December 17, 1891] 



NATURE 



61 



For the preliminary experiments which are to be described it 

 wasjthought desirable to use the 12-inch equatorial. Accordingly, 

 a cap, provided with two adjustable slits, was fitted over the 

 objective, and provided with a rod by means of which the distance 

 between the slits could be altered gradually and at will by the 

 observer, while the distance was measured on a millimetre scale 

 attached to the sliding jaws. This arrangement, which was con- 

 structed under the supervision of Mr. F. L. O. Wadsworth, 

 of Clark University, is shown in the accompanying diagram. 

 Fig. 3. 



With this apparatus the satellites of Jupiter were measured, 

 with results as given in the following table : — 



Table I. 



No. of y 



Satellites. 



August 2 ... 1*29 ... 



August 3 ... I "29 ... 



August 6 ... I '30 ... 



August 7 ... 1-30 ... 



III. 



IV. 



Seeing. 



Me^n. 



I'29 



119 



I -21 

 118 



119 



1-88 ... 1-68 ... Poor. 



1*59 ... 1-68 ... Poor. 



I 69 ... 1-56 ... Poor. 



177 ... 171 ... Good. 



173 



1-66 



These are the values of the angular diameters of the satellites 

 of Jupiter as seen from the earth. To reduce these to Jupiter's 

 mean distance these values are to be multiplied by 079, which 

 gives for the final values — 



II. 

 o"-94 



III. 

 i''-37 



IV. 

 i"-3i 



It was found impossible to see the reappearance of the fringes 

 on increasing the distance, yet the results of Table I. show that 



NO. I 155. VOL. 45] 



For the sake of comparison these values are recorded in the 

 following table, together with those given by Engelmann, 

 Struve, and Hough, and the last column contains some results 

 kindly furnished by Prof. Burnham with the 36-inch on the 

 same date (August 7) as the last of the series by A. A. M. : — 



the disappearance could still be sharply marked. Indeed the 

 concordance of the observations made under different circum- 

 stances on different nights was even closer than was expected. 

 With a larger telescope both the brightness of the fringes and 

 their distance apart will be increased, and it may be confidently 

 predicted that the accuracy will then be even greater. 



The values given in the second column, " Engelmann," are 

 probably more reliable than the succeeding ones, but it is well 

 worth noting that the results obtained by interference agree with 

 the others quite as well as these agree with each other. 



It should also be noted that the distance between the slits was 

 about four inches. It may therefore be stated that for such 

 measurements as have just been described, a telescope sufficiently 

 large to admit a separation of four inches — say a six-inch — 

 suitably provided with adjustable slits is fully equal to the largest 

 telescopes now used without them. 



It is hoped that within a few months the 36-inch equatorial 

 will be supplied with a similar apparatus and observations begun 

 for the definite measurement of the satellites of Jupiter and 

 Saturn and such of the asteroids as may come within the range 

 of the instrument. 



In concluding, I wish to take this opportunity of expressing 

 my appreciation of the courtesy of Director Holden in placing 

 all the facilities of the Observatory at my disposal, and of the 

 hearty co-operation of all the astronomers of the Observatory, 

 especially the valuable assistance of Prof. W. W. Campbell in 

 making the observations. A. A. Michelson. 



Mount Hamilton. 



THE SAMOAN CYCLONE OF MARCH 16, 1889. 



'T'HE Samoan hurricane of March 16, 1889, is one of those 

 ■*• historic storms that have been rendered for ever memorable 

 by the episodes of disaster and gallantry that attended them ; 

 by the escape of H.M.S. Calliope, which forced her way out of 

 Apia harbour in the teeth of the hurricane, amid the cheers of 

 the brave American sailors, who, themselves face to face with 

 imminent death, forgot for a moment their own dire peril in 

 their admiration of the daring and successful act of seamanship 

 that rescued their more fortunate brothers. Mr. Everard Hay- 

 den, of the U.S. Hydrographic Office, has lately issued a pre- 

 liminary Report on this storm, which, despite the regrettable 

 meagreness of the data at his command, has, nevertheless, a 

 certain scientific interest, inasmuch as less is known of the 

 cyclones of the Pacific than of those of most other tropical 

 seas. 



The Apia storm, like the cyclones of the South Indian Ocean, 

 was evidently formed on the northern limits of the south-east 

 trades, and was one of a series that were generated in this region 

 in March 1889. The first of these, in Mr. Hayden's opinion, 

 appears to have originated on the 5th of the month, some 500 

 miles north-north-east from the Samoan Islands, and to have 

 travelled first in a south-westerly direction, recurving in the 

 latitude of these islands, but at 150 to 200 miles to the west of 

 them, after which it took a south-eastward course between Tonga 

 and Nuie. It seems to have been a storm of great severity, and 

 its passage was felt at Apia on the 6th and 7th, though not with 

 any great intensity. It was succeeded by the cyclone that forms 

 the principal subject of Mr. Hayden's Report. This, he thinks, 

 was formed about March 13, some 300 miles to the north-east of 

 the Samoan Islands, and on the 15th its centre passed either 

 directly over, or a little to the north of, Apia harbour, moving, 

 therefore, on a south-west course. He considers that it then 

 sharply recurved, and that, with greatly increased strength, it 

 passed a second time over Apia on the i6th, the day of the great 

 naval disaster. The chief facts which led Mr. Uayden to 

 this conclusion are those observed at Apia itself, for no positive 

 evidence is forthcoming from the supposed birthplace of the 

 storm, and only one ship reports the state of the weather any- 

 where to the north of Samoa. The peculiar feature of the Apia 

 observations is, that the barometer fell steadily from the 12th to 

 the afternoon of the 15th (about 07 inch), then rose (about o'2S 

 inch) during the latter part of that day, and then again fell on the 

 16th to a reading slightly lower than that of the previous day. 

 On the 15th, squalls of moderate force (wind southerly, force 2 

 to 6) were experienced, and in the after part of the dny, as the 

 barometer rose, the direction changed from south to north and 

 east. There had been no heavy sea, and it was thought that 

 the gale was over. At midnight, however, the barometer began 



