46 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. IO5 



converted into inches in order to make them comparable with the 

 values pubHshed in Smithsonian Misc. Coll., vols. 79 and 90. The 

 values of pressure in millibars are given under "Pressure at Sea 

 Level." These were prepared by W. A. Macky, Meteorologist. The 

 early observations made at Hamilton and published in the Blue Books 

 he did not C9nsider of the desired accuracy and hence omitted them. 

 In regard to the observations at Prospect and St. George he writes : 

 "Prospect Observations. — These were taken by N.C.O.'s of the 

 Medical Corps under the distant supervision of the Canadian 

 Meteorological Service to whom the original records were 

 forwarded. We have in the office here these original records 

 from the beginning of 1922 but the earlier ones were de- 

 stroyed after the necessary data had been extracted in 

 Toronto, and we have available for the years 1908 to 1921 

 only the summaries published in the Bermuda Blue Books, 

 and somewhat similar summaries published in Canada. There 

 are various more or less slight discrepancies between the two 

 published versions and I have not yet been able to check 

 fully the period before 1922. 

 "From the daily observations available from 1922 to 1931 we 

 have checked carefully all the published values. It is assumed 

 that the actual readings taken at Prospect were reasonably 

 accurate. The barometer was a Fortin which we have in 

 this office and it agrees very well with our standard. It was 

 found, however, that up until the end of March 1927 no 

 correction was made for standard gravity and also at several 

 later periods this was omitted again. The site of the barom- 

 eter was moved at least once and for more than a year after 

 this move the altitude correction appropriate to the first site 

 was used. 

 "All these errors have been corrected to the best of our ability and 

 the correct sea level pressures from 1922 to 1932 for each 

 month are given in the attached table. 

 "St. George Observations. — From April i, 1932, observations pub- 

 lished have been for St. George and all the proper correc- 

 tions have been made to these. They are given also in the 

 table herewith. To sum up: (a) the table herewith gives the 

 only mean monthly values which I consider are to be relied 

 upon, that is from January 1922 to December 1940. (b) 

 There is as far as I know no source of reliable or even 

 semireliable values earlier than June i, 1908. (c) From 

 June I, 1908, until the end of 1921 the mean values pub- 



