The Virgo, carrying with it Andrée, his balloon, and a party of geologists, left Tromsö on June 14, 1896, and nine days later a suitable place for building the balloon house was found on Danes Island. The landing of the balloon and the building of the house occupied nearly a month, and it was not until July 27 that everything was ready for a start. Unfortunately, however, the wind, which had been for the most part favourable while the preparations were in progress, now veered round, and for the rest of the summer it blew steadily from the north, when it did not drop altogether. Week after week passed by without bringing any prospects of a start, and at last Andrée was obliged to pack up his balloon and return home, hoping for better luck next year.

On May 30 he was back at Danes Island once more with his balloon, which had been undergoing sundry modifications during the winter. The house had fallen somewhat into disrepair, but it was soon put in order, and the inflation of the balloon, which was begun on June 19, was finished at midnight on the 22nd. Everything was now ready for a start, and on Sunday, July 11, Andrée decided to take advantage of a stiff breeze which had set in from the south. Standing in the car with his two companions, Frænkel and Strindberg, he gave the orders for the ropes to be cut. The balloon rapidly ascended, to a height of 600 feet, and, after a temporary drop, floated away north over the flat peninsula of Hollændernæs. It remained visible to those at Danes Island for about an hour. Then it disappeared over the northern horizon, never to be seen again.

The only news that the world ever received of Andrée and his companions after this did nothing towards solving the mystery of their fate. Of the thirteen buoys which he carried with him on board his balloon only four were ever recovered. One was picked up at Skjervö, in Norway, and was found to contain a message to the effect that it had been thrown out at 10 o’clock on the night of July 11. Another, which had been dispatched on its journey about an hour later, when the party had reached lat. 82° N., long. 250 E., was recovered off the coast of Iceland. The two remaining buoys bore no message from the explorer. On July 15, 1897, the sailors of the s.s. Alken shot a carrier pigeon which had been let loose two days previously in lat. 82° 3´ N., long. 150 5´ E., but that was the latest intelligence of the explorers that ever reached their friends at home.

Many expeditions were sent out to their rescue, and reports were brought in by natives of shots heard upon the ice and figures seen on the drifting floes. Fishermen, too, said that they had heard cries for help, and that they had seen what looked like a deflated balloon drifting on the sea. But, carefully though these clues were followed, they came to nothing, and it can only be supposed that, descending on some vast ice-field far from human aid, probably somewhere between Spitzbergen, Nova Zembla, and Siberia, the unfortunate men perished miserably of starvation and exposure.

ANDRÉE’S BALLOON IN ITS SHED

CHAPTER XXXII
THE LATER VOYAGES OF SVERDRUP AND PEARY

As we have already seen, it was as Nansen’s companion on his journey across Greenland, and as his second in command on the Fram, that Otto Sverdrup first acquired the taste for Arctic travel which, in 1898, led him to undertake an expedition on his own account. The primary object of his new journey was to complete the survey of the northern shores of Greenland which had been so brilliantly begun by Peary, and to discover once and for all whether there lay any land beyond it in the direction of the North Pole. Failing that, he proposed to examine Grinnell and Ellesmere Lands, of which vast tracts still remained totally unexplored.

Sverdrup sailed from Upernavik on August 5, in our old friend the Fram, which had been refitted for the new voyage, and was soon well on his way up Smith Sound. On reaching Hayes Sound he decided to go into winter quarters there, as it was, of course, impossible to attempt to reach a much higher latitude that season, and there was plenty of good work to be done in the neighbourhood. A couple of sledge journeys across Ellesmere Island put him in possession of many new facts concerning the geographical features of that country, and, as soon as spring came round, he began his preparations for pushing north. Unfortunately, however, the season was very unfavourable, and he soon found that, for that year at any rate, his project of following the north coast of Greenland must be put aside, so he determined to explore Jones Sound, which had never before been followed to its juncture with the Polar Sea.

In this he was entirely successful, and when the next winter came round it found him comfortably ensconced in winter quarters on the south coast of Ellesmere Land. It was while Sverdrup was away exploring the coast that the career of the Fram was very nearly brought to an untimely end by fire. Fortunately, however, she was saved, and early in August she was afloat again. All serious thought of pushing up Smith Sound had now been abandoned, and instead, Sverdrup made for Belcher Channel, at the mouth of which the next winter was spent. During the spring and summer the work of exploration went merrily on, with the result that the indefatigable Sverdrup had soon added much of the unknown coast of North Devon to the charts.