But the achievements of screw steamers are not always satisfactory at first, and time has shown some curious instances where what appeared at first sight a little thing prevented great results. To-day we know somewhat of the screw propeller, but it is very difficult, if not impossible, for the cleverest and most experienced engineer to define his knowledge or to classify his facts so as to deduce any rules from them that shall enable him to lay down fixed laws for the practical guidance of others. In past years more was professed, but still less was actually known, and that which was to be a panacea for the ills of every screw ship proved useless in many instances, and aggravated the evil in others. The patents for propellers are numerous, and some of the specifications interesting and amusing, but of them all there are less than can be counted on the fingers of one hand that have any practical value, or that have influenced the commerce of the world; and we find to-day that the propeller which gives the best results is very simple in form and its working surface a true helix. What is better understood, however, are the proportions, and in them lies the success of the instrument. It is quite true that the blades may be of such a shape and so arranged as to give bad results, but it is very difficult to alter the propeller blade now most generally used and get much improvement thereby. [Illustration: The Propeller of the North German Lloyd Steamer Havel. (From a photograph of the steamship in Handren & Robins’s Erie Basin Dry Docks, Brooklyn.)] In 1865 H. B. M. S. Amazon was found to fall short of her designed speed by nearly a knot, although the indicated horse-power was in excess of the requirements. With a four-bladed Mangin propeller, 12 feet 6 inches pitch, it took 1,940 I. H.-P. to drive the vessel 12 knots. A two-bladed Griffith’s screw of 13 feet 9 inches pitch was substituted, when 12.4 knots were obtained with only 1,664 I. H.-P. But the most remarkable case was that of H. B. M. S. Iris, which had been designed for a speed of 17-1/2 knots, but on her first trial trip, although the 7,000 I. H.-P. was exceeded, the speed was only 16.58 knots. A series of trials was then entered upon to find out the cause of this deficiency, with the result that the screws were discovered to be too large; others of 2 feet 3 inches less diameter were substituted, when a speed of 18.57 knots was attained with the same I. H.-P. Similar instances could be adduced, if necessary, to show how comparatively slight changes in the propeller can produce marked improvements in speed.