Proceedings of the British Association. 371 



exception of one solitary instance, the station being in the Pyrenees 

 and in the neighborhood of iron mines, there was no diminution of 

 the magnetic intensity at the higher stations, at least beyond the 

 limits of the necessary errors of instruments and observations : even 

 at the station where some diminution did manifest itself, the quantity 

 of that diminution was very much smaller than that which resulted 

 from the views of M. Kuppfer. 



Prof. Powell read a paper " respecting the impermeability of 

 water to radiant heat." 



A paper by Sir David Brewster, "■ On the action of crystallized 

 substances upon Light," was then read by the Secretary, Mr. Snow 

 Harris. 



Dr. Williams gave an account of an improved ear trumpet. 

 The President then said, that as there were yet a great many in- 

 teresting communications to be brought forward, the Section would 

 reassemble at eight o'clock in the evening. 



Thursday Evening. — Mr. G. W. Hall made a communication 

 upon " the connexion observed at Bristol between the weather and 

 the tide." He commenced by stating, that long and carefully con- 

 tinued observation of the weather at Bristol, together with a direct 

 interest in becoming possessed of rules for anticipating its changes, 

 led to the following theory, which was strikingly correspondent with 

 facts. First, that the barometer very generally, indeed, almost in- 

 variably, undulates at times corresponding with the changes of the 

 moon, and at these times it more frequently falls than rises. Sec- 

 otidly, that the weather is ordinarily unsettled at these periods, con- 

 tinuing so for about two or three days ; and for the most part the 

 wind becomes high at these times. Thirdly, that as the weather 

 settles (if it become at all settled, since it not unfrequently remains 

 in an unsettled state,) so will it continue until the next change of 

 moon, or rather until the recurrence of its disturbing inflnences. 

 Fourthly, that these variations occur as regularly at the quarters of 

 the moon as at the new and full, and are then as fully marked. 

 Fifthly, that the period, about five days, which determines the state 

 of the weather, is derived from the spring and neap tides, or the full 

 influence of the sun and moon upon them. — The only origin of these 

 rules, he stated, was actual observation. Very striking changes of 

 temperature and weather, from intense frost to spring mildness, and 

 then frost recurring, first led to marking this correspondence ; and 

 so closely has it been observed, and so fully established, that opera- 



