1849.] THE PIROROCO. 89 



just in time for supper, and were heartily received by Senhor 

 Calistro. After a few days more I left his hospitable roof, 

 loaded with luxuries : eggs, tapioca, a roast pig, pine-apples, 

 and sweets were sent to my canoe; and I bade adieu with 

 regret to my kind host. 



On our way down I again encountered the "piroroco" 

 when I hardly expected it. We had gone in shore at a sugar 

 estate to wait for the tide, when the agent told us we had 

 better put out further into the stream, as the piroroco beat 

 there. Though thinking he only wished to frighten us, we 

 judged it prudent to do as he advised; and while we were 

 expecting the tide to turn, a great wave came suddenly rushing 

 along, and breaking on the place where our canoe had been 

 at first moored. The wave having passed, the water was as 

 quiet as before, but flowing up with great rapidity. As we 

 proceeded down the river, we saw everywhere signs of its 

 devastations in the uprooted trees which lined the shores all 

 along, and the high mud-banks where the earth had been 

 washed away. In winter, when the spring-tides are highest, 

 the " piroroco " breaks with terrific force, and often sinks and 

 dashes to pieces boats left incautiously in too shallow water. 

 The ordinary explanations given of this phenomenon are 

 evidently incorrect. Here there is no meeting of salt and 

 fresh water, neither is the stream remarkably narrowed where 

 it commences. I collected all the information I could re- 

 specting the depth of the river, and the shoals that occur in it. 

 Where the bore first appears there is a shoal across the river, 

 and below that, the stream is somewhat contracted. The tide 

 flows up past Para" with great velocity, and entering the Guama 

 river comes to the narrow part of the channel. Here the 

 body of tidal water will be deeper and flow faster, and coming 

 suddenly on to the shoal will form a wave, in the same 

 manner that in a swift brook a large stone at the bottom will 

 cause an undulation, while a slow-flowing stream will keep its 

 smooth surface. This wave will be of great size, and, as there 

 is a large body of water in motion, will be propagated onwards 

 unbroken. Wherever there are shallows, either in the bed or 

 on the margin of the river, it will break, or as it passes over 

 slight shoals will be increased, and, as the river narrows, will 

 go on with greater rapidity. When the tides are low, they rise 

 less rapidly, and at the commencement a much less body of 



