The result was most remarkable. The yellow liquid lava went into the salt water without making any column of steam at all; the sea bottom simply received it with its rush downward, the water boiling superhot, and the lava taking the water vapor into itself. The phenomenon was not due to the rising of dry steam, for there was no condensation cloud above. Scientists explained it by assuming a shell of lava making a tunnel under the ocean, with the crust ending just at sea level. Such a submarine arch was definitely not present, for the waves surged back and forth and, Matsumoto states, there was no sign of a submerged reef. The motion picture bears this out. What was probably going on is what happens to slag in a patented process of the steel mills, where the glow liquid is flowed over a perforated surface emitting hundreds of water jets, and the melt at 1300° Centigrade absorbs the water without making visible steam. The slag turns into a myriad of microscopic glassy spheres, becoming a kind of pumice. A peculiarity of this substance is that if it is cooled at 700° Centigrade it will pass a critical point and give up the absorbed water with explosive effects. It seems likely that Matsumoto’s golden torrent sweeping into the ocean was so excessively hot that it took up the water and continued to flow down the sea bottom as a water-charged product. The snapping and crackling effects, and the submarine earthquakes, making localized tidal waves such as those noted in 1919 when such a torrent entered the sea, may be due to the explosive cooling when the slag gives up its water. The use of color motion pictures is one of the many improvements owed to modern science, and the mapping of lava flows by airplane photography. This gave Macdonald a new weapon for surveying the volume accumulation at the time of the 1950 outflows, for from air photographs he got exact outlines of the flows. These, checked against calculated thicknesses, gave him volumes which could be compared with volumes of older flows proportionate to areas. These calculations showed that nothing since 1868 has yielded such large volumes of lava, per days of outflow.