It may seem extravagant to waste valuable apparatus on such seemingly small results; but as a matter of fact, the Shepherd-Perret journal of the summer of 1911 was epoch-making in the history of volcanology and in the work of the Hawaiian Volcano Observatory. It proved that skilled observers could dwell inside an active crater and there apply their skills in photography, chemistry, note-taking, and continuous publication. The substance of active lava lakes was proved to have viscosities and solidifications quite different from those implied by gases, and it was shown that different types of thermometers gave negative or positive results useful for the future. Above all, the notes on volcano chemistry by Shepherd and Perret demonstrated that engineering apparatus could be applied to the hottest and most continuously active pit in the world. Their success was at the relatively small expense of a journey and a few machines. Brun of Geneva had set an example of similar work, but Perret’s curve of rise and fall added a more detailed record of the Kilauea pit from day to day than had ever been made before. An observatory is a place of observation and measurement, whether the things observed are glaciers, rivers, stars, the weather, or volcanoes. The motive of observation in modern science is either the quality of what happens or the quantity expressed in lengths and degrees and rates of speed. Remembering the precedent of Vesuvius, I was confronted in Hawaii with the necessity of determining how a volcano should be observed, the need to measure changes in a single volcano, and the need for permanent records of what those changes are. We chose measuring instruments, photographic equipment, and thermometers, and I invented a note-taking system which was compiled into a single record book, from field notes taken uniformly by many different assistants. [Illustration: _7. Volcano House from Observatory, 1913_] [Illustration: _8. Island in Halemaumau lava lake, 1911. Photo by Perret_] [Illustration: _9. Hawaiian Volcano Observatory, 1912_] [Illustration: _10. Jaggar in seismograph vault beneath Volcano Observatory, 1916_] The textbook needs for volcanology are records of the shape, height, number, distribution, temperature, and differences among volcanoes. How gaseous is lava? how radioactive is it? how often does it erupt? and how dangerous is it for human beings? With reference to the source, crack or crater, we need knowledge of how the earth crust is ruptured, how deep are the fractures, and how much accompanied by earthquake is the wedging upward of lava in those cracks.