March, 1913. 



KNOWLEDGE. 



97 



of the carrying currents Professor Omori estimates 

 at about five or six miles, this being the height 

 generally reached by the smoke-column during the 

 explosions of the Asama-yama. 



In two cases, the form of the sound-area is still 

 more remarkable, for it consists of two detached 

 portions. The shaded areas in Figure 93, represent 

 these portions for the eruption of December 25th, 

 1910. One portion includes the Asama-yama near 

 its western margin, the other lies about fifty or sixty 

 miles to the west. It will be noticed that the direc- 

 tion of the surface-wind was north-west in the 

 eastern portion, and south-west in part of the other, 

 so that there is no connection between the form of 

 the sound-area and the direction of the surface-wind. 

 Here, again, the anomalous form of the area must 

 therefore be connected in some way with the trend 

 of the upper winds. 



A somewhat similar anomaly was observed in the 

 audibility of the minute-guns which were fired from 

 battleships lined at Spithead, when the body of her 

 late Majesty Queen Victoria was borne from Cowes 

 to Portsmouth. At places in the immediate 

 neighbourhood of Spithead, and as far as fifty miles 

 from it, the guns were almost or quite inaudible. 

 From sixty to eighty or ninety miles they were 

 clearly heard, so clearly that at a distance of eighty- 



four miles, labourers in the fields put down their 

 spades and listened. Beyond ninety miles, records 

 were less numerous, but one came from near Wood- 

 bridge, in Suffolk, at a distance of one hundred and 

 thirty-nine miles, the regularity of the booms of the 

 minute-guns allowing no doubt as to the observation.* 

 In the neighbourhood of Spithead and Portsmouth, 

 the wind at the time was from the west or north- 

 west, or " offshore " ; at the great distances at which 

 the sound was heard the wind was southerly. Since 

 the velocity of the wind increases, as a rule, with the 

 height above the ground, it follows that, in the 

 direction from which the wind is coming the sound- 

 rays are bent upwards, and pass over the heads of 

 observers at a moderate distance. In the opposite 

 direction, they are bent downwards, so that sound- 

 rays which started upwards at a moderate angle are 

 brought down again to observers at a considerable 

 distance. Thus, the sound-rays were first of all 

 refracted by contrary winds over the heads of 

 observers between ten and fifty miles, and were after- 

 wards brought down by favourable upper currents so 

 that the reports were clearly audible from sixty to 

 more than a hundred miles from Spithead. It is 

 probable that a similar explanation may be given of 

 the two detached portions of the sound-area in the 

 case of the explosions of the Asama-yama. 



'On the audibility of the minute-guns fired at Spithead on February 1st [1901] : 



Pages 124-125. 



Knowledge," Volume XXIV, 1901, 



NOTES. 



ASTRONOMY. 



By A. C. D. Crommelin, B.A., D.Sc, F.K.A.S. 



THE OCCULTATION OF THE STAR MAYER 588 

 BY JUPITER'S THIRD SATELLITE, GANYMEDE, 

 1911, AUGUST 13TH. — This very rare phenomenon was 

 predicted by Herr Banachiewicz, and observed at several 

 stations in South America, Professor Ristenpart having cir- 

 culated requests to observers with a view to deducing the 

 dimensions of Ganymede. Thirty-two had volunteered to 

 help, who were spread over thirty-seven degrees of latitude, 

 from Arequipa, S. 16°, to Punta Arenas, S. 53°. Several 

 observers noted that the disappearance and reappearance 

 were sharply defined, and could be accurately timed. The 

 time intervals give chords of the disc of Ganymede in various 

 latitudes, and the surprising fact emerges that it must be 

 considerably larger than was hitherto supposed. The observa- 

 tions are best satisfied by an elliptical outline, with equatorial 

 radius two thousand three hundred and thirty-one miles, polar 

 radius two thousand one hundred and thirty-one miles. 

 Dr. Ristenpart does not, however, attach great weight to the 

 ellipticity. Ganymede would appear to be as large as Mars, 

 whereas former estimates made it only equal to Mercury. 

 One puzzle that the new result affords is the very low density 

 that it implies. The mass of Ganymede is one thirty-ninth of 

 the Earth, or one quarter of Mars, so if its size is equal to 

 Mars its density is only one quarter as great, or, roughly, the 

 density of water. It is true that the density of Saturn is still 

 less, 0-7 of water, but the visible globe of Saturn is believed 

 to be largely vaporous, which is unlikely to be the case with 

 Ganymede. 



THE SPECTROSCOPIC BINARY 9CAMELO- 

 PARDALIS. — Several observers have noted abnormal 

 behaviour of the calcium lines in the case of spectroscopic 

 binaries. For example, Hartmann in the case of S Orionis, 

 Daniel, Schlesinger, Duncan and Slipherin the case of p Scorpii, 

 find that the H and K lines of calcium do not share in the 

 large displacement of the other lines, though they have 

 probably a small displacement of their own. The suggested 

 explanation is that the calcium vapour is not in the stars, but 

 surrounds them as a cloud. Mr. O. J. Lee, at Professor 

 Frost's suggestion, selected 9 Camelop. as a suitable star 

 for further investigation of this question (Astrophysical 

 Journal, January). Its declination is 60° North, and it 

 culminates at midnight in December, so that very long series 

 of photographs can be taken. Further, the H, K lines are 

 strong and well defined. He deduces that the period of 

 oscillation of the calcium lines is the same as that of the 

 other lines viz., eight days, but the amount of oscillation less. 

 He concludes that the calcium vapour forms an ellipsoidal 

 sheath round the principal star, and that the spectroscopically 

 effective regions are near the zero velocity points. He 

 deduces for the joint mass of the system only one four- 

 hundredth of that of the Sun, or two and a half times that of 

 Jupiter. This seems an improbably small mass for so distant 

 a body (its parallax is given as "-026, implying a distance of 

 one hundred and twenty light years), but the author admits 

 that some of his assumptions are tentative. In any case, the 

 existence of external calcium clouds in the case of several 

 spectroscopic binaries has been rendered highly probable. 



A MEASURE OF SOLAR RADIATION FROM FREE 

 BALLOONS. — The same publication contains an account by 

 Professor Very of an interesting research : the service that 



ii 



