18 Ocean Magnetic Observations, 1905-16 



the price of the latter. The azimuth circle is the Negus pattern, and is described in con- 

 nection with the sea deflector, page 24. (See Plate 5, Figs. 1-4.) 



Some experimental declination observations were also made with a Kelvin compass. 

 This is the dry, silk-suspended card compass, designed and patented by Sir William Thom- 

 son (Lord Kelvin). The Kelvm azimuth instrument for this compass was made by the 

 Kelvin and James White Company. (See Plate 3, Fig. 5.) 



Practically every modern azimuth device was given a trial, but none was found equal 

 to the requirements of all the varied conditions encountered. In general the simplest 

 devices were found to be the best. With bright Sun and a fairly smooth sea, good results 

 can be obtained by a careful observer with any of the best azunuth cu-cles in use. For 

 the varied conditions encountered, the observers on the Galilee gave preference, in general, 

 to the Ritchie azunuth circle mentioned above, this ha\dng both a prism-reflection device 

 for fairly bright Sun and a direct-vision method. Apparently there was considerable room 

 here for improvement, for it frequently happened that, under conditions which still per- 

 mitted securing satisfactory azimuth observations on land, none could be made at sea, 

 because the Sim was either too high or too faint to admit of obtaining good results with the 

 available azimuth-devices. 



Every known appliance was found subject more or less to the error arising from motion 

 of the card while the magnetic bearing of the celestial body was bemg taken. To eliminate 

 this error it was necessary to extend the observations over a sufficiently long interval so 

 that, on the average, it could be assumed that the effect of the various motions of the card 

 had been eliminated. This had to be done more or less blindly, however, as one could never 

 tell just at what point of the arc of motion the magnetic azimuth of the stellar body was 

 obtained. Furthermore, all azimuth appliances had movable parts subject to wear with 

 frequent use, such as the axes of mirrors or of prisms and the wear of the azimuth circle 

 on the bowl. Likewise, graduation errors of the card had to be considered. The result 

 of the defects in the usual ship's compasses and azimuth circles was the introduction of 

 "apparent" deviation corrections, not due to the ship's magnetism, but to purely instru- 

 mental causes. To be able to separate the "apparent" from the "true" deviations, it was 

 necessary to go thi-ough an elaborate series of shore observations whenever the vessel 

 reached port. The ship instruments were invariably dismounted and used ashore along- 

 side of the customary magnetic outfits for land work. As far as known, it xoas in the work 

 of the Galilee that the two sources of deviation-coefficients — those due purely to ship's mag- 

 netism and those due to defects in the magnetic instruments — were first systematically sepa- 

 rated. (See, for example, pp. 60-62.) 



The difficulties of securing azimuth observations at sea were further increased by the 

 meteorological conditions frequently encountered in the Pacific Ocean, viz, clouds and fog. 

 For example, on the experimental cruise from San Francisco to San Diego in August 1905, 

 we went out to sea 150 miles to get beyond the fog prevailing on the coast at that time 

 of the year, and not until the fourth day out did we secure the azimuth observations 

 necessary to give declination results. 



The problem of securing the magnetic bearings of celestial objects, and hence results 

 for magnetic declination, was the most serious one encountered in the steady progress of 

 the magnetic survey of the Pacific Ocean. In time of cloud or fog, results for magnetic 

 incUnation and mtensity could be obtained, but none for magnetic declination. Thus on 

 some portions of the cruises of the Galilee there were very few opportunities to secure 

 declinations. There may have been encountered on these portions weather that would 

 ordinarily be characterized as fine weather for navigation, permitting the securing of suffi- 

 cient sextant observations for navigating the vessel, but the observers still failed to obtain 

 satisfactory magnetic-declination results in sufficient number, chiefly on account of the 

 instrumental difficulties above mentioned. 



