302 DISCOVERY REPORTS 



This applies to Hoseason Island, Smith Island, Low Island and Snow Island, all of which 

 we tried to fix in passing by sights taken at sea, and also to the Melchior Archipelago in 

 Schollaert Channel and Port Lockroy in Wiencke Island, at both of which places we were 

 able to land but unfortunately had no day when adequate sights could be obtained. On 

 the passage from Deception Island to Melchior Archipelago and passing Austin Rocks, 

 whose position could now be taken as correct, we obtained a series of six sun sights, three 

 before noon and three after ; at the same time, as well as at other times, cross-bearings were 

 taken to all tangents of islands and peaks in sight. At dusk a set of six practically simul- 

 taneous star sights was obtained. The sun sights, when worked out and plotted, agreed 

 together very well, as did the star sights, and the same is true of the results of the bearings 

 taken : but although the observations were checked repeatedly by myself and others, the 

 results obtained by the various methods all differed. In view of this discrepancy, for 

 which so far no explanation has been found, it would seem advisable to treat the results 

 of observations taken at sea with extreme caution. 



A few soundings were obtained in Port Forster (Deception Island), at Melchior 

 Anchorage, and later in Admiralty Bay (North or Visca Anchorage), which should prove 

 useful. 



On our passage from Melchior to Port Lockroy a running sketch survey was carried 

 out by means of the compass and log, the errors of both of which had been so long 

 observed as to be known very accurately.^ 



On one day whilst at Port Lockroy (Plate XL I, fig. i) the ship went for a short cruise ; 

 O'Connor, Smith and I stayed behind, and after carrying out a rough sextant triangula- 

 tion, sounded out the harbour, using the scale and meridian given by Charcot. A large 

 number of soundings were taken with the small Lucas machine from a pram. It is an 

 excellent harbour; but the surrounding glaciers, by calving, make it uncomfortable and 

 occasionally dangerous. The soundings under these glaciers were therefore reduced to a 

 very few, owing to pieces of ice frequently breaking off. 



In addition to the surveying, a large number of soundings were obtained at sea with 

 careful astronomical positions. The Lucas machine, with 6000 fathoms of wire, driven 

 by a Brotherhood steam engine for heaving in, was always employed in making these 

 soundings. Depths up to just under 3000 fathoms were obtained, although it has to be 



1 The method, although well known, may be described briefly. A departure from a fixed position having 

 been plotted, compass bearings are observed to all prominent objects in sight, such as capes, islets, points 

 and edges of glaciers. Care must be taken to write the objects down carefully and fully, so that no mistake can 

 occur when next they are observed, for all objects change their appearance greatly as the direction alters, and 

 should therefore be kept under observation continually. At the same time as the compass bearings (which, 

 owing to the change of position of the ship, should not include too many points) the time must be taken and 

 the patent log read, while the course should also be carefully recorded. This procedure is continued at 

 intervals to the end of the passage, when the position reached, if not already known, should be fixed by 

 the best means available. In the present survey the latitude of Melchior Observation Spot was observed, and 

 the longitude accepted from previous observers. Port Lockroy was accepted entirely from the position 

 assigned to it by Charcot. The courses steered, after being carefully corrected, were plotted and the bearings 

 laid off, the result being squared in with the end positions. All appeared to give satisfactory results, the 

 current being negligible. 



