The balsas are clumsy, and difficult to paddle.  The favorite method of locomotion is to pole or, when the wind favors, sail.  The mast is an A-shaped contraption, twelve feet high, made of two light poles tied together and fastened, one to each side of the craft, slightly forward of amidships.  Poles are extremely scarce in this region—­lumber has to be brought from Puget Sound, 6000 miles away—­so nearly all the masts I saw were made of small pieces of wood spliced two or three times.  To the apex of the “A” is attached a forked stick, over which run the halyards.  The rectangular “sail” is nothing more nor less than a large mat made of rushes.  A short forestay fastened to the sides of the “A” about four feet above the hull prevents the mast from falling when the sail is hoisted.  The main halyards take the place of a backstay.  The balsas cannot beat to windward, but behave very well in shallow water with a favoring breeze.  When the wind is contrary the boatmen must pole.  They are extremely careful not to fall overboard, for the water in the lake is cold, 55 deg.  F., and none of them know how to swim.  Lake Titicaca itself never freezes over, although during the winter ice forms at night on the shallow bays and near the shore.

------
Figure

A Lake Titicaca Balsa at Puno
------

------
Figure

A Step-Topped Niche on the Island of Koati
------

When the Indians wish to go in the shallowest waters they use a very small balsa not over eight feet long, barely capable of supporting the weight of one man.  On the other hand, large balsas constructed for use in crossing the rough waters of the deeper portions of the lake are capable of carrying a dozen people and their luggage.  Once I saw a ploughman and his team of oxen being ferried across the lake on a bulrush raft.  To give greater security two balsas are sometimes fastened together in the fashion of a double canoe.

One of the more highly speculative of the Bolivian writers, Senor Posnansky, of La Paz, believes that gigantic balsas were used in bringing ten-ton monoliths across the lake to Tiahuanaco.  This theory is based on the assumption that Titicaca was once very much higher than it is now, a hypothesis which has not commended itself to modern geologists or geographers.  Dr. Isaiah Bowman and Professor Herbert Gregory, who have studied its geology and physiography, have not been able to find any direct evidence of former high levels for Lake Titicaca, or of its having been connected with the ocean.

Nevertheless, Senor Posnansky believes that Lake Titicaca was once a salt sea which became separated from the ocean as the Andes rose.  The fact that the lake fishes are fresh-water, rather than marine, forms does not bother him.  Senor Posnansky pins his faith to a small dried seahorse once given him by a Titicaca fisherman.  He seems to forget that dried specimens of marine life, including starfish,