of the Flagstaff Observatory. 101 



my plans as to the mode of registration; and as no institution 

 of a similar nature has as yet been established in the southern 

 hemisphere, I did not for a moment doubt but that my efforts 

 would be crowned with success ; consequently I took mea- 

 sures adapted to place the Observatory, from the beginning, 

 upon a right footing : a system of hourly observation and 

 registration having been organised, which has gradually come 

 into operation. 



The meteorological instruments have been registered hourly 

 from the 1st of March, 1858. The atmospheric-electrical 

 tension has been hourly observed from the 15th of April last, 

 and other horary variations in terrestrial magnetism from 

 the 1st of May. The delay in starting some of the operations 

 was owing to the fact that the different buildings became 

 only gradually available for the purpose, so that complete 

 hourly registrations have only been carried out in every de- 

 partment from the 1st of May. 



The order in which the registrations are made at present 

 is as follows : — 



At 1 m. 30 s. previous to the full hour the barometer is 

 read. 



At the hour itself, the instruments for horary variations in 

 terrestrial magnetism. 



At 1 m. 30 s. after the hour, the dry and wet bulb, black 

 and white bulb, and soil thermometer are read. 



At 2 m. wind, rain, clouds, &c. 



At 5 m. the electrical tension of the atmosphere is observed. 



Thus we bring the whole set of readings within the short 

 space of 6 m. 30 s., and this time is strictly adhered to, both 

 night and day. 



This plan of observing is only changed at times of great 

 disturbance, either in magnetism, electricity, or in the at- 

 mosphere generally — all necessary instruments being then 

 recorded from 5 to 5 minutes, and even, if required, from 

 minute to minute. If we now take into consideration that 

 550 observations are thus registered in a day, it will be at once 

 evident that we must take such steps as will keep all the ac- 

 cumulating facts continually in view, and this is effected by 

 making the most necessary reductions at once, and winding 

 up every five days 5 work by determining the means for every 

 hour, and from them again the means for the period. It must 

 be borne in mind that, while doing so, it is incumbent on the 

 staff of the Observatory to copy all single registrations again, 

 for the purpose of classifying them. It can only be by 



