ' and 

 ' William Scoresby ', and from a number of factory ships. We have to thank Capt. H. E. 

 Hansen for very kindly providing us with a large body of data consisting of extracts 

 from the logs of some of the factory ships for the seasons 1930-1, 1932-3, 1933-4, 

 1934-5 an d 1935—6. These are more numerous than the observations of the Discovery 

 Committee's ships, but the latter are better distributed in time and space, and include a 

 number of winter observations. A small number of observations are added from miscel- 

 laneous sources where such records have helped to fill gaps in the above data. Among 

 these are some unpublished records kindly supplied by Capt. Sverre Nielsen of ice 

 conditions found by whalers approaching the South Shetlands at the beginning of the 

 whaling season, certain observations of the 'Pourquoi Pas?' in the Bellingshausen Sea 

 (see Bongrain, 1914, plate v), and ice conditions in the Ross Sea given in the Antarctic 

 Pilot. In the charts separate symbols are used to distinguish the observations (a) of the 

 Discovery Committee, (b) of factory ships, and (c) of other expeditions. 



We have included in the charts only those observations in which not only the position 

 of the ice-edge but also the date are known, for the mean latitude of the ice-edge may 

 vary considerably even in the same month, and an undated record is of little value. 

 Furthermore, the plottings are intended to represent the actual outer boundary of the 

 pack-ice belt, and any records such as abstracts from logs in which it is merely stated 

 that pack-ice is in sight are disregarded unless there is something to show that the ship 

 was actually at or near the true ice-edge, and not some distance within an area, for 

 example, of open drift ice. 



Not all the accessible data have been used. It would no doubt be possible to obtain 

 additional observations from factory ships, and a search throughout the reports of all 

 Antarctic expeditions would enable us to plot a few more positions. These would, 

 however, add little or nothing to the regions and times of year at which additional 

 observations are most needed, and would multiply the records elsewhere to an extent 

 which would make it difficult to plot separate observations on charts of the scale which 

 we have found convenient for the purposes of this paper. In any more detailed descrip- 

 tion of ice conditions within a limited area no such records could of course be neglected. 



Separate charts are given for nearly every month, but, since observations are scarce in 

 the winter, May and June records are put together in Plates XCIII and XCIV and July 

 and August in Plate XCV. For most months also (September to March) there are 



