In 1896, the balloon and all the supplies were taken to Dane’s island, near Spitzbergen, but as the desired south wind did not blow, Andrée returned to Sweden. In May, 1897, Andrée and his companions again went to Dane’s island. A balloon house was built, engines were set up for producing hydrogen gas, and in due time the balloon was inflated. By July 1 everything was ready. Andrée and his two companions now waited anxiously for a south wind, which they believed would blow them to the North Pole. They waited ten days, until, on the morning of July 11, 1897, a strong, steady wind from the south was blowing. Great was the excitement on Dane’s island when the men began to tear down the house where the balloon was imprisoned, and attach the car. At 2.30 P.M., July 11, 1897, Andrée and his two companions, Nils Strindberg and Knut Fraenkel, jumped into the car, and gave orders to cut the last ropes. [Illustration: ANDRÉE BEGINS HIS JOURNEY.] Slowly the immense, airy structure rose to a height of a few hundred feet, and sailed in a northerly direction out over Dane’s gate. Then it dropped suddenly, as if it had received a current of air from above, and almost touched the sea. Andrée threw out some sand bags, when the balloon rose again to a height of about three thousand feet, and sailed away in the same northerly direction. About an hour after the start, it was lost to sight in the clouds. Some days later a carrier pigeon was shot in the rigging of an Arctic schooner off Spitzbergen. The pigeon had a message from Andrée tied under its wing. The message was dated July 13, and stated that the balloon sailed one hundred and forty-five geographical miles to the northward, and then headed east. It had traveled forty-five miles eastward when the pigeon was sent out. From that day to this, no other message has been received from Andrée. Andrée believed that his balloon would float for six weeks, but the men who watched the start, said that it lost much gas and much ballast before it passed out of sight. They thought that it might have floated about fifteen days. Two thirds of the guide ropes, upon which Andrée depended for steering, were also lost at starting. At first the balloon traveled about twenty-five or thirty miles an hour. At this rate of speed, sailing northward, Andrée should have reached the pole in less than two days. But every ray of sunshine, every puff of colder or warmer wind, cause a balloon to rise or fall, and the methods of guiding and handling the delicate appliance are not yet thoroughly understood.