1 18 Annual Report of the S. A. Institution. 



in the prosecution of other sciences, and of all which men pur- 

 sue merely through their power of analysis and generalization 

 exercised on phenomena over the occurrence of which they 

 have no control, its improvement perhaps needs the most in- 

 tricate research, and its perfection the most lasting and widely,, 

 diffused observation ; to measure its various powers and prog- 

 nosticate their variations, will be, if brought under the control 

 of human science, the loftiest and most beneficial of its results. 

 But it is one which peculiarly requires the aid of many widely 

 scattered observers, and of vigilant observation. All the 

 climates of the world are under its influence, and by it are 

 • united into one system, so that no great change in its functions 

 in one district, can be entirely of local effect, but must be 

 either the symptom or the cause of extended variations. Ere v 

 a perfect system of climate for one country can be constructed, 

 these variations must be observed in all others, and their causes 

 traced to their origin ; and perhaps the real reason why such 

 attempts have hitherto been so futile, is, that they have been 

 founded on the partial data derived from local observations. 

 The little shifting eddiea of the stream have alone been watched 

 without looking to the swell or fall of the wide and far travelled 

 currents which give them birth. Though separated by a very' 

 distinct boundary, the great atmospheric commotions or the 

 changes which occur beyond its mean movements in one 

 hemisphere, cannot fail of influencing the other; and hence the 

 obvious advantage of universal observations at the same times 

 aud hence the value, for purposes of comparison, due to obser- 

 vations in past years. One of the most instructive means of 

 investigating the variations of these currents, would be to con- 

 struct a map of all the directions which the wind has assumed 

 over every part of the world at the same instant, and it will 

 be, perhaps, only when such a spectacle can be presented for 

 every day, over many successive years, that men will have 

 skill to form rules for determining the changes of coming 

 seasons. With such aid. of sufficient extent and continuance 

 it would no longer be chimerical to predict with no hesitating 

 accuracy, the'cyclc of the seasons. There are means, we may 

 remark, for the attainment of this result, more, abundant and 

 important than men have yet employed. When we consider. 

 the. number of ships constantly traversing the ocean, with their 

 existence dependant on the vigilance of their observations, and 

 almost aH in the practice of recording minutely, the changes 

 which they experience, there is'presented to us in the journals 

 of their voyages, an ample fund of the information we need ; 

 and perhaps a day has not elapsed for these many years, in* 

 which there might not have been derived from these floating ob- 

 servatories, the means of determining the course and termina- 



