METEOROLOGY. 25 



There seem to be no data to warrant the extension of siunmer temperature of 70° much 

 north of the Missouri river, as has been done by Mr. Blodgett in his last work, and it is accord- 

 ingly made to run near the "coteau" north of that river, which rises several hundred feet 

 above it, forming the ridge between it and the Saskatchewan. On the other hand, west of the 

 mountains, the line of 70° is substituted for G5° for Bitter Root valley, the prairies near Flat- 

 head lake, and thence down the valley of Clark's Fork and the Columbia, from Fort Colville 

 south. Tlie rapidly decreasing altitude compensates for the greater distance north of part of 

 this tract; but allowance should be made, in some places, for the local influence of surrounding 

 mountains and forests. 



The w'holo of the Great Plain, and the unwooded hills and valleys north and west of the 

 Great Bend of the Columbia, are assumed to have a mean summer temperature above 70°, 

 wliicli is partly jirovcd by a comparison of temperatures on the same days at Fort Dalles and 

 in the Yakima valley. 



A series of observations at Fort Walla-Walla, twenty-nine and a half miles east of the 

 Columbia, at the confluence of the Walla- Walla with that river, altitude 1,396 feet, shows that 

 the lowest parts of the plain have a summer temperature as high as 73^, and a line has been 

 drawn to approximately include this district. 



The following are the means of observations at this j^ost, (uncorrected:) 



Fort Wdla-Walla, 1857-' 58.— Summer, 73.00; winter, 34.11; autumn, 52.59; spring, 51.85. 

 The winter, compared with the mean for four Avinters at Lapurai, on the Kooskooskia, and five 

 at Fort Dalles, appears a little below the average for the latitude and elevation. 



From the two former records the mean for the Great Plain is assumed as above 35° in wnnter, 

 though records at Fort Colville are wanting to complete the curve, and the northern parts of 

 the region included may be a little colder. 



25° is retained as the mean for the Bitter Root valley, although the second winter's observa- 

 tions, as already shown, raises it to 27.(3. The lines of 05°, C0°, and 20° are approximately 

 located, by allowing 3° for each 1,000 feet, at points whose elevation is known, and then 

 connecting them. 



There is too little known of the geography of Oregon to make a safe connexion between the 

 known points in Washington Territory and Utah. 



No records for any point, except the Dalles, exist from which the summer lines can be 

 projectedjn the interior of Oregon, and too few altitudes are known to make even an approxi- 

 mation by reduction from better known regions. 



West of the Cascades and Sierra Nevada the lines are drawn from the same data recorded in 

 Mr. Blodgett' s book, but altered to conform more closely to the surface of the country. The 

 fixed points at which they were taken being usually in valleys, or on the low land along the 

 coast, with high mountain ranges intervening. The lines are made to include only so much of 

 the country around each post of observation as, from elevation and surrounding circumstances, 

 can be considered as having a similar climate. 



Fort Jones, the only very elevated post, (2,570 feet altitude,) when compared with Oregon 

 City or Vancouver, both north of it, shows the influence of local situation on climate, and the 

 impossibility of connecting the isothermal lines naturally in so undulating a country. 



The following table gives a comparison of the mean temperatures on a lino of stations between 

 latitude 4G° and 48°, from the mouth of the St. Lawrence to Puget Sound, those points being 

 selected which are nearest' the great route of travel by the St. Lawrence, the lakes, and the 

 d 



