320 Terrestrial Magnetism. 



far as they have been communicated and published, are to be 

 found mostly in the papers of the American Philosophical Soci- 

 ety. But few of mine have yet been made public. The field is 

 a very broad one, and the amount of labor yet to be performed is 

 very great. As these elements are subject to a constant and pro- 

 gressive change, besides an annual and diurnal fluctuation, the 

 exact laws of which are as yet unknown, time and a multiplicity 

 of observations at each separate station are required before any 

 generalizations can be well established. 



I have this year commenced making monthly, and sometimes 

 even hourly, observations at this place. In our money making 

 country, I can procure little or no assistance in so unprofitable a 

 business, and my hourly observations are almost too laborious to be 

 continued. My friend and correspondent. Prof Loomis, has col- 

 lected together such observations as have been made, and has pub- 

 lished them in tables and in the form of a chart in your Journal,* 

 but so few have been the observations, and in them generally no 

 attention paid to the annual and diurnal changes, that such a chart 

 must necessarily be only an approximation to the truth, except at 

 the few points which have been particularly examined. In the pre- 

 sent volume of this Journal, pp. 49, 50, Prof. Loomis has, upon 

 ratlier hypothetical grounds, marked his own observations, mine 

 and Prof Courtenay's, with "apparent errors," to a considerable 

 amount. Now most readers will. understand by this, that the re- 

 sults of the observations are absolutely out of truth, or disagree 

 with nature to the amount noted. A careful examination of the 

 article shows that this was not his meaning, for the standard by 

 which these " errors" are made to appear, is more questionable 

 than the observations themselves. Prof Loomis, from a com- 

 parison of the most ancient with the most recent observations in 

 our country, supposes that he has obtained the average annual 

 decrease of the magnetical dip in the United States. He then 

 applies this quantity as a correction to previous observations up 

 to the present year, projects lines of " equal dip" in the direction 

 indicated by two or more points thus determined, and by so much 

 as late observations disagree with these calculations, he has noted 

 them in " error." The only objection which I offer to this mode 

 of expressing difference, is that it will not generally be under- 



* Vide this Journal, Vol. xxxiv, p. 290, Vol. xxxix, p. 41. 



