Poas crater was very different, with a crater lake of boiling water surrounded by bright-colored horizontal layers of ash. We found buried bombs from a recent eruption which had punctured the soil with holes one or two feet across. There was wild adventure for me in being given a horse at 4 A.M., equipped with a rotten saddle, which slipped when I mounted him. The horse resented me in the early morning darkness, having just left his grain, and immediately bucked off both me and the saddle. More adventure followed. On the ride up the mountain and in the midst of the forest we encountered a jaguar trap which had recently caught two big cats. It was a pen, roofed with logs baited with a fowl, and disguised with brush; a shutter fell and closed the opening when the bait was touched. On the way down we had a terrific tropical thunder storm, with sheets of cold rain, and I got chilled to the bone and was sick with dysentery for two or three days. There are a dozen volcanoes like these two on the backbone of the Costa Rica rocky mountains. They trend in a ragged line from the Panama boundary on the southeast, to Nicaragua on the northwest. All have records of explosive activity, but lava flows are rare. Beginning at Nicaragua the line of the Cordillera, capped with volcanoes, continues through Salvador, Honduras, and Guatemala; and some of the lower ones have lava flows. Cosequina is famous among them; and conspicuous as a frequently active volcano is Santa Ana in Salvador, one peak of which is Izalco, the index volcano of Central America, erupting frequently. Other index volcanoes are Kilauea for Hawaii, Stromboli for Italy, and Bogoslof for the Aleutians. The next line of volcanoes, also trending northwest, extends from Guatemala into southern Mexico. The Costa Rica line overlaps the northeast side of the Nicaragua-Salvador line, and this in turn overlaps the Guatemala line, and so on. The chains of volcanoes are over an echelon of cracks, surmounted by heaped-up lava peaks on the continental divide. From the point of view of experimenting with volcanoes, the exploration of the Cartago earthquake and Poas and Irazu craters and a study of their relations typified the unsatisfactory combination of upheaved mountains of strata and of volcanic eruptions and underground friction. This extends all the way along the Cordillera from Patagonia to Alaska. I say unsatisfactory because from the science standpoint, the action of eruption or earthquake is far scattered in time and place, and only local observatory geophysics and traveling scientists will do the work. Cartago is directly at the foot of Irazu Volcano, but the volcano did not erupt simultaneously with the earthquake. In the same way Messina is at the foot of Etna, and Tokyo is at the foot of Fujiyama; and the great earthquakes do not accord with eruptions. Sakurajima in 1914 was an exception, it had a quake after outbreak.