NATURE 



JOI 



THURSDAY, JUNE lo, 1875 



THE METEOROLOGICAL OFFICE 



Quarterly Weather Report of the Meteorological Office. 

 Published by the authority of the Meteorological Com- 

 mittee, January 1869, to September 1873. Hourly 

 Readings from the self-recording instruments at the 

 seven Observatories in connection with the Meteoro- 

 logical Office, January to September 1874. 

 T'^HPI self-recording instruments which have been in 

 -L operation at the seven Observatories of the Meteoro- 

 logical Committee since January 1869, may be regarded as 

 the best and most complete anywhere existing for record- 

 ing continuously the atmospheric pressure, temperature 

 humidity and rainfall, and the velocity and direction of 

 the wind. To ensure correctness in the work, and accu- 

 rate tabulation of the results, minute regulations with 

 respect to the officials at the outlying Observatories, the 

 assistant at the Central Observatory, and the director of 

 the Central Observatory, were laid down in the Com- 

 mittee's Report for 1868, p. 62, Thus, as regards the 

 thermograph, twenty-seven regulations were laid down, 

 one of the most important of these being the 25th, by 

 which it was provided that forty remeasurements from 

 each month's curves were to be made at Kew, the central 

 Observatory ; and a table is given (page 39 of the same 

 Report) of the results of measurements which were 

 specially designed for the detection of small errors in the 

 thermograph tabulations, from which it appears that re- 

 finements as minute as the one-hundredth of a degree 

 of temperature were taken cognisance of in the results. 



Tracings of the curves, and five-day and monthly re- 

 sults of the tabulations, though not the tabulations them- 

 selves, have been published, beginning with ist January, 

 1869; and since then many and great improvements have 

 been made in representing the curves on the Plates, all in 

 the direction of greater clearness and precision, for which 

 the Committee deserve our best thanks. Among the 

 many valuable results of these curves we may point to 

 the high temperature at Glasgow on the 21st April, 1873, 

 in connection with the remarkable changes in the direc- 

 tion and force of the wind which occurred at the time • 

 to the heavy rainfall at Valencia on the 2nd July, 1873, in 

 connection with the changes of wind, temperature and 

 pressure ; and to the minute oscillations of pressure at 

 almost all the Observatories on the 3rd and 4th July, 

 1873, in connection with the changeable weather at the 

 time. In these connections the absence of any observa- 

 tions of clouds is, however, a serious defect. 



One of the principal objects for which the seven Obser- 

 vatories were established was to furnish the data of 

 observation for the determination of the meteorological 

 " constants " for pressure, temperature, rainfall, &c., for 

 different parts of the British Isles. This being now the 

 seventh year in which this expensive system of observa- 

 tion is going forward, it may be well to inquire how far 

 the information, as published by the Meteorological Office, 

 meets the requirements of the problems to be solved. 



Assuming that the curves are correctly traced from the 

 photographs, we may inquire whether the figures tabu- 

 lated from these, under the regulations referred to above, 

 Vol. XII.— No. 293 



be satisfactorily accurate. No hourly values having been 

 printed before January 1874, the question can only be 

 answered by an examination of the printed monthly 

 maxima and minima, with the days and hours of their 

 occurrence, as compared with the curves. The following 

 Table, giving the extreme readings of the thermometer 

 for each of the Observatories for January 1869, is here 

 reprinted verbatim from the Quarterly Weather Report 

 for 1869, Part I., p. 34 : — 



Each datum of this table we have compared with the 

 temperature curves for the month, measuring each obser- 

 vation four times, viz., by the side scales of each curve 

 from below upwards, and from above downwards. Setting 

 aside every reading which does not differ from the mea- 

 sured reading so much as o°-4 of a degree, and the discre- 

 pancies which appear to arise from the unequal shrinkage 

 of the paper as indicated by the results of the four mea- 

 surements, there are in the above Table twelve errors, the 

 maxima at Falmouth and Kew being doubly wrong, the 

 amount at the given hours being wrong, and the date of 

 occurrence being also wrong. The following is the Table 

 as corrected, the corrected readings being shown by 

 asterisks : — 



It will be observed that the errors are of three sorts — (i) 

 errors of temperature and errors of the date of occurrence 

 of the maxima and minima, including (2) errors of the 

 day of the month, and (3) errors of the hour of the day. 

 Similarly the other months of 1869 have been examined, 

 with the result that forty-one errors of temperature varying 

 from o°'4 to 9°'6 f have been detected, that the day of the 

 month, as printed, is wrong on twenty-two occasions, and 

 that the hour of the day is wrong in nine cases, in which 

 the temperature and day of the month are correct — in all, 

 seventy-two errors. The Tables and curves for ten 

 months, taken indiscriminately from the other years, have 

 also been examined. 



In the Tables for April and June 1870 (p. 37 of the Quar- 

 terly Weather Report of that year) there occur six errors in 

 each of these months, and in the Table for March 1871 

 (p. 26 of Q. W. Report for 1871) there occur seven errors. 

 In none of the twenty-two months examined are there 



+ The minimum temperature at Glasgow for October is given as 39*'9 at 

 p. 109 of Q. W. Report, instead of 3o*-3 as by the curve of temperature 

 (Plate cxix.). - 



