Plumb-line in the Hawaiian Islands. 307 



year — and this all falls during the time the trades blow — only one 

 good night's observation conld be obtained in a month's stay. 

 At Ka Lae the station was eight miles from the water supply ; 

 the road leading part of the way over the rough lava known 

 as the " aa." All these circumstances made it necessary to 

 make the best possible use of good weather, when so favored, 

 and not unfrequently over one hundred pairs of stars were se- 

 lected for observation on a single night. The greatest number 

 obtained, however, was seventy-five, these being taken at 

 Kaupo in less than eight hours' observing. Satisfactory work 

 was done, allowing only one minute and a half between pairs, 

 and thirty seconds between stars of the same pair. In this 

 time the micrometer was read and the instrument revolved 

 180° in azimuth. The whole number of observations was 

 about 1500, being on the average more than one hundred for 

 each station. 



The mean places of these stars are being deduced by Mr. 

 Henry Farquhar, at the Coast Survey office, and will be very 

 reliable — about twenty of the best modern catalogues being 

 consulted. Many of the stations, however, being nearly in the 

 same latitude, pairs of stars occur which are common to sev- 

 eral places ; so that the difference of latitude of these stations 

 is independent of any error in the declinations. 



For the gravity work two pendulums were supplied by the 

 Coast and Geodetic Survey. They were of the reversible pat- 

 tern one measuring a meter between the knife edges and the 

 other a yard. They were to be swung on top of one of the 

 mountains, at the sea level, and at Honolulu, to connect the 

 work with that done in 1883 by the IT. S. Eclipse Expedition. 



The mountain chosen was Haleakala on the island of Maui. 

 It rises slightly over ten thousand feet above the sea and has 

 on its summit one of the largest extinct craters of the world : — the 

 crater itself being half a mile .deep and twenty miles in cir- 

 cumference. At the southwestern corner of this gigantic pit 

 and within fifty feet of its perpendicular walls the instruments 

 were set up and the force of gravity determined. 



The Yard Pendulum is common to both the determinations 

 of 1883 and those of 1887. 



The scheme for conducting the observations was identical 

 for each station and was strictly adhered to as far as the circum- 

 stances would permit. The pendulum was swung day and 

 night without interruption from the beginning to the end of 

 the observations. Stars were first obtained for time, and the 

 pendulum was started at the time of reversing the telescope. 

 This being the mean epoch of the star observations, the oscilla- 

 tions were thus referred directly to the stars without depend- 

 ing on the rate of the chronometer. The same stars when 



