On the other hand the natural harbors of the Panama route have successfully met the demands of commerce for the last four hundred years. On the Pacific end practically no harbor improvements will be necessary. On the Atlantic the present needs are satisfied, but the large steamers of the future may require deepening which can be done and the resulting channel easily maintained since there is no persistent filling in process such as characterizes the Greytown harbor. Volcanoes have long been plentiful in Central America, especially near the proposed Nicaragua canal. Nicaragua Lake, so geologists say, owes its separation from the Pacific to a great upheaval. There is now an active volcano near which ships would have to pass. From January 1, 1901 to April 30, 1904, a period of forty consecutive months, the instruments of the Instituto-Fisico Geografico, located 60 miles from the locks of the proposed canal, recorded 43 tremors, 91 slight shocks and 35 strong shocks, some of which lasted 16 minutes. Similar observations at Panama for the same period revealed only 6 tremors and 4 slight shocks, the longest being for a period of only 10 seconds. The lock gates of a canal might very easily be injured by earthquakes; and common sense would dictate that other things being equal, the canal should be placed where the shocks are fewest. Strong trade winds rush through the San Juan gorge at all seasons. The rainfall near the Atlantic is enormous, averaging from 260 to 270 inches per year, and rain may be expected any day. In the western part the fall is only 65 inches, and there is also a well defined dry season. Clear vision is essential to safe passage through the canal and it is extremely doubtful if it could be obtained under the above conditions. Still more serious perhaps is the excessive curvature of the channel for 50 miles of its course. It is impossible to reduce the curvature to the limit which experience on the Suez canal has proved necessary for safety and speed. Furthermore the channel must carry off to the sea the drainage from 12,000 square miles of territory. This cannot do otherwise than create currents and eddies unfavorable to navigation. The Panama route has no continued strong winds; the curvature is comparatively favorable; the annual rainfall is from 140 inches on the Atlantic coast to about 60 inches on the Pacific, with a definite dry season of three months; and the concensus of expert engineering opinion is that there need be no objectionable currents if proper provision is made for the regulation of the Chagres river. This phase will be discussed later as will also the question of curvature. Much has been said about the advantages furnished by Lake Nicaragua which covers about 70 miles of the Canal route. However, for 29 miles of that distance, an artificial channel through soft mud would be necessary, and dredging would probably be practically continuous for maintenance.