The first trial of this vessel was made in July of 1900, and was singularly unfortunate. The winch by which the sliding weight was operated broke, and the balloon was so bent that the working of the propellers was interfered with, as was the steering. A speed of 13 feet per second was attained, but on descending the airship ran against some piles and was further damaged. Repairs were completed by the end of September, 1900, and on a second trial flight made on October 21st a speed of 30 feet per second was reached. [Illustration: Zeppelin, pre-war type.] Zeppelin was far from satisfied with the performance of this vessel, and he therefore set about collecting funds for the construction of a second, which was completed in 1905. By this time the internal combustion engine had been greatly improved, and without any increase of weight, Zeppelin was able to instal two motors of 85 horse-power each. The total capacity was 367,000 cubic feet of hydrogen, carried in 16 gas bags inside the framework, and the weight of the whole construction was 9 tons--a ton less than that of the first Zeppelin airship. Three vertical planes at front and rear controlled horizontal steering, while rise and fall was controlled by horizontal planes arranged in box form. Accident attended the first trial of this second airship, which took place over the Bodensee on November 30th, 1905. ‘It had been intended to tow the raft, to which it was anchored, further from the shore against the wind. But the water was too low to allow the use of the raft. The balloon was therefore mounted on pontoons, pulled out into the lake, and taken in tow by a motor-boat. It was caught by a strong wind which was blowing from the shore, and driven ahead at such a rate that it overtook the motor-boat. The tow rope was therefore at once cut, but it unexpectedly formed into knots and became entangled with the airship, pulling the front end down into the water. The balloon was then caught by the wind and lifted into the air, when the propellers were set in motion. The front end was at this instant pointing in a downward direction, and consequently it shot into the water, where it was found necessary to open the valves.’[14]