KANSAS ACADEMY OF SCIENCE. 49 



The natural method of treating these observations is to mark on a map 

 the declinations at the several places, interpolate to find points where the de- 

 clinations are the same, and draw isogenic lines, i. e., lines of equal declina- 

 tion. I have done this on a working map, and the results are far from 

 satisfactory. On the supposition that the reports are all correct, the lines in- 

 dicate — 



1st, That the lines are comparatively regular in the extreme eastern part 

 of the State, and in the western organized counties. 



2d, That there is a large area of small declinations, containing the valley 

 of the Kansas as far west as Ellsworth, and extending further from the river 

 to the south than to the north. It contains the part of the Arkansas valley 

 north of tnat river between Hutchinson and Great Bend, and the valley of 

 the Republican. In Marion and Morris counties the declination is much 

 smaller than in the rest of the area. 



3d, The declinations in the valleys of the Saline and Solomon are greater 

 than those in the neighboring portions of the Kansas valley, and correspond 

 approximately to those in the western counties. 



4th, The southern parts of the State are not included, as only the eastern 

 counties of the southern rows have reported, and have already been included 

 in the statement concerning the declination in the eastern part of the State. 



These results are entirely different from any others of which I have any 

 knowledge. The map showing "the lines of equal magnetic declination in 

 the United States for the year 1875," published by J. E. Hilgard, Superin- 

 tendent of the U. S. Coast Survey, indicates that the isogenic lines in Kansas 

 run regularly, with an average direction of about N. 20° E., and other maps 

 indicate the same. If, then, the reports of the county surveyors are correct, 

 there is here an interesting field for investigation; but the reports are not 

 numerous enough to allow any thorough discussion. 



It seems to me, however, that we will get much nearer the truth bv 

 making a different supposition, viz.: that the reports are not all correct. In- 

 correct results can arise from many causes. The following seem to be the 

 most likely to occur: 



1st, Imperfect instruments. When a county surveyor uses a shaky instru- 

 ment, or one whose needle is sluggish and unreliable, I know of no way to 

 remedy the defect, other than by getting a good instrument. If constant 

 errors are suspected, comparison with a good needle might do away with the 

 <lifficulty. 



2d, Carelessness. By carelessness, I do not mean hasty and poor work. 

 Of such a charge each one should be considered innocent until proved 

 guilty. But I do mean the neglecting to humor the needle, to let it swing 

 freely in the box before coming to rest, and to protect it from accidental 

 local attraction. An ax or chain inadvertently left near the instrument, or 

 a pair of steel-rimmed spectacles worn during the observation, will sometimes 

 cause errors amounting to two degrees. 

 4 



