42 DISTRIBUTION OF RAIN. • 



June 22. — In our course this morning, we struck one of the principal 

 branches of the Salt Fork near its source, and followed it down upon 

 the left bank to its confluence with the main stream. Below the junc- 

 tion the stream was fifty yards wide, but only about one-fourth of its 

 bed covered with water. This branch of Red river, like the other, heads 

 in the border of the "Llano estacado," and directly at the source is an 

 elevated hill with abrupt vertical sides, terminating in a level summit; 

 below this, upon the south bank, are two round mounds that can be 

 seen for many miles. 



We were much gratified in finding the water at the head of this 

 branch, as in the north fork, sweet and wholesome. This settles the 

 question that these branches of the river do not take their rise in salt 

 plains, as has heretofore been very generally supposed. On the contrary, 

 at their sources, which are in the eastern borders of the "Llano estacado," 

 the water is as pure and wholesome as can be desired. And this charac- 

 ter continues upon all the confluents until they enter the gypsum forma- 

 tion, when they become impregnated with salts, that impart a new char- 

 acter to the water, which continues to its junction with the Mississippi. 



A solitary cotton-wood, with an occasional clump of willows, constitute 

 the sylva, of this portion of the river. The soil in the valley is an 

 arenaceous red alluvium, and would be productive wilh the aid of arti- 

 ficial irrigation. 



The bluffs bordering the valley are, at this place, about one hundred 

 feet high, and composed of a deep red clay, overlaid with a stratum of 

 drift ; and this surmounted with a capping of calcareous sandstone from 

 five to fifteen feet thick. 



Upon the rocky bluffs bordering the river we found silicified wood in 

 great quantities, strewed about over a distance of two miles. The petri- 

 faction was most perfect, exhibiting all the fibres, knots, and bark, as 

 plainly as in the native state, and was quite similar to the cotton-wood. 



This evening we have another rain coming from the northwest, which 

 will increase our chances for finding water in advance. 



As it will be seen by a reference to the meteorological tables, our 

 barometer has, in almost every instance, been a certain index to the 

 weather from the commencement of the march. Sometimes, indeed, it 

 has exhibited a most extraordinary depression of the mercury for two or 

 three days previous to a storm ; but in no instance has it failed to rain 

 before the instrument would resume its usual range. 



During the last three summers which I have spent upon the plains, 

 as has been before observed, I have seen no rain of consequence from 

 about the middle of May to the middle of August. And after passing 



