Dr Forry on the Climate of the United States. 215 



increase of 30° 83', for example, atFort-Snelling, comes upon 

 a mean winter temperature of 15° 95', while at Fort-Sullivan, 

 on the same parallel, the increase of only 17° 16' follows a 

 winter temperature as high as 22° 95'. Between northern 

 and southern latitudes, this contrast is still more marked ; for 

 while, at Fort-Snelling, there is a difference of 13° 46' between 

 the months of February and March, and at Key West only 

 1° 56', the temperature of February at the former Ls 18° Q&, 

 and at the latter 72° 15'. 



What month expresses the nearest equivalent to the mean an- 

 nual temperature ? — This is a question which has excited 

 considerable controversy, and in regard to which there still 

 exists a difference of opinion. By Kirwan it is strenuously 

 contended that the month of April expresses this equivalent ; 

 while Humboldt, on the other hand, shews, by extensive ta- 

 bular statements, that October is better entitled to this cha- 

 racteristic ; and on either side of this question are arranged 

 many other authors of lesser note. As the laws of Nature 

 are universal, these phenomena, like all others, must be 

 susceptible of systematic arrangement ; and lest it may be 

 thought presumptuous in the author to attempt to decide be- 

 tween such high authorities, he will state in advance that 

 the diverse systems of climate presented in the northern re- 

 gions of the United States, on the same parallels, afford a 

 means of comparison doubtless heretofore unequalled. 



As respects the controverted question, whether April or 

 October expresses a nearer equivalent to the mean annual 

 temperature, the following deductions is clearly authorised 

 by the tabular abstracts appended to the writer's work on 

 " The Climate of the United States," &c. In excessive cli- 

 mates, tlie mean temperature of April is generally as high as 

 that of the year, while that of October is considerably higher ; 

 and in modified climes, it will be found that the former is gene- 

 rally as much lower as the latter is higher. Now, this relation 

 is precisely what might have been anticipated from a consi- 

 deration of the preceding facts ; for as the vernal increase of 

 temperature is always much greater in excessive than in modi- 

 fied climates, it follows that if, under any circumstances, April 

 expresses a nearer equivalent than October, it must be when 



