Most water-tube boilers had failed to satisfy the demands made upon them; a result sometimes brought about by their inventor's attempts to secure good circulation. The circulation in many boilers had been so violent that dry steam could not be produced. Alfred Yarrow of Yarrow and Co., Poplar, had carried out a series of experiments with the object of producing a water-tube boiler suitable for marine use; his water-tube boiler consisted of a cylindrical upper part of about 20 inches in diameter by 6 feet long, and two lower prismatic water pockets about 6 feet in length. These three chambers were sub-divided longitudinally, so as to give free access to their interior. The upper cylinder was half filled with water, and was connected to the water pockets by a large number of straight galvanised steel tubes slanting downwards, at an angle of about 30 degrees, forming in section an inverted letter "V", of which the fire grate formed the base. The flames rose from the furnace and passed to the right and left, between the slanted tubes and round the top cylinder to the funnel. These tubes could not be overheated because they were always filled with water. Alfred Yarrow had designed his boiler in response to the demand for higher pressures and not because he did not have faith in his locomotive marine boilers.