436 



NATURE 



{March 7, 1889 



perhaps can be observed. It must appear ungracious to question 

 a theory that accords so completely with the natural history of 

 coral islands, but even this theory reqitires a geological concession, 

 and that is stability. Coral islands, it may be supposed, after all 

 only differ from other oceanic islands in being crusted over with 

 ■coral, so that we cannot see their original state, and the question 

 is whether we can grant such long periods of stability t6 them, 

 from our experience of other oceanic islands, which are free from 

 coral and can therefore be observed. Nearly all oceanic islands 

 are volcanic, and it is probable that their elevation coincides more 

 or less with the period of volcanic activity somewhere along their 

 line. It is obvious that coral islands are not formed during this 

 phase, because no theory would then hold good ; the peaks would 

 grow through and cany up the coral, which might leave only such 

 small traces of its existence as we find in a single spot m 

 Madeira. It would not be unreasonable to suppose that if the 

 expansive and elevating force were withdrawn the peaks would 

 slowly subside, and that if there are some lines of elevation, there 

 must be others of subsidence, unless the earth is as a whole 

 growing in bulk. Darwin claims the existence of areas of. sub- 

 sidence, and that these are eminently favourable to coral growth, 

 and it is quite apparent that if the Island of Madeira were to 

 sink, as it has undoubtedly risen, its last appearance in a coral 

 sea would be as an atoll. We shall never see the interior 

 structure of a stationary or subsiding coral island, and can only 

 look for a re-elevated example with a crust that has been pro- 

 tected from solution whilst dead and submerged, and yet not 

 sufficiently so to mask the core. 



In submitting geological considerations I am not questioning 

 any of Mr. Murray's observations, which are in every way 

 admirable, though it does appear to me doubtful whether atolls 

 could increase outwards in deep water on their own talus, in face 

 of the dissolution of dead coral that is claimed to take place in 

 the interior of the lagoons, and yet more so in deeper water. 

 J. Starkie Gardner. 



The Sun's Ccrona, 1889. 



A GENERAL Statement of the successes of the Western Eclipse 

 Expeditions on January i has already appeared in Nature, 

 More photographs of the corona were taken than ever before— 

 many of them indifferent and worthless, but an unusually large 

 number of great excellence. The best that I have so far seen 

 were taken with 5- inch telescopes, by Mr. W. H. Pickering at 

 Willow, and by Captain R. S. Floyd at Lakeport, both in Cali- 

 fornia. The latter's lens was newly made by Clark, on the 

 Stokes-Pickering plan, convertible from optical to photographic 

 use by reversing the crown lens. 



Until the photographs can be well collated, there is little use 

 in presenting them ; and the difficulties arising in this work are 

 by no means easy to meet. 



With the drawings, however, the case is different. These 

 were made in great abundance, and I have received sufficient 

 responses to my printed instructions to afford very satisfactory 

 conclusions as to the appearance of the corona. The state of the 

 sky was practically everywhere favourable throughout California, 

 Nevada, Idaho, Wyoming, Montana, Dakota, and Manitoba. 



The instructions for sketching the corona were printed in three 

 sections : (i) drawings of the corona as a whole ; (2) drawings 

 with small telescopes of the filaments about the solar poles ; (3) 

 sketches of the outlying streamers along the ecliptic. These 

 latter were made with the assistance of an occulting disk, set up 

 at such distance from the eye of the observer that it would sub- 

 tend an angle of 65'. It was supposed that disks of this size, 

 being much larger than those used on any previous occasion, 

 would hide very nearly all the inner corona, and leave the eye 

 free to follow the faint outer filaments to their farthest limit. 

 The magnitude and brilliancy of the inner corona, however, 

 were such as to convince me that the disks might better have 

 been one-fourth larger. 



From the best of all the drawings now available, of the three 

 classes, Mrs. Todd has prepared the accompanying sketch of the 

 corona. This was done wiikout knowledge of the details shown 

 on any of the photographs, and it may be taken as an index of 

 the sort of results which may be derived from the co-operative 

 plan of figuring the optical corona. It is also instructive in 

 studying the differences between the optical and the photo- 

 graphic corona. David P. Todd. 

 Amhetst College Observatory, February 22. 



[No sketch was received with this letter. — Ed.] 



The Meteoric Theory of Nebulae, &c. 



There would appear to be a difficulty in the theory of the- 

 meteoric constitution of nebuloe,i ^^c., which, as far as I am 

 aware, has not been mentioned. _ , 



It is, namely; the fact that some gas— probably part of it 

 permanent— exists in the nebula along with the moving masses 

 in translatory motion. Making allowance for the relatively 

 small effect of gravity on the gas, due to the diffuse distribution 

 of the matter, and consequently having regard to the probable 

 tenuity of the gas ; it has nevertheless, I find, been estimated 

 by Joule (" Scientific Papers," vol. i. p. 539) that meteors are 

 first observed at a height of 116 miles in the earth's atmosphere. 

 He estimates that 0*0003 oif a grain of air is contained in a column 

 of air one mile long, and one square foot in cross section at that 

 height. This, I find by calculation, amounts to i/iooo millionth 

 of an atmosphere in round numbers as to density. 



So that if in some nebulfe the gas had something like this 

 small density, the bodies, or masses moving in translatory motion 

 according to the kinetic theory, would (if their velocity were at 

 all comparable to that of those colliding in the earth's atmosphere) 

 behave as meteors, or inflame ; and so apparently be rapidly con- 

 verted into gas. Even if they did not inflame ; no doubt the heat 

 consequent on friction would be considerable. It might be sug- 

 gested, perhaps, that the mass of these bodies in some nebulae 

 may be so great that they do not lose their translatory motion 

 rapidly, even if they leave a luminous track. In any case it is 

 evident that this stage of evolution is not a lasting one, and, to 

 my mind, it seems that it is less permanent than is perhaps 

 generally supposed. 



I find that Mr. G. H. Darwin, in his paper in the Philosophi- 

 cal Transactions, 1889, above alluded to, suggests the hypothesis 

 that the " metallic rain " generated by the condensation of the ^ 

 incandescent vapour of iron could "fuse with old meteorites' 

 whose surfaces are molten." It seems to me that the rate of 

 translatory motion, calculated by him at 5^ kilometres per 

 second, is scarcely allowed for here. How, it may be asked, 

 could such " metallic rain " fuse on bodies colliding against it 

 at this velocity ? Some are movihg lat a less velocity, no doubt ■; 

 but some are moving at a greater^:' 



The tempefature equivalent to this value for translatory motion 

 (5^ kilometres per second) is, I find, 36,000° C. (about) ; i.e. 

 this would be the temperature if the translatory motion alone 

 were entirely converted into heat. Clausius has calculated, I 

 believe, that in a gas the ratio of the whole energy (which in-, 

 eludes translational and vibrational energy of molecules) is to 

 the vibrational energy alone as the specific heat at constant 

 pressure is to that at constant volume." If this be the case, a 

 very large proportion of the translatory motion is resolved into 

 internal motion— that motion which emits the waves of heat 

 analyzed in the gas. Must not the same be true of meteo.ic 

 masses : or is not the principle (ratio) mdependent of \\it scale, 

 or number of molecules clustered about a centre, and moving as 

 one lump in the motion of translation? In some complex gases, 

 at least fifty to sixty molecules may be clustered about a centre to, 

 form a lump. Then if more (as in a meteorite) are so clustered, 

 it appears that the same must hold true, as regards subdivision of 

 the energy between translatory motion and vibratory motion 

 (heat). If so, by the great temperature equivalent of tlie 

 translatory motion (viz. 36,000° C. above estimated), the 

 meteorite would rapidly be dissociated into separate molecules, 

 by the subdivision of the energy according to the above principle 

 —just as the more firmly united constituents of the lumps {i.e: 

 compound molecules) of gases would be dissociated, even if 

 moving at but a fraction of the above translatory velocity. 



Is it supposed perhaps that the length of path between en- 

 counters (giving time to cool ?) in meteorites constituting nebulas, 

 prevents this,?, This point is not apparently gone into in Mr. 

 Darwin's paper. But if the meteoric mass has time (nearly) to 

 cool down, or lose, by radiation into stellar space, the heat 

 generated at' each successive collision, then it would seem that 

 the translatory motion would be somewhat rapidly lost by con- 

 ' I allude specially to Mr. G. H. Darwin's paper, "On the Mechanical 

 Constitution of a Swarm of Meteorites," of which an abstract appeared in 

 Nature of November 22 and 29, .1888 (pp. 81 and 105)- ihe paper is con; 

 tained in full in the Philosophical Transactions, vol. clx.\x., X889. 



= It may be curious to observe that, if a meteoric swarm whose mass 

 equalled that of the sun, were contained within an impenetrable envelope, 

 whose radius equalled the ra,dius of Uranus's orbit (nearly) the mean den- 

 sity of the meteoric swarm would be one five-millionth of an atmosphere 

 only ; which represents a fair " vacuum " of a Sprengel pump. This degree 

 of distribution of matter may be represented by a. centimetre cube ot iroii 

 p'aced at the corner of a cube 139 metres in the side. 



