It is very desirable that all the observatories keep the same time as exactly as possible, in order'to class the meteorologic pheno-me~a to note accurately the occurrence of earthquakes and to study theIr propagation, etc. In this case we measur€, rather the duration than determine the absolute time, and we can arrive at a greater precision if the clocks are in accord. E. P. COTTON, Late Commissioner of Lands and Director of Sun-eys, Nigeria.