376 THE VOYAGE OF THE 'DISCOVERY' [Sept. 



From our hills we could see two possible roads to the 

 south. One lay outside the White Island and promised the 

 smoother travelling, but necessitated a considerable detour. 

 The other was more direct, and led towards the high black 

 cape which we commonly called the Bluff; it passed between 

 the 'White' and 'Black' Islands, and though it seemed to 

 contain some rough places, I thought it worth exploring, on 

 the chance of saving the longer distance. 



But in making a spring journey to observe these routes it 

 was obvious that as one or the other must eventually be taken 

 by the main party, in either case that party must pass around 

 the Bluff, so that it was advisable that the southern recon- 

 naissance party should carry enough food to be able to 

 establish a depot at the Bluff. 



Besides these early efforts at clearing the routes for the 

 main journeys, one other matter claimed our attention in the 

 spring programme : we had still to communicate with the record 

 at Cape Crozier. It was advisable that this should be done 

 before the longer journeys were undertaken, but I thought 

 it might be left until after the reconnaissance parties had 

 returned. 



As a preliminary to the commencement of the spring pro- 

 gramme, I decided to make a short trip to the north with the 

 dogs and a party of six officers and men, mainly in order to 

 test the various forms of harness which we had on trial, and 

 to find out whether the dogs pulled best in large or small 

 teams ; but incidentally there were many minor topographical 

 features in this direction which we could not see clearly from 

 the hills, and which we now wished to make sure of. 



On September 2 we started in a blizzard and camped in 

 conditions of some discomfort ; on the following day we pushed 

 on past the Turtle Rock and found ourselves brought up by 

 the long tongue of a glacier. Although this was but eight 

 miles from the ship, from our hills we had only been able 

 to make out a wavy, indistinct shadow, showing how extra- 

 ordinarily limited is the distance at which one can detect ice 

 disturbances. 



